I don't know if it was mentioned before, but please try to make it easy (or at least possible) for us to get development devices without having to sign up for Sprint, like Google is doing with their Android Dev Phones. It would be specially useful for those of us not in the US.
Andrew Wooldridge
· 11 months ago
Given that apps are written in HTML, CSS, and js - is there some way to let folks who want their sourcecode hidden (I'm not one of those people) so they will feel better about releasing an app on the platform?
Anderson
· 11 months ago
A universal, official, time-limited trial mechanism. Any developer who chooses to use the option, during distribution setup, gets the functionality at the OS level. In addition, please encourage developers to provide trial versions in general.
This will help developers get more sales and higher prices by allowing people to see value in a prospective app and reduce the purchasing uncertainty inherent in many mobile apps. Also, it should help to reduce the frequent problem of crap apps from overshadowing the high-quality ones.
Shorin
· 11 months ago
Um... i hope this is the area for the "feedback" mentioned above...
I would think that this is Palm's chance to "catch up" to other mobile platforms that have been meeting standards in user expectations out-of-the-box. (cough... Windows Mobile...cough). Palm needs to make this out-compete the various Microsoft-integrated platforms by out-integrating them with Microsoft (and etc) products natively. Also add all those features that just make the thing more useful (USB Mass storage mode). Also.. mmm.. hotsync needs to speed up... seriously. I never sync my centro because it never ever finishes (i'm guessing it would take an hour or more if i let it go). I know that's some heavy criticizm for something that is so close to coming out (and thus these considerations will probably be ignored), but I really want Palm to survive... I'm even considering applying to Palm when I graduate because it is my favorite mobile platform that I will always stand by. I don't want to see it die out before I get that ultimate GPS+Wifi+Bluetooth super Palm OS phone!
One last thing.... I'd like for my community-created apps to be runnable on this thing. I don't think you can run ARM code through HTML, Javascript or CSS. Is there any chance older apps will be compatible? (TCPMP - the free one, will need to be rewritten in the worst case; and TCPMP I swear is one of the hardest programs to compile from source, in my experience).
Thanks for taking my complaint ... and good luck with this new release!!
Jeffrey W. Baker
· 11 months ago
Reiterating my comment from before: please ensure that a user can install whatever he wants, without the intervention of a third party "app store". Having to deal with third party distribution makes life difficult for developers of private, intranet-facing applications. It's the #1 reason my organization can't adopt the iPhone. We don't want out private applications in anybody else's hands.
Justine Pratt
· 11 months ago
A successful new store should resolve the issues we've experienced in the past:
1) Ease of use for the customer to install (and try) on device or alternately purchase right away. 2) Ease for the customer to buy the app after trying (single click type methods are good for impulse buying and ensuring the process isn't abandoned). 3) Ease for the developer to communicate and provide updates to customers. 4) Commission rates that allow for the developers to succeed without extremely high volumes (i.e. the current 50-75% commissions are not conducive to a good developer economy). 5) An avenue for developers to promote their applications (paid or unpaid) on the store. 6) User awareness of available third-party software, and encouragement for them to install it.
Any methods that help these areas would seriously improve existing conditions that are faced by third-party developers.
Good to see the conversation continuing!
Steven Fisher
· 11 months ago
I don't have any comments about the store yet, but I did want to say that someone needs to look at the blog's stylesheet. The header is interfering with clicking links.
ashebanow
· 11 months ago
We'll get that fixed asap, thanks.
Joe
· 11 months ago
I'm sure you're trying to do this anyway, but just getting out the SDK ASAP would be a huge benefit.
And I personally think trials are a great idea and something you don't really get on the other platforms other than Palm/Garnet and Windows Mobile.
Sean Massa
· 11 months ago
Make the SDK freely available as soon as possible. The sooner I have it, the sooner I can start populating your webOS platform with killer apps!
William Beh
· 11 months ago
I second getting the SDK ASAP. It could help bringing a set of application out as soon as Pre is out.
Laurent
· 11 months ago
release the sdk most of developpers are not in US and are not able to have a contract with Sprint, so allow developpers to buy unlocked hardware and install applications without any appstore but with a direct copy to the device think about iphone jailbreak and symbian hacked security and do not waist your time in this kind of things, it's only a matter of challenge for hackers so let people choose if they want to install unsigned apps ot not publish the OS source as you made for PalmOs in the good old days, there is no better samples think about java think about flash
Justin
· 11 months ago
Comment #3 is golden, see lots of feedback on the Apple AppStore.
Internal app distribution is also important; perhaps an Enterprise-installable appstore, or certificate program, to manage distribution/licensing?
The Installer.app and Cydia route is also very interesting for Jailbroken iPhones. A common installer program that the user can add application sources to. SO if my company has private apps, allow me to enter the URL of their internal app server/authenticate the device to it and install from there.
In addition to trial periods, the ability to "return" a program may be interesting.
I'd also second #2; there's surely some mechanism to obfuscate the code, but would like to see some additional info on the app packaging (a la SDK) :) to know how much of a concern that is.
Finally, please keep the dialogue going. More conversation = better affinity from developers = more apps early on = greater opportunity for marketshare.
Leonard Lin
· 11 months ago
I'd agree that getting an SDK out sooner rather than later would be the best thing. Having that behind a PDN signup and alpha NDA would be fine, but it's hard to really comment on something w/o being able to play around with it (from the leaked shots on BGR, it looks baked enough and you're likely to get a lot more useful feedback from developers (not to mention apps) w/ a lot more people banging on it).
I'd also second developer oriented phones.
some SDK thoughts:
* hopefully the JSON-bus will allow an app to access/add observers on all the hardware/sensors and system events (ie, being able to say do processing when a message comes in, or when joining a new network, etc would be great)
* related, it'd be interesting if there were hooks to intercept events - ie, to be able to say automatically block calls based on recipient and location as an example use case.
* system calls like that would of course require a pretty robust security model - it seems like the right combination for that is a modal security model that requires user approval for access of specific services like location, telephony network access, etc (allow/deny w/ a "remember" setting). Policy/System-level lockdown would probably be wise for security concerns as well (only run signed code, no user-downloadable apps except via admin deployment, etc)
* Running bg tasks w/o cluttering up the card views? not a huge loss if not, but since the cards are linear, i could see it getting really crowded really quickly...
* I don't really care about backwards compatibility (while I had an original USR Pilot and used Palm's for years, my last Palm was a Treo 600), but it seems like porting something like POSE would be trivial if there was access to an X11 layer
Store suggestions:
* Collaboratively filtered/social recommendations for apps
* "Beaming" of apps; Zumobi in particular was showing off some really slick functionality for this last year
BTW, Congrats for directly engaging the community - that's a huge thing and I hope it continues. Looking forward to developer forums, and docs.
Dev Community suggestions:
* Allowing comments on docs a la the PHP model
* Using the dev blog w/ mini-tutorials highlighting interesting functionality w/ attached project files
* Allowing/encouraging devs to have an area for contributing sample code
yasth
· 11 months ago
Support for code signing, but also support for certificate trusting would be nice. Also please make the point of entry t your stre inexpensive. If I am going to develop a freeware app I will be disinclined to do so if I have to pay $100 to distribute it.
Revenue share should follow Apple if at all possible.
Support for Trials is essential. I could care less about coupons, though being able to drop price, and bundle (i.e. multipack) would be nice.
Updating functionality should be official and at least semi automatic without being annoying.
A Palm pick of the week or some other way for new/updated software to be noticed would be very nice. Also a dev contest to kick things off, and maybe occasional student/hobbyist contests (things that focus on sheer cool, irrespective of practicality, you'll get your tip calculators regardless, but wiimote controlled internet enabled sword fights are something different) might do well to maintain buzz.
Taisuke Odajima
· 11 months ago
Please develop somekind of Web API of the store (like the Amazon one) which would allow WEB developpers to create different services. It would be nice to have something more than simple RSS feeds.
Berthold
· 11 months ago
As a user: - Make installation of apps easy - Make uninstall of apps simple and complete
As a developer: - Please provide the SDK as soon as possible - Provide lots of examples and howto's - Provide standard re-usable components (also GUI) - A good emulator/simulator with debug capacity - Some kind of automatic random testing of the app - Provide tools to create installer packages - Use open standards for data storage (no PDB's thanks) - Hierarchical storage of data and file system
Keep up the innovation!
Jan Ehrhardt
· 11 months ago
I'm not sure, but I heard, that webOS will use WebKit as a rendering engine. Further information about this (which version, which features, which JavaScript engine) would be great. It would be great to get the SDK to play around (even with a beta) and then tell you, what I'm missing. For distribution I would like something like a Apples App Store, but without the restrictions, Apple has. Don't check each app, don't remove apps, when they are doning things your software also does. So I think freedom for developers and users would be best.
Numair Faraz
· 11 months ago
Don't ask for input. Android should show you that this just results in a lot of diffuse, pointless stuff, and you end up in a situation where many people feel that their "critical" requirement isn't met. Instead, come up with something fast, keep enough restrictions that you're not putting end users at risk of mediocrity (a MUCH bigger issue than dealing with the issues of developers, 90% of whom are little more than the virtual equivalent of gold-rush speculators), and iterate constantly. Give thorough rationale for all of your decisions and changes - that's the big area where Apple kinda misses the mark.
Seriously though, developers are the absolute worst group of people to be asking about questions such as installation processes. You need to put together a group of 12 year old girls, and their moms, from around the world.
Hippe IT
· 11 months ago
automatic version and dependency management like apt is delivering on Linux would be great. Or at least something as user friendly like the Firefox plugins.
John Moore
· 11 months ago
The biggest thing I've have been looking for in my next phone is some sort of auto-sync function. When I go home I don't want to take the time to add all my new pictures and data from my phone onto the computer I just want it to be there. I've overcome this lack of ability through various online services but it seems like this should be something handled up front. I realize this can pose a potential security risk to my data however manually moving this data around is no longer acceptable to me.
Shreekar Shetty
· 11 months ago
I am from INDIA . I don't kno if u guys sell ur phones here cos i frankly have seen none of them here . i have been following ur recently demo'd phone n bein a developer myself see huge potential in it .... my only request is don't concentrate ur phone n apps only on the US of A .... India and China r huge markets ( having more mobile phone users than the whole population of north america ) .... give ur phone a global edge .... google is way behind in that ... apple is far too dumb to understand the words "globalized phone" ....that leaves only windows phones as ur competition. Even NOKIA might do the same .... don't loose out to competition .... this a second beginning for PALM ...u guys should use it with a well planned n broad minded approach ... contact me if u need any other info bout the asian markets ... i ain't an expert ... i just follow technology closely .....
Kai
· 11 months ago
I have some suggestions:
1. Be honest with developers from Day Zero about where you want the platform to go. If you want it to be business productivity...say so. If you want to make a Gameboy...say so.
2. *Seriously re-consider* this notion of a "store"...or make it *optional*. Really. As an ISV, the biggest problem I have with...some other platform, is the notion that I cannot have a relationship with my customers that allows me to meet their needs because of middleman interference, can't add features or fix any bugs in anything like a reasonable or timely manner.
3. Do not waste precious resources on doomed-to-failure codesigning/drm schemes and tell us it is for "stability". We've seen exactly how stable and secure other platforms are for all of the wasted effort.
4. Codesigning: Do NOT make this more trouble than it is worth for developers. Your platform is going to need developers to *want to write for it* and not fight the toolchain to get the job done for *debugging*...let alone release.
5. It is Palm...consider the savvy user. While no one wants a platform that is a breeding ground for malware and phishing and such, please do NOT make a system that hamstrings users AND developers due to trying to fight this "spectre of evvvvviiilll". Allow users to set up their own Chains of Trust. If an app isn't signed, warn them...don't stop them from having a choice. This will cut out a large motivating factor and justification for subverting your platform.
You are of course, on your own with the whole carrier exclusivity thing :)
6. Do not play to the Lowest Common Denominator for a quick cash grab. Your platform will then become like another one; full of potential, ultimately ruined by being a dumping ground for kiddie apps and dreck in a race to the bottom. Planning on having a store? Do EVERYONE A FAVOR and don't even have a sub $4 price tier. If carriers are willing to cut deals to sell *ringtones* for $2, "AnswerTones", etc for a reasonable price...if digital music stores think a 3 minute music video is a $2 purchase...why not raise the bar here? Software development costs time and money and requires assets. You want webOS running the apps that the other guys aren't getting? Set up your system so that there is a reason *not* to write so-called "ringtone apps" in the hopes of turning a "quick buck".
7. Get an SDK out before March if at all possible. If devs are going to have to suck it up and buy hardware and contracts on a carrier that isn't the one they are using, they are going to need time to work it out...as well as any bugs, etc.
8. Make SURE your Dev Program is not a communications nightmare. If we need feedback, and have to pay for it...then answer our questions. It really, really sucks to be treated as a *liability* and not a *partner*...ya' dig?
9. While documentation might take awhile to get up to par...at least make sure your sample code is well commented. Most eager devs are going to want to "hit the ground running" as it were...don't make us curse you :)
10. If you all *insist* on being a conduit to the end user, remember that they are *our customers too* please. If you *have* to have a vetting process (sigh) please, PLEASE make it transparent to the developers! There is nothing more frustrating than the...'black box' methodology employed on another platform I am familiar with. No one wants their QA team pestered several times a day by eager devs and the way to nip this in the bud would be to provide adequate feedback to them throughout the process. "In,waiting,waiting some more,waiting some moooooooooor, zzzzzzz, Hello? Hello?!?!,Done" is really not enough.
Basically, all you have to do is NOT make the process a "love/hate" one and put the tools in our hands. We'll do right by you if you do right by us.
I am really...REALLY...rooting for this platform. I think Palm has so far put a lot of thought not only into the core implementation, but the user experience as well and it deserves the attention it has received. It deserves to be a contender.
But know this: while everyone made fun of Ballmer jumping around like an idiot screaming "Developers!"...we all as developers (even on other platforms) knew he was right.
We want to be your friends. Let's be friends :)
Alex
· 11 months ago
Like Guillermo said in #1, an easy access to the device for developers would be great. I do love my iPhone, but its so stupid that we have to jailbreak the thing to develop for it.
caje
· 11 months ago
Having tools to backup applications, and hopefully data, are both important. In the past, with my old palm Zire 71 pda, I collected something close to 50 apps. After the inevitable hardware crash, I'd have to restore my programs and data, and having a local backup was very handy. Providing unlimited downloads of previously purchased apps would also be nice.
Gothy
· 11 months ago
+1 for and early look SDK and dev phones
Stefan Constantinescu
· 11 months ago
Several things I would like to see:
1: Adding my own search engines to the universal search engine that pops up when typing at the home screen. If you want ideas on how to implement that then check out either Google Chrome or Opera Mini. The latter has a really nice function where if you click a text entry field that is attached to the search function of a website, you can click on options, add search engine. That search engine is now in your list of available search engines. Note how many search engines Opera Mini has by default too, Google and Wikipedia are great, think about adding IMDB, Amazon, etc.
2: There should be an RSS feed anyone can subscribe to to follow all the new applications being added to the Application Store.
3: There should be a forum on this website so that developers can talk to each other. Host an IRC server as well so developers can talk to each other in real time if need be.
4: Have a screen saver or something similar to that effect so that I can rapidly check the time and date. People my age (mid 20s) do not wear watches anymore. I would not want to unlock my device and squint at the thin bar on top to check the time. Also when I say show me the time and I date, show me the actual day as well. It would be nice to know it is Wednesday or something.
5: Make the SDK available ASAP is a given.
6: Switch this blog's RSS feed to full, there is no reason, zero, zip, nada, that this should be a partial RSS feed.
7: Improve the application startup time of your application. In the videos I saw the transitions were well and fantastic, but the startup time for things such as the contacts application or the web browser looks down right horrible. Granted the software is beta, but you should really fix this. Use a BlackBerry Bold or a Nokia E71 for reference.
8: The OMAP 3430 is a fantastic processor, please let a special developer or two create a media playing application that can handle the files that everyone my age grab from the internet, natively, without the need to transcode. You know what I'm talking about: XviD files.
9: Look at the Nokia Mobile Web Server. It is an S60 application you can install that literally turns your mobile phone into a server. What's the benefit? Say you take a photo, if your friends are subscribed to your RSS feed then your photo will be sent to their reader. The best benefit however is the ability to edit PIM data in a browser. I am not talking about sync, this is real time file editing, over the internet. The browser essentially acts as an IP based screen and keyboard for my device. Something tells me this is what you tried to do with the Foleo, but the market didn't want it or you failed to tell the story of the Foleo properly.
10: I hope there is an option in the browser to save web pages so I don't have to constantly go to a website to get a piece of text I want.
11: I hope there is copy and paste in the browser. I live in Finland so copy and pasting these long funky street names into Google Maps is a slow process of switching back and forth between the two running applications to get the whole 20 character word.
12: Call Google. Make sure that if the first person who buys a Pre and goes to gmail.com gets something that is similar to what iPhone users have. Should be easy to do.
13: The Application Store should use the notifications feature you presented during CES to tell users about updates.
14: The music application should have a similar navigation structure as the iPod.
15: Firmware updates should come over the air and I hope you have the right software in place so that in case the device loses power, or the user drops the device, whatever, the device does not become a paperweight due to a faulty firmware update. Look at how the HTC G1 operates in that regard.
16: I know you like natural designs, but shiny black plastic is a terrible fad that is close to dying. Please bring out something with matte paint, or better yet, aanodized aluminium like the Mac Book Pro. Look to your roots, the Palm Vx.
17: Look at what Nokia did with WOM World, a word of mouth marketing company. They sent review units out to bloggers, popular forum members and popular people on Twitter all over the USA, Europe and Asia. Most of those users loaded Flickr with pictures, YouTube with unboxing videos and wrote a ton of blog posts.
18: You need to make your platform sound more attractive to developers who are in this for the money. Their ratio is 30% Apple, 70% developer. Why not make it 20/80 or even 10/90?
19: Let me pin applications to the home screen. By that I mean let me make a "card" that can not be thrown out of the "deck of cards." This would be useful for weather applications and so on.
20: If you ever need a "community manager" or whatever they call "social media specialists" these days then shoot me an email. I loved each and every Palm I've ever owned and I'm glad to see you guys are coming back!
thomas hirsch
· 11 months ago
I'd quite like to see a report of which suggestions (posted here and otherwise) have been implemented or considered. Developer transparency ftw!
I have a slight query about the fees and such. In sunny old England, you often get plans that involve text (SMS) messages, internet data access and calls, as separate entries. However, as the Pre bundles all the messaging types into one place, it is possible you'd get charged differently for the different things, so you'd use an IM, facebook message or email rather than a text. How do Palm plan to strategise this?
Also, release dates and pricing please! And I know the States are the largest market but it wouldn't hurt to acknowledge "the rest of the world" and talk about details for those for whom Sprint means nothing.
Ricardo
· 11 months ago
Since this is a new platform and will be international, we expect it to support multiple languages. To do this, not only the phone must support Unicode (perhaps UTF-8?), but also the whole system must be ready to deploy / manage multiple versions of the same application in several languages.
It's awful how some sites hosts the same application in different languages as different applications! This soon becomes a mess: when you search their app database, instead of receiving "n" results, you get "n * (languages available)".
It's hard to comment more since we have no idea how the internationalization of apps will be / can be implemented: can an app be distributed in several languages being in a single package? Will users be able to download just the translation they want of an app if it becomes available weeks after the release of the first version? Will this be managed independently by the webOS, or by each programmer through each of their applications?
Keep up with the good work, folks!
Valere
· 11 months ago
About the Application repository, it could be a really good thing to be able to change the Applications repository location to another.
This way we could be able to create professional application repository for professional and internal uses (for easy install,deploy,..)
Think about how the apt-get works in Debian, that's could be very very usefull!
Thomas
· 11 months ago
Please allow advanced users to configure the device for direct installations. I know you want to get money out of the whole appstore thing, but companies and power users want to have the phone under their control (like with Windows Mobile). So don't tell me or them what to install. Certification is great, but there must be a way for me to say "Ok this app is from an unknow source (maybe myself?) and it is not signed, but I definitely want to install this!". Don't make the same mistakes apple made.
And I know it's a bit off topic, but I think I heard this whole "we only support HTML and Javascript for development" thing not so long ago from apple. Can we use the OMAP3430 processors OpenGL ES® 2.0 and OpenVG™ features with HTML and Javascript? Can I write/port a PalmOS Emulator with Javascipt?
Fernando Lins
· 11 months ago
I think it is important to let users from anywhere to download/buy the software and to provide payment methods that are more practical for international users, such as Paypal, not relying only on international credit cards. Also, when distributing the SDK, make sure there are enough examples of how to achieve several different behaviors and actions on the OS. As said above, the earlier the SDK is out the better. And please provide a framework for syncing desktop and mobile software :)
Gavin Baker
· 11 months ago
From what I've heard, I think the Pre could be the blend between utility that WimMo offers with the fun the is the iPhone.
Features past cut and paste, tethering, apps running in the background are assumed, the biggest hurdle will be how the pre deals with media. One of the beautiful features about the iPhone that I use all the time is the iPod. While I primarily use it to listen to podcasts, I won't really be all that interested in needed hypothetically run 3 programs to download and sync my new podcasts. e.g. iTunes integration if possible.
Michael Houghton
· 11 months ago
I think the main concern I have is that any approval mechanism
- doesn't seriously create delays deploying an app
- doesn't stop me deploying an app on a first-class basis to a small number of clients. I build bespoke web apps and I'd love to build serious small-scale webapps for clients but (using the Apple App Store as an example) a store that has an approval scheme that involves a test for 'wider utility to the market' is obviously going to stop me pushing out a useful app for a few dozen people, and yet the 'ad hoc' alternative they offer is likely to limit me should an application become just too popular to be delivered that way
- won't need endless re-approvals for minor bug fixes.
In short I think I'd prefer to see three tiers:
-stores where you (or your partners) approve everything, decide what they are going to sell and not sell
- a system for approving _me_ as a small-scale developer for Symbian-style code signing. (In practice if your users have the ability to install an app which is unsigned, with suitable caveats displayed that would be even better, but I accept that this isn't going to be popular
- an 'enhanced browser' equivalent to BigFive etc. that can be used for truly ad-hoc mobile admin screen development.
As a final plea can I ask you to co-operate with projects like PhoneGap, where feasible? I understand that you want to push your platform (and I'm seriously pleased to see it) but developers of websites have to deal with the devices their customers have already deployed.
harold campbell
· 11 months ago
Please take into consideration that your devices will be sold in other countries outside of the US (e.g. Jamaica and the rest of the Caribbean); as such, we should (JUST AS EASILY) be able to opt into your developer programs and also be able to purchase content (app, music, movies, etc) from which ever store or application that is created.
Judah Diament
· 11 months ago
an Aptana plugin for Pre would be nice, but first and foremost - get the SDK out ASAP so we can kick the tires
Jon at zhiing
· 11 months ago
With smart phones, it is all about location. I hope Palm realizes they are going to need to have to offer location as a seamless integration for developers within the SDK. If you think about it, that is one of the tipping points of iphone. GPS is not good enough because it barely works. I hope Palm allows Skyhook Wireless integration, so those of us developing location applications can get on deck and offer rich experiences. Not just consumer and social apps, but enterprise as well.
Tom W
· 11 months ago
I would like to see the ability for providers to collect recurring payments for their apps. Both the Android and iPhone stores do not currently allow this but we believe that it makes sense to provide recurring updates to the application for a recurring cost.
Eric Bresie
· 11 months ago
This might be a mix including some wish list items but...
As an interim for SDK release, are there any specific javascript frameworks (dojo, etc) or other APIs (QT, Gnome, Java/JavaME, LWUIT, Google Gears, GWT, etc) that would be widely used as part of this? This might allow people to start developing.
Maybe some understanding of the underlying kernel level info that the webOS is built upon would also be nice (if it's Linux based, then I suspect some of this may already be getting addressed in a Linux mailing list or some other place)
If much of this is HTML, CSS, and Javascript based, providing some form of templates and/or lots of example for development would be nice as well.
Good javascript debugging support would be nice. Maybe some loggin capabilities to help debug things.
Good security management to insure any applications don't start distributing information it shouldn't.
Documentation...documentation...documentation. (Getting started, Programming Guide, Framework, API, "Palm Zen", updated for "Palm Pre Zen").
Make sure that any application specific files (data files, resources, etc) are easliy identifiable and removable to prevent having left over files all over the place if the application is removed.
I gather applications have a web presence somewhere and a cache/local presence of applications. Clarification on how these thing occur might be nice.
Automatic updating of local versions recommended (although rolling back to previous versions might also be necessary if the new "feature" don't agree with a person's tastes).
Bill MacAdam
· 11 months ago
It is essential for my company to be allowed to maintain our current distribution system. We sell reference data and it is not intended for consumer distribution. It would violate our contracts with our data providers if we placed our applications in an app store environment.
There has been no discussion of databases on the device. It would be helpful to know how data is stored within webOS.
I am hopeful that there will be a way for users to continue to run existing PalmOS applications. We have roughly 2000 users and most of them are technophobes who hate change. It will surely make for a more graceful transition.
Finally, I would like to echo the concern for source code protection.
Stephen
· 11 months ago
I work for a company that has been developing mobile software for 10 years. We support just about every mobile platform. The #1 issue on any platform for us is installation. On some platforms we estimate that we are losing 40% of possible revenue due to tech support costs, dissatisfied customers, and return customers because installation is so problematic. This is the #1 reason why the iPhone is so appealing to us. We have had 0 installation issues on iPhone with 10x the usage of BlackBerry where we are constantly dealing with install issues.
The second point I would like to make is ease of 3rd party app discovery. Once again this is a huge reason we have been successful on iPhone. People can find our app. The extreme opposite is symbian. Most people who own a symbian phone don't know what they have, let alone that it can run 3rd party apps. Our iPhone app way out sells our symbian app even though symbian has 10x the market penetration of iPhone. So if you make a VERY good app store 3rd party developers can actually make more money on WebOS than on some other platforms even those WebOS won't have as much market penetration right away.
Finally, make sure that future versions of WebOS are backwards compatible. Don't end up with something like Symiban s60 v1, v2, and v3 that all require different builds of an app. This is difficult for us to maintain and it is confusing to customers. If you do end up having to do this please make sure that your store makes this completely transparent to users.
As you can see we most concerned about how the user discovers and gets their app than about any 1 feature of WebOS. We have found this to be one of the main determining factors in mobile 3rd party app profitability.
Thanks!!! Stephen
Casey
· 11 months ago
I'll second Jon's comment about location, having 3 apps on the iPhone and 1 on Android, it's all about location. Without location, the mobile platform is little more than a small laptop. Really differentiates and enhances the mobile experience.
Thomas Boerkel
· 11 months ago
Application installation over Bluetooth (from PC or other PalmPre) must work.
Speaking of Bluetooth: rSAP profile is necessary for business users.
Backup of everything to local PC must work. Over USB or Bluetooth. I guess not everything is accessible in USB drive mode.
Native SDK is a must. Why make the same mistakes as Apple? Also, this is needed for a PalmOS emulator or cool games.
Where is the video camera application, like in my Treo 680?
Thomas Boerkel
· 11 months ago
Please give us a JavaVM!
ashebanow
· 11 months ago
Hey folks, thanks for the great comments. But please, let's try to keep the focus on the issues around application distribution. We'll open additional topics to discuss issues like the SDK as time goes on.
george
· 11 months ago
Hi I develop a transit application for the Toronto area and would love to see Skyhook location services on the device.
Ame
· 11 months ago
The biggest thing I need regarding applications and distribution....they need to be available through the Palm site as well as "in the cloud".
I would like to see some kind of "sync" app to get things off of my home computer that is NOT synced "to the cloud" and onto my Pre. So if that's an app that runs on the Pre and has a platform both on Mac/PC, that connects to some kind of online hub, great. In a sense like the .Mac/MobileMe setup.
Chris
· 11 months ago
Here are the things I like about the Palm OS platform, that need to be carried over. 1. Open development. Anyone can develop an app, and they can charge for it or not charge for it. 2. Open distribution. Anyone can distribute an app form their own website, and are not required to go through an App Store. This is great for people looking for beta testers of their software, and trying to debug issues. 3. Open Platform. Palm releases just about every bit of information on the Palm OS platform, making it easy for developers to write just about any kind of program, including enhancements, and bug fixes to the OS. (This seems to be also the case for Android).
New things. 1. Ease of install. The App Store method introduced by Apple is a great idea, but is too restrictive. Its the only way to distribute an app. This is great for people wanting to make it easy for users to find an install their apps, but should not be the only solution for installing an application, as was mentioned above.
The HTML Jave it great for quick and easy development, but a C++ API is going to have to be in the roadmap, or most of the really good developers are going to look elsewhere, like Android, Win Mobile, or the iPhone.
Jaron Horst
· 11 months ago
* Application installation and updating
I like the PalmSource install method personally. One, compressed package delivered OTA that automatically removed the old version and replaced it with the new one. Easy to work with and saves the end-use a lot of extra headache! A compressed, installer/updater/uninstaller would be nice with OTA updates.
* Ecommerce (purchase, trials, coupons, etc.)
After the iPhone mess, I would personally like to see Palm have an online (on-phone) store but not to limit/lock users into only being able to use it. An open market would be better in my opinion.
For stability, especially if a kernel-level SDK is developed, it would be nice to have programs signed as approved by palm. However, let users be warned of unsigned/beta software before installing. Kindof how Microsoft does their driver-signing for know, safe drivers.
Anti-pishing and walware detection I believe should be highly considered as most of the above take advantage of security holes in an OS to install and hide themselves! As much as it is a users problem for getting the malware, Palm, as the manufacture (especially since companies can't write native apps!) needs to be pro-active about protecting their customers.
* Browsing and searching for applications
Doesn't really matter to me!
-----
Also, I would like to add my support for a native SDK in addition to the Mojo SDK as I would like to see companies like Hobbyist Software or StyleTap develop nice, kernel-level applications for the Pre.
matro
· 11 months ago
- provide the SDK as soon as possibile, even if it's beta;
- provide SDKs for trials and sw upgrade;
- allow non-US developers to produce as soon as possible;
- provide developers the Pre hardware device before worldwide distribution and possibly at lower price: I can develop good quality software, I can't afford a Pre if it costs as the iPhone. you made this with Treo in the past but only inside US: please allow non-US too;
- appstore facility is ok, but allow installation with no limits, as an advanced option at least.
jan
· 11 months ago
I have 4 points to make:
1. The hardware is probably capable of 3d, and real time sound synthesis/ capture. I pressume the camera can also record movies. I'd really really like to see that palm makes C objects for these 'power features' and expose these objects in javascript API's.
That has never been tried before: on the iphone, if you want to do 3d stuff, you'll need objective C. I think that allowing javascript developers to take part in developing 'the real cool applications' will make the palm platform strong!
2. What I am worried about: if the entire Mojo GUI is programmed in html/js, how can you prevent evil developers from developing a screen that looks just like an OS screen. (Phishing on the device it self). I hope you have some solution for that.
3. Allow free app catalog entries for free open source programs!
4. a developer should be able to install his app on his own real phone. (Not just run it on an emulator).
Steve
· 11 months ago
1) Get an SDK out ASAP... as in yesterday. I don't care if it is beta...or only an 80% solution. If you want us to care about this platform, stop just focusing on the handful of developers that you have brought under you kimono.
2) HTML and JS? Seriously? That a nice sandbox...especially for casual development, but I have business level apps that I could start porting tomorrow...IF I had access to a JVM on the phone. There is no way that we are converting some of these multi-platform applications to HTML and JS. Could we put the back end on a server and just host the presentation tier on the device? For many of these applications, that approach is very undesirable. There has got to be a way out of this sandbox. You did build on Linux...right?
3) For the sandbox, you want me to write and deliver apps that are essentially source code? Seriously? I would expect industrial-strength obfuscation at the very least..if not some sort of encryption with an encryption-aware JS engine.
4) Give us a rich widget and widget group library. The value of user interface consistency cannot be understated.
5) Cut and paste support (with associated event- and state-aware menus) everywhere...at least for text.
6) I am also very much behind the idea of an OS-implemented software trial mechanism, but am concerned that once it is broken (and it will be), we will all be screwed.
Steve
· 11 months ago
Ohh...okay, just for distribution? Okay, well all of the comments above AND allow installation from either your catalog OR a third-party. I like the ability to include a link to a PRC that someone can hit from blazer and then install...but I cannot stand the five dialog boxes that they must dismiss to make it happen.
The ability to go around the central app catalog is going to be critical for business applications distributed within the enterprise.
Kai
· 11 months ago
I know, I know...distribution :)
I think tho, Palm needs to, before even remotely asking us about this, make very *very* clear what the deal is with Javascript/html/css development to clear up misconceptions that are running rampant.
Most devs I've talked to are making a LOT of assumptions about how limited they are going to be...because of previous platform limitations.
You guys should probably address this ASAP, to set the proper expections and *manage them*
Brian
· 11 months ago
Don't lock down the developers under a strict NDA. Have a forum area that allows developers to support each other, and have them monitored by Palm so that they can chime in as well with support. As many examples as possible, start simple and build on them.
JasonB
· 11 months ago
1. Having an app store is one thing. You can see how successful it is with the iphone and G1 Android. However, I would really appreciate having the option to download an app from a third party online store like handango.com with the ability to install it ourselves.
2. If you were to come out with an app store on the phone, I would really like you to allow us to view what type of services the app will be using and also provide user ratings for that app.
3. How about uninstalls... If someone was to uninstall an app that saved data to memory, could you have it setup to prompt us if we would like to keep the existing data or delete the files associated with that app?
William
· 11 months ago
- Give the users an easy way do download trials, an easy way to purchase/download apps, a way to leave feedback about apps, a way to remove apps from their phone and get them back again (without having to buy the app again).
- It would be nice to be able to install apps another way besides the App Catalog. This would allow companies to develop proprietary apps and distribute them privately.
- +1 on not distributing apps as source code.
- It would be cool for an application's information page (in the App Catalog) to allow for things like images of the app (like the iPhone) and video if you've got it.
- Syncing is a must. You should also be able to sync all your apps to your computer(s).
- All-in-all, the most important part is that it's easy for users to find and obtain the apps they want. This will ensure that people start buying/downloading apps and will continue to do so, ultimately building a community around the Pre. That's why the iPhone is doing so well. Protecting dev's work is also important.
* Non-distribution-related suggestions
- +1 for releasing the SDK asap. I don't think I have to explain why that's a good idea. And by asap I mean yesterday.
- Games have proven to be a big money maker on the iPhone, and a lot of those games are pretty graphically intensive. An SDK that allows devs to make games like that would be great.
- A soft keyboard would be nice for sideways typing.
- Flash? Pretty please?
david
· 11 months ago
- Release SDK as soon as possible, developers need some time to prepare software, test every capability; - Allow getting the device (Palm Pre) for not US only developers. There are a lot of developers who are not located in US and they still would like to develop applications.
Mark
· 11 months ago
Regarding this blog, a full text RSS feed would be much appreciated.
warren shufeldt
· 11 months ago
as a developer dedicated to applications highly tailored to either a single client or a very small number of clients i would like to be able to install directly to my own phone or to my clients phones, eg this is a belly-to-belly, )-( , environment - small is beautiful and simple
Titus Blair
· 11 months ago
Hey Palm I am excited about the SDK! You should really look into Skyhook for your location based services, we are releasing some cool apps like JobCompass shortly using their technology.
John Moore
· 11 months ago
I would really like a application downloader that allows you to queue up the applications and content you wish to download. For those times when you learn about a new app but don't have a higher speed connection. Or if you're just to busy to dedicate time to a new program, but would like to check it out some Sunday morning.
Mathew Cucuzella
· 11 months ago
As Linux desktops are growing in popularity (and apparently being the designated Linux advocate) there are two must-haves for me to consider this phone (which I'm very interested in):
1. Although having access to multiple cloud based services is a modern idea, having direct import, sync and backup to my Linux desktop (through Bluetooth or USB) is critical for me to make sure I have local copies of my data. 2. The SDK must be able to run under Linux.
IMHO this would put Palm ahead of all other phones in an increasingly important area.
Ian Cinnamon
· 11 months ago
I am the CEO of Cinn Dev, which does iPhone development.My navigation apps have been ranked #22 of all paid apps on the App Store (check out Car Finder). One of the keys to my success was the indoor hybrid location systems available from Skyhook Wireless. With Skyhook, my customers can find their precise location indoors where GPS in not available. I would bring Car Finder over to the Palm platform if you were able to bring Skyhook's system to the Palm platform.
Nick
· 11 months ago
SDK can probably compute information on how processor- and radio-intensive an application is. (so that developer can optimize) App catalog should give a hint at this.. give information to customers on the power consumption of an app.
As a customer, if I can choose between too fancy "always on" applications.. I need : price info rating (as in app store) effect on my battery life
In the end, I might prefer to pay for a more optimized application.
Palm should encourage optimized and energy savvier apps, which are essential to a better user experience and won't deteriorate palm's products perception.
Mario
· 11 months ago
Please make some way for private apps, maybe a developer can put it directly on the device (I understand the issues with this) or make it so that an app can only be downloaded by poeple with some sort of key or maybe a specific "PRE". The device will sync with a computer directly with out the internet Right? Just cheking... Now that we clear that, I'm super excited since the code is from very common sources please make it so that we can write a app for the device and other for the computer and can be sync, and of course a link (in a cloud or in a company intranet server) between several computers running the same app for organization and coordination.
All the above is beacuse I allready thinking of apps for companies, that can be made to fit perfect.
Also, don't forget us, the outside of US developers. Give us a hand ok. Thanks
Dave Brown
· 11 months ago
The poster (#19) who said that developers are the wrong group to ask about the store is right on. Notice how most of the posts are asking for the SDK? Sure, most of us probably know what HTML/CSS/ECMAScript are and can do in a browser... but what I want to know is whether there's a prayer of creating, say, a usable VNC client for Pre.
But on the topic you wanted, let the store be at most a premium distribution method. Don't lock developers into it.
Ian Baird
· 11 months ago
Any support for Skyhook-style service when GPS is unavailable?
mike
· 11 months ago
I could REALLY use recurring billing in a palm app store. Specifically the ability to bill x dollars monthly for an app that provides real time sports information. One time purchases are fine for games or standalone apps but not for apps that have some back end service that the app depends on. This opens up a whole new segment of applications that just make no sense to build for the iphone and it's one time purchases.
Trials would be great as well.
Stefan Constantinescu
· 11 months ago
I thought Palm hired people to count the number of taps (in this case touches) to get something done.
If I lift my finger off the screen just pop up a fancy box that has two options: copy and paste.
None of this tapping a menu, clicking edit, and then selecting the copy or paste function.
I know this is the wrong post to leave such a comment, but I'm hoping someone at Palm will read this and forward this to the right people.
I know you're better than this, prove me right!
wally
· 11 months ago
The most important thing to me is the ability to install any app I want. It is okay if you don't accept some apps into the App Catalog, but just let me download an app off any developer's website direct to my Pre and I will be happy. This is such a big differentiator from the iPhone to me, and is a big reason I love my Treo. There's no reason you can't have the central, high-visibility, super-easy-to-use App Catalog and allow sideloading for power users at the same time.
Looking forward to the SDK!
Alice
· 11 months ago
So many questions could be answered if you release the SDK so we can give development a whirl. Next, web development can be very insecure if done incorrectly. Make sure you have something in place to keep schmucks and malicious folk from doing things they shouldn't. Lastly, don't force distribution through an app store and don't gag developers.
Steffen Frost
· 11 months ago
We are developers for a social network for carpool and ride sharing on mobile platform called Carticipate. Location awareness is obviously important. Besides GPS, which doesn't work indoors, cell tower triangulation and reverse IP geolocation functionality via Skyhook is a must.
awaybbl
· 11 months ago
1) Flexible distribution: app store, side-load, "my site", other distro hubs.
2) Community: app reviews (only if they actually own a copy)
3) No limit on number of installed apps (limit should only be based on available storage)
4) Ability for users to categorize/sort/move/lock/unlock/rename apps and their location.
5) App approval: Allow developers to post any and all apps. Let the market decide, not some process bogged down in red-tape and uncertainty.
6) Beta's, Timed Trials, Private distibution: Allow developers an easy method to distribute to private groups without restriction. Prior to app final distribution, developers need a good way to get their app out for public testing.
7) Of course give developers all the tools possible. Templates/emulators/IDE/SDK/API/CodeBuilders, and anything else that they need.
8) A bigger piece of the pie: If Palm wants developers to move their apps from iPhone, their commission needs to be competitive with Apple's. The 30/70 split is excessive, as well as their $250/region(7!) holdback. A better approach would be 20/80 with $100/month holdback for entire payout.
Wladimir
· 11 months ago
Any chance of using Python (or even perl6) instead of Javascript to develop?
In case it isn't, any technical chance to port the ParrotVM to a system like WebOS?
ern0
· 11 months ago
TOC 1. Native API 2. Sync/Backup
1. Native API
Sooner ot later, there will be native apps hacked to the phone. So, it's better if *you* release the native API.
PalmOS (from 5.x) runs M68000 code as default, and ARMlets can be called from the 68K emulation. For example, there's an image wiever written in 68K - except the JPEG decoder, which is an ARMlet (also, it contains the 68K alternative routine for Dragonball systems); and it runs fast as hell.
WebOS should use similar mechanism. The interface should be as thin as possible, and it should forbid access to things which are accessable via JS API. Then, the holy day, when the newest Palm machine comes out with a new architecture (processor/family), only this thin layer (API + processor) must be emulated by the new version of the WebOS operating system, and only these small routines must be rewritten (or just recompiled) by sw developers.
(I don't wanna live in a world, where I can't play Bejeweled with my Palm. Okay, it's not my favourite game, but others'.)
2. Sync/Backup
I have a Treo650 now, and I don't sync. I am happy with the weekly backup to card.
What about a small client which runs on the PC/Mac, and can receive the backup requests initiated from the phone? This small stuff should listen on your internet server as well, for a small fee (or for free: "We know that this gadget is the most important toy of you, dear customer, so we provide you not just reliable hardware but professional backup solution. We know, that your next phone will be a Palm too, so we're in continous partnership, and that's why we provide it for no extra fee. It's a kind of marketing cost for your next phone buying.").
Also, if something make apart professionals from amateurs, it's the backup solution. Professionals take care of their data. Let turn Palm customers to Professionals.
William Pietri
· 11 months ago
Hi! Could you please make your RSS feed include the full post, rather than just a teaser? Most people force a full click-through because they're after the ad revenue. For people in you situation, who are trying to reach as many people as possible, it's better to let the ideas flow as widely as possible.
Thanks!
ashebanow
· 11 months ago
We're working on it. According to the wordpress admin panel it is a full text feed. The developer site team even sent me the php code and it looks like it should be full feed. Very strange. Anyone who knows wordpress ins and outs who could offer tips?
Nicolas Bouliane
· 11 months ago
It would be primordial to offer testing devices for a low price without a contract with Sprint. There are craploads of developers outside the US
Nicolas Bouliane
· 11 months ago
As said earlier, give the developers a bigger piece of the pie. 20% will get you far more support than 30 or 40%.
Free development tools too, of course.
FuturDreamz
· 11 months ago
OS-wide plugin support would be nice, so you can have things like a text expander that automatically changes a pre-set code that you type anywhere into a long phrase like an address.
Dimitry Z
· 11 months ago
I would like to see an local storage system similar to Google Gears where a web app's data and code can be used when the device is not connected (and updated when the device is). Also, and this is a long shot, can registered developers get a developer discount on the Pre? That would help you distribute the phone to people who would then build valuable tools for it.
Anyway, I hope this phone makes it here, to Canada. I wouldn't want to drive south and have to unlock it. This is an exciting product but unavailability would mean at least 1 less developer.
Dimitry Z
· 11 months ago
ashebanow, if you want the full post to appear in the feed, you can either change the 'rss_excerpt_length' option in wp_options table to 0 or you can replace the following line:
The latter solution will give you full HTML formating in your feed.
Leonard Lin
· 11 months ago
A quick pull of the feed looks like it's being redirected to FeedBurner. Are you sure you don't have "Summary Burner" accidentally enabled?
Adam Backstrom
· 11 months ago
@ashenbanow Might not be a WordPress issue. Is Feedburner trimming the full feed?
I would want to be able to choose my delivery method. I am not sold on the "app store" concept, I would like to know that I can distribute my apps to my users without an intermediary party. I like the model from my Palm V and Tungsten days: I found apps via the web and installed them myself. I'm not against an app store, but I want to install other apps if I so choose.
ashebanow
· 11 months ago
Thanks everyone for the pointers on the full-text issue. I passed them on to the folks who actually have admin privs on the system :-)
Prakash
· 11 months ago
1. The dialer app should run in a separate thread. No other apps should interfere with normal phone functionality. Currently Centro hangs sometimes while receiving calls. 2. Sandbox functionality for 3rd party apps. 3. Code security verification or something... 4. Desktop development IDE like Visual Studio/eclipse (!) 5. Application uninstaller. 6. Animation Framework/Small Physics engine
Prakash
· 11 months ago
7. Online backup functionality 8. Online sync with/like Live Mesh Platform 9. Sync todo and memo with some website like rememberthemilk 10. firmware update/software update from Palm Desktop instead of connecting directly to internet. 11. Security Center application which has settings to say, block all internet calls..etc., 12. Firewall alerts
Stu
· 11 months ago
With regard to the app store, don't forget about all the Palm forums out there (TreoCentral,MTDN, Treonauts, 1SRC, Brighthand, etc.), that rely on software vendors to pay the bills (for the servers, writers, etc.). These forums do a lot for Palm in troubleshooting problems, promoting new products and applications, etc. It would be a shame to put them in dire straits due to Palm keeping all the software vendor income.
William Sun
· 11 months ago
East Asian Language (Traditional and Simplified Chinese, Japanese and Korean) Input Method APIs should be built-in. So developers can port the needed input methods the the webOS (to name a few: changjei, pinyin, zhuyin, hiragana, Chinese or Kanji writing recognition, etc). What even better is to build in some common and free-to-use input methods.
Chuck McKinnon
· 11 months ago
For me, the three biggest things needed from a development standpoint are:
1) An SDK, as soon as you can manage it 2) An emulator 3) The ability to intercept low-level commands like gestures that currently trigger Palm-built apps.
Like others, I'm based outside the U.S. (in my case, Canada) so access to an unlocked phone for a reasonable price at launch would be very beneficial.
For the ecommerce store, the ability to:
- Invite people to opt in/out of mailing lists (in my case, one for free upgrade notifications, and another for new product / bundling announcements). - The ability to bundle, offer discounts, and to call a remote server to GENERATE MY OWN REGISTRATION CODES so I can offer free trials and not be locked in to whatever (certain-to-be-hacked) "protection" Palm may have planned.
Really, the ideal would be to make the Palm app store optional. I think because you're on the receiving end of so many support calls, it gives your executives a distorted view of how stable/unstable the majority of existing Palm devices tend to be. It really depends on the kinds of applications one installs, and I don't think the rest of us should be penalized by forcing us to use an app store of your making.
To speak bluntly, I don't want to subsidize Palm to the tune of double-digit percentages of my sales. Your existing partnership with PalmGear is usurious; it's over 50% of revenue.
Code signing is fine; again, I hope you allow for the fact that many of us have Comodo or other code-signing certificates we use for our Windows installers or conduit DLLs. Having to purchase a separate one for use in your store would be redundant and probably more expensive than it needs to be.
In general, keep the platform truly open and don't try to fight us as we help make the Pre a bestseller. We all want to see this succeed.
jp
· 11 months ago
I was wondering about the potential for viruses and mal ware. I am not computer programmer but i was wondering if it would be possible for a "shady" developer to hide something in their code that would steal poeples information and send it over the web. i know you guys already look for these kinds of things in the software already but my concern comes from the pre's ability to multitask.... maybe now developers may be able to hide parts of malicious codes in seperate programs. Once the are all up and running at the same time something maybe activated. i am not sure how plausible that maybe but who knows what people can come up with... thanks
flash god
· 11 months ago
Let Flash run on WebOS!
Since it's already running on Android and Windows Mobile anyways, by not allowing it to run you guys will seriously miss out on the best virtual machine EVER. This would fix one of the biggest criticisms of Web OS which is that Mojo doesn't do much for developing excellent games on mobile. With flash, all kinds of existing games could be ported and new ones will surely arise. Don't be fools like Apple, get flash to run on it, flash 10 is on its way to mobile (especially android and winmo) in 2009, don't miss out!
Nigel Runnels-Moss
· 11 months ago
I could write an entire book on how to, and how not to implement a new development platform. I spent the last year and more banging my head against Apple's iPhone development programme whilst leading a commercial software development project, which was severely hampered by Apple's draconian and ham-fisted strategies.
1) NDA - learn from Apple's mistakes: you want developers to talk to one another, to create their own resources, to leverage each other's experiences. If your NDA is like Apple's was, then I as a developer couldn't even tell anyone I was developing on the iPhone. Don't make the same mistake please.
2) Distribution - be flexible. Learn from Apple's mistakes: if you try and lock things down all you will do is alienate a large portion of your developers, who will go develop for some other platform. If you must have an app store, go ahead, but it should be one of many mechanisms for getting apps onto the device, not forced upon the user.
3) Concurrency model - learn from Apple's mistakes: if you want to limit background processing then there are many ways to do so without adopting a 'only one app running at a time, unless it's an Apple app' policy, which just makes things frustrating for the user and hamstrings the developers. We have many resource allocation/management mechanisms from back in the good old days of cycles being a precious resource - use one or more of them.
4) Iterating the public beta SDK - learn from Apple's mistakes: trying to guess what the hell would be broken every morning when you got into the office because the latest SDK drop arrived unannounced with an unannounced set of often major and poorly documented changes is just a stupid waste of time. Publish ahead of time when the next version will be released, and what API changes and bug fixes it will have in it, at least a week ahead of time, preferably two - development teams for projects any larger than trivial size need time to plan around changes to the SDK and not waste precious development resource, e.g. recoding stuff that we never would have done had we known that feature was being removed in the next SDK iteration.
5) Community - do it right. Employ people that have created successful developer platform communities in the past, learn from their experiences. Learn from Apple's mistakes: they had a great Mac developer community with driven passionate developers. Somehow they managed to avoid utilising any of their successful experiences of the Mac developer community when starting the iPhone programme, much to the developers' confusion and annoyance.
Off to go write a book now. Good luck, I'd love to see Pre succeed.
Igor S
· 11 months ago
1) Layered API: both the currently announced Web model AND more native APIs, perhaps under different distribution/security processes, but available.
2) Rich API. Full access to the platform's data and events: timers, PIM data (including PIM item modification events), calls, messages, location - shouldn't be limited compared to, say, Windows Mobile, Android or PalmOS.
3) Services: ability to create modules that run in the background, maybe without UI, survive restarts and low memory conditions (e.g., restarted by the system after being dumped out of memory), and are accessible from other applications/services via some protocols (like DLL, COM, CORBA, RMI, etc.)
4) Code and data persistence - no imposed dependence from the availability of a broadband (or any) connection. In other words, an application that doesn't require connectivity should be able to be started and work offline.
5) International availability of the devices.
JasonB
· 11 months ago
Now that the Palm Pre fire has died down. I definitely recommend that you guys post some new material very soon to keep those who are interested in the phone at the edge of their seats before we start to forget how great and awesome this phone is.
h0tzenpl0tz
· 11 months ago
release that phone!
ipsi
· 11 months ago
In terms of application distribution, I want to see the following:
1) Ability to install programs how I want to, without having to go through the App Store. Useful for Beta Testing or installing a program I found while browsing the internet and don't feel like hunting down in the app store.
2) Provide a sensible way of ordering applications and a solid search feature. I don't really care if I have to tap through to an advanced screen to get it, I just want it to be there. I want to be able to search by company, title, tags, price (user-defineable ranges please!), ratings, etc. And order them by similar.
3) Provide the ability for developers to sell bundles of content for an application, and for them to offer upgrade bundles as well, ideally with automated detection of what the user has installed already, and only offer them valid upgrade options. As a hard-and-fast example, take PlecoDict (http://www.pleco.com).
They offer Chinese/English dictionaries (and related) software. They currently have 8 dictionaries (I think) available for sale. The total price is over US$200 for all of them and not everyone wants to pay that much (and not everyone needs or wants them all), so they currently offer bundles. Buy 2 basic dictionaries for cheap, then you can upgrade and buy the more advanced/expensive ones later on. Or buy the more advanced ones first then buy the others later if you miss them. In each case, the price varies for the package and for the upgrades. More importantly, if you buy bundle B1 (including dictionaries D1, D2), you then pay a different price to get dictionaries D3, D4 than you would have if you had purchased bundle B2, including Dictionaries D4, D5, so it can get a little complex, but is also necessary.
4) Provide an easy way to do trials of applications but don't lock developers in. Again, taking Pleco as an example, some of their contracts may not allow them to make the full dictionary available for free, even if it's as part of a time-limited trial, so they instead reduce the functionality of the program (quite significantly) but let you use it for an unlimited amount of time.
5) Allow developers to re-use the App Store for a different purpose. Take eReader as an example - currently they have a huge number of books on sale, and I would love to be able to search through them all on the device and purchase them then and there. This would require the app store to be customizable and either pointed to a different database or to a different server or whatever. You would obviously need to allow developers to add extra search fields, like ISBN, Author, Publisher, Year, etc. It would make it quicker for developers to finish their program, and probably result in it being more full-featured.
6) Provide a 'Company' page accessible through the App store that lists any and all applications they've made available on it. Also allow for Open Source developers to have their own page, ideally with an optional 'Donate' button on it (some people would prefer money went to actual charities rather than themselves).
7) Provide a seperate category for OSS software, but mix it in amongst the regular results as well (while allowing people to exclude free apps) so that people who might not know what is or why they should investigate that category can still find potential solutions.
That's all I've got for distribution ideas at the moment. In terms of what else I want, there's the following:
8) Unicode support. UTF-8/16 is essential. If you haven't already got it supported then it's going to be hell to add it later...
9) An unlocked GSM model of the phone available for international developers, ideally running quad-band 3G (850/900/1900/2100). Why Quad-band 3G? Because not everyone lives in a country with EDGE support, and I can't imagine using this phone with only GPRS support. It's one thing to develop an app using an emulator, but quite another to actually use it in the real world and I suspect most international developers will struggle to make usable programs until they've actually got a phone they can carry around with them and see how their app performs in day-to-day use. I'm down in New Zealand, and I don't expect my carrier to be carrying this phone for some time after the GSM launch. While it's Vodafone, and if Vodafone UK picks it up I expect <acronym>VFNZ</acronym> will, it won't be a simultaneous launch, and may be several months behind the VFUK launch. It will also be more expensive, and I don't particularly want a carrier-branded phone.
This is more important than it might be with other phones because it's marketed pretty heavily as an 'always connected' phone, and importing the Sprint CDMA version won't be an awful lot of use as there aren't many Wifi hotspots where I live, and none are free. I imagine that's true for other developers as well.
10) An SDK that runs on Linux. I understand that you're basing this off Eclipse. Eclipse runs just fine on Linux, and the phone itself is Linux at the core, so I'm really hoping that I'll be able to compile and develop applications on Linux. I don't want to have to use WINE, I want native Linux support. I can deal with it if I don't (Windows in a VM), but I'll be sad if I have to do it that way.
In terms of what others have been asking for, I'm not terribly worried about code encryption/obfuscation (and will be turning that off if it's present), nor do I care about low-level development access. I would like that both of these things are present (I would like to see Pleco be able to be ported, and that requires that both low-level development access and code encryption/obfuscation are present. The first purely for speed, the second because of contractual obligations. Same for eReader, really), but I don't expect to make use of them myself.
Anyway, thanks for reading. I'm looking forward to this phone, and I hope it's out soon.
Paulo
· 11 months ago
- Release the SDK - with Windows, Linux and Mac version. - If possible, a phone simulator on SDK. - Show a lot of examples for learning
William Sun
· 11 months ago
We are still having some concern of the performance of Javascript-based apps... Will the source code compiled to machine code so that the performance be maximized?
As a user (not a developer,) I can't live without a PDF viewer App called 'PalmPDF'(Henk Jonas, metaviewsoft.de) I assume a single guy (or small group) in Germany created a single, free app that makes my Treo invaluable. -And an iPhone unthinkable. I keep years worth of handwritten notes on a SD card, with me at all times. Please consider allowing us users to have access to these small, 'home-brewed' apps that have always made the Palm platform unique, open, and versatile.
JB
· 11 months ago
From a current iPhone Developer:
- Let developers upload apps without having to wait for Palm to check them like Apple does. It's very frustrating having to wait for Apple when you have a fix that you need to get release ASAP.
- Release the SDK as soon as possible.
- Don't be like Apple and take complete control of the store. Google's model is much better. Give developers freedom.
- Provide devices without having to sign for a Sprint contract/plan (like Google does)
Keep up the good work. From what I've seen the Pre looks like a real winner.
Janice Kreidel
· 11 months ago
I would just like to echo what other developers have said: As a user I don't like "stores" that I HAVE to use. I used to buy from a really large "store" until they "forced" me to install their store on my 755p. Actually I would prefer to buy directly from a developers site. Been a Palm user since the III and want to keep using Palm's due to their great ability to use what I want to use on MY phone.
Mike Kabala
· 11 months ago
This has probably been mentioned, but just in case it hasn't, I'd like to see some sort of device emulator available for the PC so that I can develop and test most of my application there instead of having to load it on the device itself every time I make a change.
midtoad
· 11 months ago
1. Ability to connect a bluetooth keyboard, similar to the capability present in devices like Palm TX. This feature is lacking in the iPhone. It's so simple to add this feature, and will greatly increase the productivity of the device for those people that want to enter text of more than, say, 1 paragraph in length.
This bluetooth keyboard capability is also present in Nokia phones.
leonara
· 11 months ago
First,thx u guys bring this excellent gift to all palmfans! I come form china, Use Palm PDA and treo over 10 yaers. I have a few questions to ask right here which is the idea(thinking) of majority chinese users.
[b]1. [/b]Can new Webos support any Asian language,such as Simplified Chinese、traditional Chinese or Japanese?by which way to make it come true? original unicode?
[b]2. [/b] I konw u guys have already pulled out of china market,if the new webos support chinese, when u guys would come back to china as the #1 sellphone market of the world? form 01-01-09,china already to release 3G licenses,include CDMA2000 and WCDMA, I think the network is not a problem.
[b]3.[/b] The new Webos can't run any oldOS prc,maybe the only way is use simulator.Do u guys have any plan to make one which simulator can support chinese and let us run old app-old PIM and read personal information?
Kai
· 11 months ago
As for supporting an application infrastructure:
1) A central app store is quite welcome to ensure ease of distribution for low cost applications. Apple does it mostly right, the only improvement on their model would be to provide both a vetted area and a non-checked area, so that approval processes would not need to delay the release. This would also allow bug fixes to be released quickly.
As for the payment models that should be supported for applications, single app, bundling, upgrades and money-back are the only important ones that need support (trials are the same as money back).
2) There needs to be a way to distribute applications outside of the app store, especially for in house applications. For these, there should be a way to sign phones as company phones, and provide them with a link an in house app store. This would also enable applications to be locked down to in house phones, if so desired. (And on the other side, provide a company white list of approved apps for use on the phone)
A further use of company certified phones could be to support volume deals on applications, a company could use their in house software to buy licenses in bulk, and the normal app store could then use the company certificate to distribute these applications and provide the license management.
For testing applications, direct download would be the easiest way to support it.
3) For updates, there should be some kind of update service that checks the download location of each application for updates automatically. For this you might need to define the format of an update file to provide the update information, to be maintained by the application provider. This update check should be able to do automatic updates.
4) All applications should be signed, so that specific signatures could be blocked from the phone. There would be an opt-in palm blacklist, and you could block them yourself on either the phone or the in house store. There would be two classes of signatures, one free without any checking, and one where you have to pay a small fee, but where the contact details are verified. This is essentially security by auditing, you do not prevent people doing harm, but you know who they are to punish them later if they misbehave.
5) Since applications cannot be checked thoroughly in an cost efficient manner, there should be three access levels to the functions of the phone:
green: cannot do much harm, has web access, can add/change information to synergy (I imagine synergy to provide a join of multiple tables, and applications can add/maintain their own table to be included in the shared synergy view), have their private database, can exchange information only via cut and paste, and have access to rough location information only (at the county level).
yellow: more dangerous, gain access to detailed location information, can read all shared databases, but not modify these
red: access to everything there is available on the phone.
In addition, an application should be able to permanently gain read or read/modify access to a database by asking the security manager for permission. The permanent permission is key here, you should be bothered by the security only once during installation.
Moshe Katz
· 11 months ago
Please make two emulators: 1. For us to test Pre Apps on the computer. 2. For us to run old apps on the Pre.
These are my only prerequisies to buying a Pre
tph
· 11 months ago
Will there be any functionality similar to Microsoft's System Center Mobile Device Manager? www.windowsmobile.com/MobileDeviceManager - security/password policies, file/device encryption - remote management - OTA deployment - application dependency management - allow/deny only certain applications etc.
Bastiaan
· 11 months ago
Maybe it would be wise to make everything adaptable. With this I mean: real dynamic contacts. Most people have about 5 contacts they call or text most. These contacts should have some sort of speeddial function, or be at the top of the list automatically. So the people you call/text the most should always be on top of the list automatically.
beeno
· 11 months ago
First of all... VERY excited about the potential of the WebOS platform. I'm an enterprise developer and agree 100% with the point made by comment #13. We investigated being able to use iPhone's to deploy an internal mobile client solution, but the AppStore model made that impossible for us, including the learning curve for our developers. So... we had to stay with our Windows Mobile based devices :( It would be a very desirable feature if we could be able to deploy our own Enterprise App Catalog for WebOS based solutions.
It would be wise, Palm and developers alike, not to underestimate the WebOS platform for internal enterprise solutions. This is potentially a huge market being dominated by Microsoft at the moment, but as we all know, domination doesn't necessarily mean superiority. Using web based technologies is a smart move, since this can potentially enable even our group of web developers to easily extend their expertise to a mobile platform.
Sorry, struggling to stay on point here, but a question was raised earlier regarding database support. Please could this be answered, as it would obviously make a big impact on the viability of Pre in the Enterprise.
Back to deployment... An App Catalog is not a bad idea, but app distribution should definitely not be limited to just that model. That having been said, lack of regulation or control could potentially lead to a whole bunch of nasty apps breaking my Pre. If I want to download and install an app, I'd like to know if there's a degree of risk involved. This is where security certificates issued by Palm to approved apps can prove helpful. That can give developers the opportunity to develop a partnership with Palm and be rewarded for writing good apps. But, if you still want to write bad apps, or couldn't care less about that program, no one will stop you from writing apps... but your users will be notified that your app is not certified.
Just a thought to throw in the pool.
Now... where's that SDK!?
Stefan Constantinescu
· 11 months ago
The RSS feed has been fixed, thank you!
Bill Principe
· 11 months ago
Please. PLEASE. Write a "Palm Card" application that allows old Palm OS software to run on a Pre card. Yes, I know. Most old Palm stuff will not take advantage of the cool features of the Pre, and some will probably not run under any compatability feature, but if you want to attract current Palm/Treo/Centro users, you had better allow them to keep their familiar things. Then, as new software is developed for the Pre, we can drop the legacy software.
Jeff Kendall
· 11 months ago
** Bundle StyleTap or provide the feature asked for in post #119. The Pre as presented appears to violate the promise of a path forward for Garnet users on Palm's next OS.
** If apps are html, can they be written in php? If the pre supported php which in turn had access to the filesystem, I would think there is not much an app could not do.
** Also some means of source obfuscation is required to securely distribute apps.
** Even if apps can be written in php, which reduces source code exposure to "view source" in the browser, is there a way to make sure users can't view the php source in some sort of file viewer on their Pre? I can't see a developer spending a lot of time and $ writing an app if it can only be distributed as plain text without any protections. For instance, a user could easily comment out the trial period or copy protection portions of the javascript, php or html.
Peter
· 11 months ago
1. Strongly request to release the SDK to community as soon as possilbe even a beta version 2. We need to see what can be done with Mojo to decide whether we are going to support palm pre or not. 3. Don't emulate the legacy plam software on palm pre , this will affect new software sale, that eventually affect the altitude of programmer to support palm pre or not.
Don
· 11 months ago
In some ways this is putting the cart before the horse.
I'd rather see the SDK before I even worry about app distribution. Allowing the use of native code would make the SDK much more useful and could solve, for some people, the point raised in #2 (yes, I am one of those people). Provide a alternate SDK that would allow developer's to chose the best tool for the job or allow the development of alternative tools/languages. It should not be as straitjacketed as Apple or Android platforms (what remained nice about PalmOS).
Well OK: 1) allow installation of code directly on the phone: do not require it to be loaded off a server. 2) Make it clear at installation that software is not "certified" and the risks involved. 3) Make it possible for software developers to use the technology of your app store on or from their web site for both limited distribution (e.g. internal enterprise, specific customers, trials, etc.) and retail distribution. 4) make optional, the use of your app store.
Mike Cane
· 11 months ago
1) Do NOT stick your nose in eBooks as Apple did. Books have been established longer than tech. Tech guys should stay OUT of judging eBooks.
2) I'd like to see a Twitterfeed for new apps. That would be helpful while at the desktop.
3) Have a web-accessible version of the store. This is a huge shortcoming of both Apple and Google. Again, this is great when at the desktop (or even on the road, using a netbook!).
Stan Slaughter
· 11 months ago
Will "Palm Mojo" be available for desktop website construction as well ?
holysilence
· 11 months ago
I ask this as a Chinese Palm fan, please please please make the webOS have a good support on Chinese(display and input), and please please please make it sell in China. we want nothing more.....
Ross King
· 11 months ago
As a developer and User;
1. Update the public site to explain how the multiple Calendar/Email services work. What I want to know is can I have an Exchange (Work: Email, Contacts, Email), POP Email, Imap Email, A desktop app that syncs my home Outlook calendar and contacts but keeps them all in their own playground. With the ability to select what i want to overla[p and to add items and pick from a drop down where to save them.
2. The ability to download and install personaly written apps. The ability to self sign or have no signature and let them run once I have click on the am I sure box. (Would be great for internal apps for companies who support this device)
The SDK should be a fully functional simulator including the ability to dial phoen # and have the phone go thru the motions in making a call (Even though it is just going thru the motions)
Ross King
· 11 months ago
If it doesn support now it should in the near future
1) Ability to at least open and view Word, Excel, Open office versions also. 2) Ability to view PDF (As an extra feture would be advanced support for PDF forms processing)
Wladimir
· 11 months ago
Allow developers to run Parrot Virtual Machine on your machine!
www dot parrot dot org
The perl comminuty is eager to do stuff on new platforms(i include myself)
A "official" palm post on perlporters could do the trick. Besides, is right on time, being that Parrot (the VM that runs Java, .NET, Python, PHP. Perl, Perl6, Basic, and many more) is about to be released to 1.0 in march.
Giving the WebOS that kind of power will open the gates to a HUGE number of apps, still keeping the advantage of reusing the skills that are more popular, and adding more (A lot of people knows BASIC, a lot more PHP, etc etc)
Luis D
· 11 months ago
Hi, I jus have a couple of suggestions. I currently us a Samsung Instinct. It's an ok phone. The problem is that there is no mobile broadband. So I will really need wi-fi. Also, my wife has an HTC Touch. I honestly have to tell you that something that I wish my Instinct had is the copy/cut and paste function from windows mobile.
Kai
· 11 months ago
I'd like to see another post or follow up to this one, to be honest at this point.
I would also like to re-re-re-stress that Palm needs to start Day Zero full on developer relations; I cannot begin to go into the horrific nightmare some other platform is causing, on MANY levels, and the seeming outright contempt that is being shown to developers...a lot of which the general public has no clue about (yet).
We need more information, and preferably, before March. We need tools, guys.
Peter Morales
· 11 months ago
auto-update: I see apps not as installations but as update streams, so a robust auto-update, to keep all apps in one's phone in sync with latest revs, and with any OS revs.
segment app store: people self-organize into tribes so cater to more than one 'Palm tribe': 'experimental' apps, vetted apps, 'private' (enterprise/organization) apps, etc.
CasBoy
· 11 months ago
WebOS demo is really fantastic! But it still have a long way to go to success. It's a good opp to interact with WebOS developers to show my opinion here. I really want Palm to success! As an user in China, I care more about internationalization of WebOS. A very important reason that Palm is not as popular in East Asia (China, Japan, Korea) as in North America is bad support for Double Byte Character Set (DBCS) support. Even latest unreleased OS6 never mentioned build-in DBCS-support. We have to process & display based on 3rd party software. Althrough iPhone is not officially release in China, its build-in local language interface & display support lead to its popular in China (In my office, 5 out 30 staffs use iPhone, that is more than 15% market share. It's amazing considering about 600 million+ cell phone user, twice number of US population, in China!). 1. Build-in support for unicode is bottomline for a success mobile OS in East Asia. Extended support for other encoding system (such as GB2312, GBK, BIG5,JS-???) will be perfect! 2. Good develop enviroment for input method programing is also very important. Besides high-efficient PIM & rich software resource, an excellent input method with perfect keybord support (like MacroHard & HandEase) is a key reason for choosing treo according feedback from treo forum. But based on existing information, programer can't develop an input method with only HTML, CSS & Javascript. 3. Local installtion for non-Sprint (or rumored Vodafone) user.
Cera Stone
· 11 months ago
Previous Palm devices used to support device locking (JackFlash, JackSprat, FlashPro, etc.), but the new architecture has closed out the user who wants to modify their device (say, to start up with a particular program, or have a program always running, even if the device is reset to (now (assuming these things are possible)) factory settings) to fill a particular need (but really, this could be for ANY purpose). What I am getting at is that (referring to the Centro/Treo x) programs like Warden Security, TealLock, or SecureX cannot be put built into the operating system's image on a device except by a Palm insider! It would make the devices much less desirable by thieves (and more desirable to customers) to make device encryption >and< locking standard on all Palm Prē devices, not to mention give us peace of mind (this means using a good encryption algorithm, like Twofish, and not a decrepit one, like DES). Don't promise anything too early, either; take as long as is needed to implement these features (webOS, the developer SDK, etc.). If it takes until later this year, or even until next year, it's fine. It's okay to hurry sometimes, but not to rush. Also, it seems as though people abominate cloud computing (myself included). Dummy terminals in internet cafés and in huge organizations/corporations or banks are really the only places for such technologies. Just think of cheap people like me who pay for per kilobye cent usage of the internet on their phones (which makes downloading a megabyte's worth of data cost 10 dollars); I have a pretty good idea where this HTML/CSS/Javascript thing is going. I also think that the built-in trial functionality would be a very good idea, as Anderson (poster #3) describes.
Cera Stone
· 11 months ago
It seems as though people abominate cloud computing (myself included). Dummy terminals in internet cafés and in huge organizations/corporations or banks are really the only places for such technologies. Just think of cheap people like me who pay for per kilobye cent usage of the internet on their phones (which makes downloading a megabyte's worth of data cost 10 dollars); I have a pretty good idea where this HTML/CSS/Javascript thing is going. Previous Palm devices used to support device locking (JackFlash, JackSprat, FlashPro, etc.), but the new architecture has closed out the user who wants to modify their device (say, to start up with a particular program, or have a program always running, even if the device is reset to (now (assuming these things are possible)) factory settings) to fill a particular need (but really, this could be for ANY purpose). What I am getting at is that (referring to the Centro/Treo x) programs like Warden Security, TealLock, or SecureX cannot be put built into the operating system's image on a device except by a Palm insider! It would make the devices much less desirable by thieves (and more desirable to customers) to make device encryption >and< locking standard on all Palm Prē devices, not to mention give us peace of mind (this means using a good encryption algorithm, like Twofish, and not a decrepit one, like DES). I hope you guys take this into consideration for FUTURE devices, because it still pisses me off (although I admire Palm) every time I look at my handheld. Don't promise anything too early, either; take as long as is needed to implement these features (webOS, the developer SDK, etc.). If it takes until later this year, or even until next year, it's okay. It's alright to hurry sometimes, but not to rush. I also think that the built-in trial functionality would be a very good idea, as Anderson (poster #3) describes.
Constantine
· 11 months ago
I would like to see the ability to distribute beta versions of my application outside of the App Store like environment (as in sending an install file as an attachment) so that I can iron out all the bugs with my dedicated base of fans before exposing the application to all webOS enabled devices.
Ken
· 11 months ago
Is there any way this WebOS can be ported to the Desktop or at least a NetBook version? That would be so ideal!
Vadivelan
· 11 months ago
1. Please don't ignore developers from various countries across the world. Provide some option to purchase unlocked "pre" for development. 2. A contest like "Android Developer Challenge" will be great to attract developers in huge number and starts with a bang. 3. Most of the developers can't market the products themselves, so appstore going to be great help. Restricted installation ( allowing installation only thru store kinda ) will eliminate piracy completely.
beeno
· 11 months ago
I'm with Kai in post #131. I've been trolling this blog for weeks and no feedback,no new items, no updates on roll-out times. Come on Palm! If you're really serious about a comeback with the Pre, you're going to have to go the extra mile for us developers and give us GOOD reason to want to stay excited about your new platform. Enthusiastic developers WILL be the life blood of the Pre's success. Get enough developers frustrated and you've lost the fight before it even started.
Let's see a bit more bite to match up to the bark!
William
· 11 months ago
Every day you guys hold out that SDK from us you're just shooting yourselves in the foot. We're getting antsy. We getting tired of seeing nothing new on the blog or the developer site.
We're ready to start building apps. What's the hold up? A reply would be nice.
heyrobertdavis
· 11 months ago
Make it possible to post applications for a client as a third-party developer: the iPhone app store won't let me develop an app for a client, then upload it as the client from my location. The folks at Apple seriously told us we had to drive to our client's location and do it from their site! Please recognize the difference between app ownership and app development.
Mark Reid
· 11 months ago
I would like the "Universal Search" functionality to access built-in apps and be extensible to new apps. For example, you open the Pre and start typing. After searching the onboard databases and giving the options to search the web, additional options could be "add to tasks" or "add to calendar". New apps should have the ability to add their own options to this list also. Thanks for listening!
TreoKing
· 11 months ago
Will there be a way for developers to extend the applications shipped with the Pre? So if I wanted to add a menu item in emails or camera, would I have the ability to do so?
pietro
· 11 months ago
dears sir may you have already answer, my question is: this OS could run on a treo 650? i.e. could this upgrade and relaunch most palm device on market? and become the real challenge to microsoft?
ipsi
· 10 months ago
Bit more on non-European language support: Have it configured out of the box. Don't make me go hunting for 3rd-part software or trying to figure out which fonts I need and how to install the correct IME - it's bad enough on Windows Desktop and worse on Windows Mobile. Registry Editing just to be able to display a foreign language? Not cool!
Also, you really need to think about how searching will work for languages like Chinese. "But it's easy," I hear you say, "just match character for character!". Well, no. Not quite.
Take Chinese for example. If someone has a rare character in their name, it could take 30 seconds or a minute or more to find it in the IME. That's quite a long time if all you want to do is call them. What is far preferable in that case is to be able to enter the pinyin for their name - instead of (or in addition to) the characters - and have it come up with everyone whose name matches those characters. It's likely this will produce a number of results, but it's also possible to know multiple people with the same full name in English, though less common.
Same for Japanese and possibly Korean as well (and any other languages I've not mentioned). And let it be customizable! Provide your own default implementation, sure (in fact, please do!), but let others install their own. This would, for example, make it possible for people to map Characters to Cantonese pronunciation rather than Mandarin if they're in Hong Kong or Guangzhou or wherever.
This should also be applied to all core apps, and to 3rd part apps where applicable and possible. It would not be cool to have this functionality in the contacts listing but not in the calendar.
There's been a few people on here requesting Chinese support, and there's likely to be a lot more who'll buy the phone if it was available in China (for a reasonable price).
Though I'm likely to be buying the unlocked GSM version, assuming there's no major problems in the CDMA version and the SDK is sufficiently open and flexible. Oh, and assuming you're going to be selling an unlocked GSM version in a timely manner.
Finally, where's the next blog post? It's nice to have this one, but making a single big post and staying silent isn't what a lot of us were hoping for...
Lucas
· 10 months ago
Palm, we're waiting. Where's the SDK?
AmandaJ
· 10 months ago
Developers really need some info guys, the fire will die if you drag this out... you need to keep us excited and not let us forget how awesome this phone is.
Release whatever you have for the SDK TODAY, not tomorrow, not next week. Even if it's handwritten notes from the Palm development team... we need to know so we can make plans to support this platform or not. Really. And if that is not possible... maybe you don't have the handwritten notes yet... at least narrow the release date down... 30 days? 60 days? Come on guys... you KNOW how this works!
Henson
· 10 months ago
While I’m no serious coder, I have some strong opinions on the matter:
- Applications, once created, should be stored in a single zip file. The Palm should be able to recognize these zip files as programs (through an XML file stored inside) and automatically give the user options to install, upgrade, or remove.
- While having an official app store is important, the Pre should offer the same install options no matter how it receives a program: through a 3rd-party site, e-mail, or otherwise.
- Applications should be stored on the Pre in their own subdirectories, so that deleting ALL of an app is as simple as deleting the folder (or running an uninstall wizard that does the same thing in the background).
I’ve only made one program in Adobe AIR, but I love the way AIR recognizes an application, installs it, and offers to put links on the Desktop / Start Menu for you. Palm could do the same with their cards system.
Regis St-Gelais
· 10 months ago
Here are some of my needs, some have already been said:
1- Possibility to install applications directly from the user's PC (or at least the possibility to have our own store). Our applications are not for the wide public but rather for large corporations. They should not appear on a public store.
2- Possibility for an application to communicate directly with the user PC (thru the USB cable, not over the air) or if not possible, the possibility for the application to write to the mass storage drive that is visible to the PC.
3- Code hiding is imperative. We can't have our source code visible. It must be protected somehow. Commercial applications need to be protected. If me code is visible, you can count me out.
4- A good simulator/emulator for debugging.
5- An SDK available ASAP (even a beta, to see what’s coming) or at least an API function list or help file.
6- An official developer forum (web based and/or NNTP based)
Thanks
Regis St-Gelais
· 10 months ago
Adding to my post.
7- A unique device id (serial number)
Alberto
· 10 months ago
proper use of Synergy. I do not know if this has been covered or not, but I would like Synergy to ask me if which profile pic i would like to display on my phone to identify my contact. In other words, if it recognized that in both gchat and facebook my one contact has two pics, it would ask me to choose which pic to display. This way, whenever my friend updates their profile pic, it would dynamically change on my phone as well. Of course, still have the option to choose a pic saved on my phone, but that is old and boring.
Carlos
· 10 months ago
Please, release the SDK as soon as possible. I can't wait to start playing with it. It will be nice to have a virtual device to test the applications on the dev environment.
Nicxl
· 10 months ago
Not really Application Distribution related but, since you palm guys are supposedly reading this :
All this multitask multi app thing is great, but we NEED a way to close ALL or MULTIPLE apps at once.
After some normal activity I guess many apps will be open. For order freaks like me, that can be a problem. And there will always be the suscipion that active process will use juice and the need for the users to be reassured and that they can close the apps. Now if 15 apps are open that's really not pratical
Give us some gesture or shortcut to : - select multiple cards that we can throw away at once - kill all apps at once
Thanks!! I can't imagine myslef constantly throwing cards away which I'll do..
jman
· 10 months ago
This platform has me really excited. Please don't make the same mistakes Apple did with the iPhone OS, such as to quarantine apps so they can't talk to each other or share files. I also echo the plea that you provide developer phones as Google has done with the G1.
anonymouse
· 10 months ago
SDK SDK SDK :)
Richard Kimber
· 10 months ago
HTML 5 location & storage APIs and CSS 3 transforms & transitions. I'm not sure if WebOS is based on Webkit, but Webkit would be great.
andy_liu
· 10 months ago
my suggestion is , first release windows webOS simulator for many user , second use some application like "styletap" run old palm + symbian + WM5 some software on webOS .. let other PPC/Palm user can move to new webOS system .
We at Isaac & Young Computer Company (IYCC) maintain the IYCC Linux Distribution, and we are excited about supporting and developing for the Palm WebOS. IYCC believes Palm can differentiate itself in the market by embracing developers of both open source and proprietary software.
IYCC has no problem in principle with proprietary software (though we believe that proprietary code will soon go the way of the fax machine). However, as a developer, I don't want to depend by mistake on applications that are proprietary. Further, the more I use already written open source code, the smaller footprint my applications will have and the fewer bugs I'll have to fix. And of course, as a user myself, I want the freedom to use Free and Open Source Software to the greatest extent possible.
We have some "gotta haves", though, in order to be in a position to support the WebOS platform.
First, in order to develop the quality of seamless user experience that we deliver, we need the OS, including kernel, libraries, drivers, languages, and toolkits to remain open source to the maximum extent possible. There's really no better method of ensuring my applications integrate seamlessly than by looking at the source code of the software with which my applications interact.
Second, interoperability with Debian packages, using aptitude (or a GUI frontend to it) system to install .deb packages would *significantly* reduce our costs and speed development time. Debian's package management system is by far the best in the industry. Using the well-tested APT system will significantly reduce our costs, and provide a huge codebase of already-written apps and libraries to draw from. The RPM system is also very good, and its codebase is broad and stable, but nothing beats Debian's package management.
Third, applications that do not meet the Open Source Definition should be in a separate repository from those that do, and the user should be able to turn off the proprietary repository if the user wants so to do.
IYCC looks forward to developing for the Palm WebOS, and I personally am looking forward to being one of the first to buy a PalmPre!
BTW -- IYCC is able and willing to host and maintain repositories for WebOS, for open source and proprietary developers both. If the repository structure and package management are Debian-compliant, as I described in 161 above, we can set up a repository structure dedicated to WebOS within two business days.
Happy Trails,
Loye Young
phreich
· 10 months ago
Some thoughts: 1. The suggestion that apps be loadable outside a "store" is essential. 2. The suggestion to allow for apps to go through a "certification" process is a good one, and warning users that non-certified apps could cause damage is okay. Asking developers to pay something for the work involved to gain that certification is okay too. 3. allow for use of a memory expansion slot in devices.
Rick Simpson
· 10 months ago
I have been a Palm Users since my Palm 3X days (then 505, T-3, 650 and 700p - my wife have a TX). I have been wed to the Palm PIM / Callendar and this remains my most important tasks which is why and iphone is not so attractive.
One of the single best apps that my wife and I still morn the death of ... is DUAL DATE. The ability to quickly share our calendars over lunch with one beam was just so simple and .. well GREAT. Please bring back the ability for calendar sharing.
rick
Kai
· 10 months ago
Loye...not wanting to take this "there"...but:
This is not the place nor the platform for OSS Evangelism.
I believe there are more than a couple of Linux-based mobile platforms that in some way or another meet the "gotta haves" you posited; and from the looks of the consumer adoption, market penetration, general buzz and all measurable standards...the fact that they are OSS-y doesn't seem to be doing them any good.
As a (former) long-time advocate of OSS in general and Linux, specifically, at the end of the day, I can say that for commercial ventures aimed at end users in an environment where *quality and User Experience* trumps *quasi-religious* ideology, Palm would find itself in the same position as these other platforms if they came out waving a GNU-Style OSS banner.
Besides: only the Linux that it runs on is likely OSS; it is being used as an embedded os/HAL. The real action happens in the API's running on top of them, and I am willing to bet that Palm has no intention of free-for-all source access to that stuff...
Ken Young
· 10 months ago
With the iPhone, Apple has integrated the application download purchase and update flow with iTunes, which is great because it provides a platform that interfaces with the web and the phone directly for download and installation of applications and provides a high-speed backup and local sync conduit for music, video and photos. It's probably too late at this stage of the game, but I'm hoping Palm has a similar level of integration. It's conceivable that the application store could be housed entirely on the phone and application installation and data sync be done through synergy only, but I think it would be a much more convenient solution to have a desktop application that does everything iTunes does for the iPhone. Hopefully in a less resource intensive package. :)
Joshua Granick
· 10 months ago
Our biggest request is for a way to filter URLs (blocking bad addresses, or blocking all addresses and creating an allow list).
There are many great things on the internet, but there are many others that I don't want coming through my phone. Can you please add the ability to add even a simple level of parental controls to a webOS-enabled phone?
This is a huge deal in my family. RIM supports a similar kind of control in Blackberry devices, but I believe that Palm is better for many reasons.
We have not subscribed to internet access in our Palm phones because there is no filter. It would be wonderful to be able to purchase a Palm Pre with restrictions to allow internet access on only certain applications (such as Google Maps or yellow page searches) and to restrict the addresses available in the built-in Webkit browser to safe sites we trust.
I use Opera on my PC, and it lets you restrict access using a urlfilter.ini file. I am not sure if Webkit is designed to support a similar file, but all we need is to be able to configure a URL filter file and password protect it.
I believe that this functionality would open the door to a lot of conscientious parents who are concerned about allowing unrestricted internet access on their families' cell phones.
Rick
· 10 months ago
We need an sdk asap, distribution is secondary to sdk.
Having said that:
we need a way to make our private business apps private, to get them ONLY with a special code, or to specially authorized PRE's.
Recurring billing would be nice.
auto-version update would be nice.
as 11000 other people have said, html/css/javascript is all well and good, but I'm not putting proprietary business methods on a phone as source code. Javascript obfuscation if needed has to be industrial strength.
FutureNerd
· 10 months ago
Distribution things.
0) Less is more. The less Palm itself is crucial to distribution, the better. The fewer issues have to be solved with or in distribution methods, the better.
For instance, security. If applications are sandboxed well enough that they can't harm or spy on anything, that means fewer worries about installing something. Fewer road blocks and steps in both users' and developers' faces.
Apps shouldn't be able to even read other apps' data unless the user has knowingly (i.e. with a clear UI) set up a specific channel. Example: a GPS app should not be able to read my address book (even though that's obviously useful) unless I said so. The user should be able to allow all sorts of interconnections between apps, and between apps and things on the internet, usb drive, computer thru usb, computer thru bluetooth, etc.
If you can securely verify that an update comes from the author of the original program, then updates should be no hassle even though the user has given the software permission to touch things outside its sandbox.
1) Make a MIME type for "Palm Pre app". Clicking a Pre app link on any site means you are going through the install process. User should be able to "download only," and maybe install later, but this should be a menu item or option you turn on rather than in the face of average users.
2) Figure out how to allow Amazon to sell Pre apps.
3) Think about content (as opposed to application) installation. Seems like a lot of good content should be viewable with the built-in webbish software (including PDFs).
I'm with others here: the more standard formats you can handle, the less distribution hassle. I hate Flash, but as long as I can turn it off, or get it to quit in less than a second, I far prefer that to some indirect transcoding scheme. Figure out how to prevent plugins from stealing the CPU. Like I said: if you can handle the problem outside of distribution, less distribution problems. To put it another way: where web browsing works, distribution is solved.
Some content requires players or plugins. On the one hand, allowing the user to buy or download some content bundled with the right player is nice, but effectively forcing content developers to compile the content into a copy of the player is sad. More like a prerequisites thing or bundle in the sales sense. Prerequisites are good for things that depend on plugins, too. Also helpful for uninstalling ("You are uninstalling the only app that uses the RealPlayer plugin. Do you want to uninstall RealPlayer, too?" (yes))
Also, Apple got into the situation of censoring content based on their app-vetting process. Don't try to censor content on your own store. An "I'm 18" category is fine.
Btw, another kind of bundle is collections of files that go with something, like the directory full of css and jpg files that goes with a web page, or the Apple "packages" which are also just directories of files. Sounds like something you would have down already, just saying.
4) The phone should belong to the user, not the developers. In the following I am not saying users should need to muck about with settings files, but that the user should have the ultimate power.
That includes letting the user say no to updates, do manual downgrades, look at and change config files, prevent the app from contacting the developer's (or any) site. If the developer doesn't want the app to work in a true sandbox, the app can put up a dialog saying what it wants the user to enable. But the dialog where the user actually makes the choice must be controlled by the OS.
You might want a way for the user to give permission for the app to keep some data "hidden" from the user, but the user should be able to locate and delete any "hidden" files.
Hiding data from the owner of the phone is essentially DRM. It's impossible without special tamper-proof hardware, nearly impossible with. Don't get into the morass of trying to provide an impossible service to worried developers. That is, you can make a stab at hiding, but don't try to guarantee to developers that this is secure, and do not get involved in DMCA fights with hackers or pirates over DRM, third-party apps, or content.
Same with code obfuscation. It's fine to make a gesture at it although leaving it up to the developers themselves is better. There is no such thing as industrial strength obfuscation or DRM unless you mean the flailing strength of dinosaur industries. Don't waste effort on the DRM at the expense of regular developers who just want to deliver useful stuff.
The user needs the ability to delete all of an app's user data and config info and force it to reconfigure from scratch. (I don't mean muck with internal code files and necessarily expect the app to keep working!)
5) Obviously sync apps and content as well as data. Allow installing stuff that was downloaded (or produced) on the PC as part of the sync process.
6) Install from (and sync with) USB storage. (Might want to have content, apps, or apps' data selectively reside on a USB drive as main location (like with SD on a Treo). However, that requires syncing the USB drive with the PC. I know, this is drifting away from distribution.)
@acarback
· 10 months ago
Excellent starter and example applications in the SDK. Documentation is excellent and required. But I'm more a 'show me' and 'visual' kind of person.
Please release the SDK before the phone!
Gerry
· 10 months ago
1) Applications should be searchable by title, keywords, or description, so applications don't necessarily have to have a literally descriptive title in their product if they don't want to, and still have it be easily found through search.
2) An method to put in variable discounts depending on previous software purchased. If someone purchased my three most expensive applications, maybe I would want them to get a big discount on another one, or give away the cheapest one for free for being a great customer.
3) At the very least, a way to put variations of a product on one page instead of taking up multiple pages. Say I had a program with features ABCD. I know that most people will only want three of those features, but which one depends on the customer. It would be nice to have the ability to let them check mark which ones they want, and show the adjusted price, or at least see all the options in one page for an easy comparison. Same goes for a much simpler regular, deluxe, and premium editions of one product.
Kyle Thompson
· 10 months ago
As chair of my anesthesiology practice's IT committee we're looking for several features in our next smartphone choice (edited for this blog):
3.Good text and voice functionality. 4.Repetitive alert capability 5.Applications: a. Epocrates Patientkeeper (future consideration) 6.Battery life 12h+ at usage rates of text/voice 7.Handsfree capability? Voice dialing? (Colorado House Bill 1094) 8.Cloud backup & sync options? Encryption options for the cloud? (HIPAA compliance)
From what I've read, #3, #4, and #5a are covered.
Thanks, Kyle Thompson, M.D. South Denver Anesthesiology, P.C.
USR Pilot 5000->3Com PalmPilot Pro->Palm IIIx->Palm 515->Palm TT3->Treo 700p->HTC Touch Pro
tal
· 10 months ago
Java SDK; Would make the community much more vibrant and allow for more "serious" applications to pop. I can't think of you winning with just the webOS. Doing it later rather than sooner might be too late.
Loye Young
· 10 months ago
To proprietary developers:
Palm's WebOS is an open source operating system based on the Linux kernel. Using an open source operating system such as WebOS does not *require* that you provide the source code for your proprietary application. If users want to pay you for a cool application and they don't care about the source code, the users should be free to make that choice.
Proprietary apps need a tollbooth for downloads, while open source applications don't. For this reason, proprietary and open source applications need to be kept in separate repositories, but that separation is more important to developers than to users.
A well designed user interface can make the existence of various repository seamless. Various applications have already been written to allow users to add, modify, and remove repositories. Such applications are typically found in a "system settings" menu that most users don't bother to mess with because the defaults are usually sane.
To Kai (post 165):
If you read my earlier post carefully, you will see that I am far from a religious zealot when it comes to open source. My issues are practical and economic.
You will note that the title of this thread is "Application Distribution". My comments are addressed to the nuts-and-bolts of distributing software (of various licenses) on top of an open source operating system.
IYCC has a fair amount of expertise on these issues because we already distribute both open source and proprietary software as a part of the IYCC Linux distribution. We know well the issues surrounding a mixed proprietary and open source software stack, and we make our living serving OEMs and ODMs who want their customers to have a seamless experience.
IYCC sees Palm's WebOS as an opportunity to make mobile computing available not just to the techno-enthusiasts (ala iPhone) but to mainstream users who want their gadgets to "Just Work".
I am wondering if there is a pre-launch application development program that interested developers can participate in. We recently lauched a local search application(Superpages) for iPhone and would love to get our hands on Pre's sdk as soon as possible. Let me know.
Thanks, Uddhav
jwall
· 10 months ago
@Uddhav,
Our team is now aware of your interest in the SDK. We are currently in private prerelease, and if you haven't done so already, please be sure to subscribe to the RSS feed on the blog to keep up to speed on any news we may have about the SDK
Poetic intentions
· 10 months ago
Hey I can't wait until the Pre comes out but I have a few issues 1) I need a micro sd it's perfect for this type of phone i think thats what apple messed up on not adding that to the iphone 2)virtual keyboard is that included because I love the thing, yes a full on keyboard is great but I would love the option 3)auto update would be great having to update manually with my phone sucks plug it up update all that log on to the internet update waste o time with some sort of auto update your phone will be great.
Khaled
· 10 months ago
What about Arabic language and middle east availability ? Palm has a bad history with middle east releases
David Williams IV
· 9 months ago
I think the pre should be able to use both the finger and a stylus. Finger for easy use, stylus for accuracy.
David Williams IV
· 9 months ago
A full size SD (secure digital) card with sdhc support would come in big handy, then your information can be easily transportable, you could have more features, and you can easily view photos and videos taken on another device.
Richard
· 9 months ago
Palm Pre..... please add 1) SD card 2) ability to run Palm apps
HAL
· 9 months ago
Hi there, Palm people and PalmFans and PalmUsers! To the powers that be, I kindly suggest, or add my weight to previous suggestions, regarding features for the forthcoming (SOON :) ) Palm Pre sporting the WebOS:
Touchscreen with ability to switch between stylus or finger, at convenience of the user.
Palm OS (Garnet) full interoperability (and it was one of your "election days" promises, Palm, don't forget it)
Ability to admit native code aplications, instead of only surface-level .html and JS.
SDIO slot, with SDHC support (as in 32GB, i.e.)
GPS and WiFi (for the case you were thinking of ditching them)
The best online market resource, IMO, is a lean client app, or a simple browser bookmark. So feel free to use my opinion to endorse the death of the AddIt client and the death of the buggy MyPalm client. If much of this is now about getting online, all we need is a bookmark.
Apps installation: as of my experience, best practice is a clean download that can then be executed, but clean download goes first. That may do for e-commerce and upgrades.
Software security: OK, better leave it for Symantec or Trend Micro. Things are getting this far, anyway.
Come on, Palm people, you did it before. DO YOUR MAGIC!!!
David Lutze
· 7 months ago
Palm has dropped the ball by not making a portable keyboard that really works. I've purchased three keyboards and they have ALL been a piece of junk. The batteries wear out quickly and there is no way to plug them into a wall curcuit.
There are many college students that are tired of hauling laptops around. Many would opt for a Palm Pilot if there was a keyboard that actually worked. Even though I have had bad luck with them, I would quickly buy another if it was truly functional.
Is there anyone listening to me?!!!
boydoc
· 7 months ago
I would like to replace my Palm Tungsten but I do not need another phone. Please make a PDA available with plenty of memory for professionals to use as reference. Thanks
All Comments
This will help developers get more sales and higher prices by allowing people to see value in a prospective app and reduce the purchasing uncertainty inherent in many mobile apps. Also, it should help to reduce the frequent problem of crap apps from overshadowing the high-quality ones.
I would think that this is Palm's chance to "catch up" to other mobile platforms that have been meeting standards in user expectations out-of-the-box. (cough... Windows Mobile...cough). Palm needs to make this out-compete the various Microsoft-integrated platforms by out-integrating them with Microsoft (and etc) products natively. Also add all those features that just make the thing more useful (USB Mass storage mode). Also.. mmm.. hotsync needs to speed up... seriously. I never sync my centro because it never ever finishes (i'm guessing it would take an hour or more if i let it go). I know that's some heavy criticizm for something that is so close to coming out (and thus these considerations will probably be ignored), but I really want Palm to survive... I'm even considering applying to Palm when I graduate because it is my favorite mobile platform that I will always stand by. I don't want to see it die out before I get that ultimate GPS+Wifi+Bluetooth super Palm OS phone!
One last thing.... I'd like for my community-created apps to be runnable on this thing. I don't think you can run ARM code through HTML, Javascript or CSS. Is there any chance older apps will be compatible? (TCPMP - the free one, will need to be rewritten in the worst case; and TCPMP I swear is one of the hardest programs to compile from source, in my experience).
Thanks for taking my complaint ... and good luck with this new release!!
1) Ease of use for the customer to install (and try) on device or alternately purchase right away.
2) Ease for the customer to buy the app after trying (single click type methods are good for impulse buying and ensuring the process isn't abandoned).
3) Ease for the developer to communicate and provide updates to customers.
4) Commission rates that allow for the developers to succeed without extremely high volumes (i.e. the current 50-75% commissions are not conducive to a good developer economy).
5) An avenue for developers to promote their applications (paid or unpaid) on the store.
6) User awareness of available third-party software, and encouragement for them to install it.
Any methods that help these areas would seriously improve existing conditions that are faced by third-party developers.
Good to see the conversation continuing!
And I personally think trials are a great idea and something you don't really get on the other platforms other than Palm/Garnet and Windows Mobile.
most of developpers are not in US and are not able to have a contract with Sprint, so allow developpers to buy unlocked hardware and install applications without any appstore but with a direct copy to the device
think about iphone jailbreak and symbian hacked security and do not waist your time in this kind of things, it's only a matter of challenge for hackers so let people choose if they want to install unsigned apps ot not
publish the OS source as you made for PalmOs in the good old days, there is no better samples
think about java
think about flash
Internal app distribution is also important; perhaps an Enterprise-installable appstore, or certificate program, to manage distribution/licensing?
The Installer.app and Cydia route is also very interesting for Jailbroken iPhones. A common installer program that the user can add application sources to. SO if my company has private apps, allow me to enter the URL of their internal app server/authenticate the device to it and install from there.
In addition to trial periods, the ability to "return" a program may be interesting.
I'd also second #2; there's surely some mechanism to obfuscate the code, but would like to see some additional info on the app packaging (a la SDK) :) to know how much of a concern that is.
Finally, please keep the dialogue going. More conversation = better affinity from developers = more apps early on = greater opportunity for marketshare.
I'd also second developer oriented phones.
some SDK thoughts:
* hopefully the JSON-bus will allow an app to access/add observers on all the hardware/sensors and system events (ie, being able to say do processing when a message comes in, or when joining a new network, etc would be great)
* related, it'd be interesting if there were hooks to intercept events - ie, to be able to say automatically block calls based on recipient and location as an example use case.
* system calls like that would of course require a pretty robust security model - it seems like the right combination for that is a modal security model that requires user approval for access of specific services like location, telephony network access, etc (allow/deny w/ a "remember" setting). Policy/System-level lockdown would probably be wise for security concerns as well (only run signed code, no user-downloadable apps except via admin deployment, etc)
* Running bg tasks w/o cluttering up the card views? not a huge loss if not, but since the cards are linear, i could see it getting really crowded really quickly...
* I don't really care about backwards compatibility (while I had an original USR Pilot and used Palm's for years, my last Palm was a Treo 600), but it seems like porting something like POSE would be trivial if there was access to an X11 layer
Store suggestions:
* Collaboratively filtered/social recommendations for apps
* "Beaming" of apps; Zumobi in particular was showing off some really slick functionality for this last year
BTW, Congrats for directly engaging the community - that's a huge thing and I hope it continues. Looking forward to developer forums, and docs.
Dev Community suggestions:
* Allowing comments on docs a la the PHP model
* Using the dev blog w/ mini-tutorials highlighting interesting functionality w/ attached project files
* Allowing/encouraging devs to have an area for contributing sample code
Revenue share should follow Apple if at all possible.
Support for Trials is essential. I could care less about coupons, though being able to drop price, and bundle (i.e. multipack) would be nice.
Updating functionality should be official and at least semi automatic without being annoying.
A Palm pick of the week or some other way for new/updated software to be noticed would be very nice. Also a dev contest to kick things off, and maybe occasional student/hobbyist contests (things that focus on sheer cool, irrespective of practicality, you'll get your tip calculators regardless, but wiimote controlled internet enabled sword fights are something different) might do well to maintain buzz.
- Make installation of apps easy
- Make uninstall of apps simple and complete
As a developer:
- Please provide the SDK as soon as possible
- Provide lots of examples and howto's
- Provide standard re-usable components (also GUI)
- A good emulator/simulator with debug capacity
- Some kind of automatic random testing of the app
- Provide tools to create installer packages
- Use open standards for data storage (no PDB's thanks)
- Hierarchical storage of data and file system
Keep up the innovation!
It would be great to get the SDK to play around (even with a beta) and then tell you, what I'm missing.
For distribution I would like something like a Apples App Store, but without the restrictions, Apple has. Don't check each app, don't remove apps, when they are doning things your software also does.
So I think freedom for developers and users would be best.
Seriously though, developers are the absolute worst group of people to be asking about questions such as installation processes. You need to put together a group of 12 year old girls, and their moms, from around the world.
1. Be honest with developers from Day Zero about where you want the platform to go. If you want it to be business productivity...say so. If you want to make a Gameboy...say so.
2. *Seriously re-consider* this notion of a "store"...or make it *optional*. Really. As an ISV, the biggest problem I have with...some other platform, is the notion that I cannot have a relationship with my customers that allows me to meet their needs because of middleman interference, can't add features or fix any bugs in anything like a reasonable or timely manner.
3. Do not waste precious resources on doomed-to-failure codesigning/drm schemes and tell us it is for "stability". We've seen exactly how stable and secure other platforms are for all of the wasted effort.
4. Codesigning: Do NOT make this more trouble than it is worth for developers. Your platform is going to need developers to *want to write for it* and not fight the toolchain to get the job done for *debugging*...let alone release.
5. It is Palm...consider the savvy user. While no one wants a platform that is a breeding ground for malware and phishing and such, please do NOT make a system that hamstrings users AND developers due to trying to fight this "spectre of evvvvviiilll". Allow users to set up their own Chains of Trust. If an app isn't signed, warn them...don't stop them from having a choice. This will cut out a large motivating factor and justification for subverting your platform.
You are of course, on your own with the whole carrier exclusivity thing :)
6. Do not play to the Lowest Common Denominator for a quick cash grab. Your platform will then become like another one; full of potential, ultimately ruined by being a dumping ground for kiddie apps and dreck in a race to the bottom. Planning on having a store? Do EVERYONE A FAVOR and don't even have a sub $4 price tier. If carriers are willing to cut deals to sell *ringtones* for $2, "AnswerTones", etc for a reasonable price...if digital music stores think a 3 minute music video is a $2 purchase...why not raise the bar here? Software development costs time and money and requires assets. You want webOS running the apps that the other guys aren't getting? Set up your system so that there is a reason *not* to write so-called "ringtone apps" in the hopes of turning a "quick buck".
7. Get an SDK out before March if at all possible. If devs are going to have to suck it up and buy hardware and contracts on a carrier that isn't the one they are using, they are going to need time to work it out...as well as any bugs, etc.
8. Make SURE your Dev Program is not a communications nightmare. If we need feedback, and have to pay for it...then answer our questions. It really, really sucks to be treated as a *liability* and not a *partner*...ya' dig?
9. While documentation might take awhile to get up to par...at least make sure your sample code is well commented. Most eager devs are going to want to "hit the ground running" as it were...don't make us curse you :)
10. If you all *insist* on being a conduit to the end user, remember that they are *our customers too* please. If you *have* to have a vetting process (sigh) please, PLEASE make it transparent to the developers! There is nothing more frustrating than the...'black box' methodology employed on another platform I am familiar with. No one wants their QA team pestered several times a day by eager devs and the way to nip this in the bud would be to provide adequate feedback to them throughout the process. "In,waiting,waiting some more,waiting some moooooooooor, zzzzzzz, Hello? Hello?!?!,Done" is really not enough.
Basically, all you have to do is NOT make the process a "love/hate" one and put the tools in our hands. We'll do right by you if you do right by us.
I am really...REALLY...rooting for this platform. I think Palm has so far put a lot of thought not only into the core implementation, but the user experience as well and it deserves the attention it has received. It deserves to be a contender.
But know this: while everyone made fun of Ballmer jumping around like an idiot screaming "Developers!"...we all as developers (even on other platforms) knew he was right.
We want to be your friends. Let's be friends :)
1: Adding my own search engines to the universal search engine that pops up when typing at the home screen. If you want ideas on how to implement that then check out either Google Chrome or Opera Mini. The latter has a really nice function where if you click a text entry field that is attached to the search function of a website, you can click on options, add search engine. That search engine is now in your list of available search engines. Note how many search engines Opera Mini has by default too, Google and Wikipedia are great, think about adding IMDB, Amazon, etc.
2: There should be an RSS feed anyone can subscribe to to follow all the new applications being added to the Application Store.
3: There should be a forum on this website so that developers can talk to each other. Host an IRC server as well so developers can talk to each other in real time if need be.
4: Have a screen saver or something similar to that effect so that I can rapidly check the time and date. People my age (mid 20s) do not wear watches anymore. I would not want to unlock my device and squint at the thin bar on top to check the time. Also when I say show me the time and I date, show me the actual day as well. It would be nice to know it is Wednesday or something.
5: Make the SDK available ASAP is a given.
6: Switch this blog's RSS feed to full, there is no reason, zero, zip, nada, that this should be a partial RSS feed.
7: Improve the application startup time of your application. In the videos I saw the transitions were well and fantastic, but the startup time for things such as the contacts application or the web browser looks down right horrible. Granted the software is beta, but you should really fix this. Use a BlackBerry Bold or a Nokia E71 for reference.
8: The OMAP 3430 is a fantastic processor, please let a special developer or two create a media playing application that can handle the files that everyone my age grab from the internet, natively, without the need to transcode. You know what I'm talking about: XviD files.
9: Look at the Nokia Mobile Web Server. It is an S60 application you can install that literally turns your mobile phone into a server. What's the benefit? Say you take a photo, if your friends are subscribed to your RSS feed then your photo will be sent to their reader. The best benefit however is the ability to edit PIM data in a browser. I am not talking about sync, this is real time file editing, over the internet. The browser essentially acts as an IP based screen and keyboard for my device. Something tells me this is what you tried to do with the Foleo, but the market didn't want it or you failed to tell the story of the Foleo properly.
10: I hope there is an option in the browser to save web pages so I don't have to constantly go to a website to get a piece of text I want.
11: I hope there is copy and paste in the browser. I live in Finland so copy and pasting these long funky street names into Google Maps is a slow process of switching back and forth between the two running applications to get the whole 20 character word.
12: Call Google. Make sure that if the first person who buys a Pre and goes to gmail.com gets something that is similar to what iPhone users have. Should be easy to do.
13: The Application Store should use the notifications feature you presented during CES to tell users about updates.
14: The music application should have a similar navigation structure as the iPod.
15: Firmware updates should come over the air and I hope you have the right software in place so that in case the device loses power, or the user drops the device, whatever, the device does not become a paperweight due to a faulty firmware update. Look at how the HTC G1 operates in that regard.
16: I know you like natural designs, but shiny black plastic is a terrible fad that is close to dying. Please bring out something with matte paint, or better yet, aanodized aluminium like the Mac Book Pro. Look to your roots, the Palm Vx.
17: Look at what Nokia did with WOM World, a word of mouth marketing company. They sent review units out to bloggers, popular forum members and popular people on Twitter all over the USA, Europe and Asia. Most of those users loaded Flickr with pictures, YouTube with unboxing videos and wrote a ton of blog posts.
18: You need to make your platform sound more attractive to developers who are in this for the money. Their ratio is 30% Apple, 70% developer. Why not make it 20/80 or even 10/90?
19: Let me pin applications to the home screen. By that I mean let me make a "card" that can not be thrown out of the "deck of cards." This would be useful for weather applications and so on.
20: If you ever need a "community manager" or whatever they call "social media specialists" these days then shoot me an email. I loved each and every Palm I've ever owned and I'm glad to see you guys are coming back!
I have a slight query about the fees and such. In sunny old England, you often get plans that involve text (SMS) messages, internet data access and calls, as separate entries. However, as the Pre bundles all the messaging types into one place, it is possible you'd get charged differently for the different things, so you'd use an IM, facebook message or email rather than a text. How do Palm plan to strategise this?
Also, release dates and pricing please! And I know the States are the largest market but it wouldn't hurt to acknowledge "the rest of the world" and talk about details for those for whom Sprint means nothing.
It's awful how some sites hosts the same application in different languages as different applications! This soon becomes a mess: when you search their app database, instead of receiving "n" results, you get "n * (languages available)".
It's hard to comment more since we have no idea how the internationalization of apps will be / can be implemented: can an app be distributed in several languages being in a single package? Will users be able to download just the translation they want of an app if it becomes available weeks after the release of the first version? Will this be managed independently by the webOS, or by each programmer through each of their applications?
Keep up with the good work, folks!
This way we could be able to create professional application repository for professional and internal uses (for easy install,deploy,..)
Think about how the apt-get works in Debian, that's could be very very usefull!
And I know it's a bit off topic, but I think I heard this whole "we only support HTML and Javascript for development" thing not so long ago from apple. Can we use the OMAP3430 processors OpenGL ES® 2.0 and OpenVG™ features with HTML and Javascript? Can I write/port a PalmOS Emulator with Javascipt?
Features past cut and paste, tethering, apps running in the background are assumed, the biggest hurdle will be how the pre deals with media. One of the beautiful features about the iPhone that I use all the time is the iPod. While I primarily use it to listen to podcasts, I won't really be all that interested in needed hypothetically run 3 programs to download and sync my new podcasts. e.g. iTunes integration if possible.
- doesn't seriously create delays deploying an app
- doesn't stop me deploying an app on a first-class basis to a small number of clients. I build bespoke web apps and I'd love to build serious small-scale webapps for clients but (using the Apple App Store as an example) a store that has an approval scheme that involves a test for 'wider utility to the market' is obviously going to stop me pushing out a useful app for a few dozen people, and yet the 'ad hoc' alternative they offer is likely to limit me should an application become just too popular to be delivered that way
- won't need endless re-approvals for minor bug fixes.
In short I think I'd prefer to see three tiers:
-stores where you (or your partners) approve everything, decide what they are going to sell and not sell
- a system for approving _me_ as a small-scale developer for Symbian-style code signing. (In practice if your users have the ability to install an app which is unsigned, with suitable caveats displayed that would be even better, but I accept that this isn't going to be popular
- an 'enhanced browser' equivalent to BigFive etc. that can be used for truly ad-hoc mobile admin screen development.
As a final plea can I ask you to co-operate with projects like PhoneGap, where feasible? I understand that you want to push your platform (and I'm seriously pleased to see it) but developers of websites have to deal with the devices their customers have already deployed.
As an interim for SDK release, are there any specific javascript frameworks (dojo, etc) or other APIs (QT, Gnome, Java/JavaME, LWUIT, Google Gears, GWT, etc) that would be widely used as part of this? This might allow people to start developing.
Maybe some understanding of the underlying kernel level info that the webOS is built upon would also be nice (if it's Linux based, then I suspect some of this may already be getting addressed in a Linux mailing list or some other place)
If much of this is HTML, CSS, and Javascript based, providing some form of templates and/or lots of example for development would be nice as well.
Good javascript debugging support would be nice. Maybe some loggin capabilities to help debug things.
Good security management to insure any applications don't start distributing information it shouldn't.
Documentation...documentation...documentation. (Getting started, Programming Guide, Framework, API, "Palm Zen", updated for "Palm Pre Zen").
Make sure that any application specific files (data files, resources, etc) are easliy identifiable and removable to prevent having left over files all over the place if the application is removed.
I gather applications have a web presence somewhere and a cache/local presence of applications. Clarification on how these thing occur might be nice.
Automatic updating of local versions recommended (although rolling back to previous versions might also be necessary if the new "feature" don't agree with a person's tastes).
There has been no discussion of databases on the device. It would be helpful to know how data is stored within webOS.
I am hopeful that there will be a way for users to continue to run existing PalmOS applications. We have roughly 2000 users and most of them are technophobes who hate change. It will surely make for a more graceful transition.
Finally, I would like to echo the concern for source code protection.
The second point I would like to make is ease of 3rd party app discovery. Once again this is a huge reason we have been successful on iPhone. People can find our app. The extreme opposite is symbian. Most people who own a symbian phone don't know what they have, let alone that it can run 3rd party apps. Our iPhone app way out sells our symbian app even though symbian has 10x the market penetration of iPhone. So if you make a VERY good app store 3rd party developers can actually make more money on WebOS than on some other platforms even those WebOS won't have as much market penetration right away.
Finally, make sure that future versions of WebOS are backwards compatible. Don't end up with something like Symiban s60 v1, v2, and v3 that all require different builds of an app. This is difficult for us to maintain and it is confusing to customers. If you do end up having to do this please make sure that your store makes this completely transparent to users.
As you can see we most concerned about how the user discovers and gets their app than about any 1 feature of WebOS. We have found this to be one of the main determining factors in mobile 3rd party app profitability.
Thanks!!!
Stephen
Speaking of Bluetooth: rSAP profile is necessary for business users.
Backup of everything to local PC must work. Over USB or Bluetooth. I guess not everything is accessible in USB drive mode.
Native SDK is a must. Why make the same mistakes as Apple? Also, this is needed for a PalmOS emulator or cool games.
Where is the video camera application, like in my Treo 680?
I would like to see some kind of "sync" app to get things off of my home computer that is NOT synced "to the cloud" and onto my Pre. So if that's an app that runs on the Pre and has a platform both on Mac/PC, that connects to some kind of online hub, great. In a sense like the .Mac/MobileMe setup.
1. Open development. Anyone can develop an app, and they can charge for it or not charge for it.
2. Open distribution. Anyone can distribute an app form their own website, and are not required to go through an App Store. This is great for people looking for beta testers of their software, and trying to debug issues.
3. Open Platform. Palm releases just about every bit of information on the Palm OS platform, making it easy for developers to write just about any kind of program, including enhancements, and bug fixes to the OS. (This seems to be also the case for Android).
New things.
1. Ease of install. The App Store method introduced by Apple is a great idea, but is too restrictive. Its the only way to distribute an app. This is great for people wanting to make it easy for users to find an install their apps, but should not be the only solution for installing an application, as was mentioned above.
The HTML Jave it great for quick and easy development, but a
C++ API is going to have to be in the roadmap, or most of the really good developers are going to look elsewhere, like Android, Win Mobile, or the iPhone.
I like the PalmSource install method personally. One, compressed package delivered OTA that automatically removed the old version and replaced it with the new one. Easy to work with and saves the end-use a lot of extra headache! A compressed, installer/updater/uninstaller would be nice with OTA updates.
* Ecommerce (purchase, trials, coupons, etc.)
After the iPhone mess, I would personally like to see Palm have an online (on-phone) store but not to limit/lock users into only being able to use it. An open market would be better in my opinion.
* Security (code signing, testing, anti-phishing, malware, etc)
For stability, especially if a kernel-level SDK is developed, it would be nice to have programs signed as approved by palm. However, let users be warned of unsigned/beta software before installing. Kindof how Microsoft does their driver-signing for know, safe drivers.
Anti-pishing and walware detection I believe should be highly considered as most of the above take advantage of security holes in an OS to install and hide themselves! As much as it is a users problem for getting the malware, Palm, as the manufacture (especially since companies can't write native apps!) needs to be pro-active about protecting their customers.
* Browsing and searching for applications
Doesn't really matter to me!
-----
Also, I would like to add my support for a native SDK in addition to the Mojo SDK as I would like to see companies like Hobbyist Software or StyleTap develop nice, kernel-level applications for the Pre.
- provide SDKs for trials and sw upgrade;
- allow non-US developers to produce as soon as possible;
- provide developers the Pre hardware device before worldwide distribution and possibly at lower price: I can develop good quality software, I can't afford a Pre if it costs as the iPhone. you made this with Treo in the past but only inside US: please allow non-US too;
- appstore facility is ok, but allow installation with no limits, as an advanced option at least.
1. The hardware is probably capable of 3d, and real time sound synthesis/ capture. I pressume the camera can also record movies. I'd really really like to see that palm makes C objects for these 'power features' and expose these objects in javascript API's.
That has never been tried before: on the iphone, if you want to do 3d stuff, you'll need objective C. I think that allowing javascript developers to take part in developing 'the real cool applications' will make the palm platform strong!
2. What I am worried about: if the entire Mojo GUI is programmed in html/js, how can you prevent evil developers from developing a screen that looks just like an OS screen. (Phishing on the device it self). I hope you have some solution for that.
3. Allow free app catalog entries for free open source programs!
4. a developer should be able to install his app on his own real phone. (Not just run it on an emulator).
2) HTML and JS? Seriously? That a nice sandbox...especially for casual development, but I have business level apps that I could start porting tomorrow...IF I had access to a JVM on the phone. There is no way that we are converting some of these multi-platform applications to HTML and JS. Could we put the back end on a server and just host the presentation tier on the device? For many of these applications, that approach is very undesirable. There has got to be a way out of this sandbox. You did build on Linux...right?
3) For the sandbox, you want me to write and deliver apps that are essentially source code? Seriously? I would expect industrial-strength obfuscation at the very least..if not some sort of encryption with an encryption-aware JS engine.
4) Give us a rich widget and widget group library. The value of user interface consistency cannot be understated.
5) Cut and paste support (with associated event- and state-aware menus) everywhere...at least for text.
6) I am also very much behind the idea of an OS-implemented software trial mechanism, but am concerned that once it is broken (and it will be), we will all be screwed.
The ability to go around the central app catalog is going to be critical for business applications distributed within the enterprise.
I think tho, Palm needs to, before even remotely asking us about this, make very *very* clear what the deal is with Javascript/html/css development to clear up misconceptions that are running rampant.
Most devs I've talked to are making a LOT of assumptions about how limited they are going to be...because of previous platform limitations.
You guys should probably address this ASAP, to set the proper expections and *manage them*
Have a forum area that allows developers to support each other, and have them monitored by Palm so that they can chime in as well with support.
As many examples as possible, start simple and build on them.
2. If you were to come out with an app store on the phone, I would really like you to allow us to view what type of services the app will be using and also provide user ratings for that app.
3. How about uninstalls... If someone was to uninstall an app that saved data to memory, could you have it setup to prompt us if we would like to keep the existing data or delete the files associated with that app?
- It would be nice to be able to install apps another way besides the App Catalog. This would allow companies to develop proprietary apps and distribute them privately.
- +1 on not distributing apps as source code.
- It would be cool for an application's information page (in the App Catalog) to allow for things like images of the app (like the iPhone) and video if you've got it.
- Syncing is a must. You should also be able to sync all your apps to your computer(s).
- All-in-all, the most important part is that it's easy for users to find and obtain the apps they want. This will ensure that people start buying/downloading apps and will continue to do so, ultimately building a community around the Pre. That's why the iPhone is doing so well. Protecting dev's work is also important.
* Non-distribution-related suggestions
- +1 for releasing the SDK asap. I don't think I have to explain why that's a good idea. And by asap I mean yesterday.
- Games have proven to be a big money maker on the iPhone, and a lot of those games are pretty graphically intensive. An SDK that allows devs to make games like that would be great.
- A soft keyboard would be nice for sideways typing.
- Flash? Pretty please?
- Allow getting the device (Palm Pre) for not US only developers. There are a lot of developers who are not located in US and they still would like to develop applications.
environment - small is beautiful and simple
1. Although having access to multiple cloud based services is a modern idea, having direct import, sync and backup to my Linux desktop (through Bluetooth or USB) is critical for me to make sure I have local copies of my data.
2. The SDK must be able to run under Linux.
IMHO this would put Palm ahead of all other phones in an increasingly important area.
One of the keys to my success was the indoor hybrid location systems available from Skyhook Wireless. With Skyhook, my customers can find their precise location indoors where GPS in not available. I would bring Car Finder over to the Palm platform if you were able to bring Skyhook's system to the Palm platform.
App catalog should give a hint at this.. give information to customers on the power consumption of an app.
As a customer, if I can choose between too fancy "always on" applications.. I need :
price info
rating (as in app store)
effect on my battery life
In the end, I might prefer to pay for a more optimized application.
Palm should encourage optimized and energy savvier apps, which are essential to a better user experience and won't deteriorate palm's products perception.
The device will sync with a computer directly with out the internet Right? Just cheking...
Now that we clear that, I'm super excited since the code is from very common sources please make it so that we can write a app for the device and other for the computer and can be sync, and of course a link (in a cloud or in a company intranet server) between several computers running the same app for organization and coordination.
All the above is beacuse I allready thinking of apps for companies, that can be made to fit perfect.
Also, don't forget us, the outside of US developers.
Give us a hand ok.
Thanks
But on the topic you wanted, let the store be at most a premium distribution method. Don't lock developers into it.
Trials would be great as well.
Why is copy and paste so complicated? http://www.youtube.com/watch?v=luMEMEBiL_g
If I lift my finger off the screen just pop up a fancy box that has two options: copy and paste.
None of this tapping a menu, clicking edit, and then selecting the copy or paste function.
I know this is the wrong post to leave such a comment, but I'm hoping someone at Palm will read this and forward this to the right people.
I know you're better than this, prove me right!
Looking forward to the SDK!
2) Community: app reviews (only if they actually own a copy)
3) No limit on number of installed apps (limit should only be based on available storage)
4) Ability for users to categorize/sort/move/lock/unlock/rename apps and their location.
5) App approval: Allow developers to post any and all apps. Let the market decide, not some process bogged down in red-tape and uncertainty.
6) Beta's, Timed Trials, Private distibution: Allow developers an easy method to distribute to private groups without restriction. Prior to app final distribution, developers need a good way to get their app out for public testing.
7) Of course give developers all the tools possible. Templates/emulators/IDE/SDK/API/CodeBuilders, and anything else that they need.
8) A bigger piece of the pie: If Palm wants developers to move their apps from iPhone, their commission needs to be competitive with Apple's. The 30/70 split is excessive, as well as their $250/region(7!) holdback. A better approach would be 20/80 with $100/month holdback for entire payout.
In case it isn't, any technical chance to port the ParrotVM to a system like WebOS?
1. Native API
2. Sync/Backup
1. Native API
Sooner ot later, there will be native apps hacked to the phone. So, it's better if *you* release the native API.
PalmOS (from 5.x) runs M68000 code as default, and ARMlets can be called from the 68K emulation. For example, there's an image wiever written in 68K - except the JPEG decoder, which is an ARMlet (also, it contains the 68K alternative routine for Dragonball systems); and it runs fast as hell.
WebOS should use similar mechanism. The interface should be as thin as possible, and it should forbid access to things which are accessable via JS API. Then, the holy day, when the newest Palm machine comes out with a new architecture (processor/family), only this thin layer (API + processor) must be emulated by the new version of the WebOS operating system, and only these small routines must be rewritten (or just recompiled) by sw developers.
(I don't wanna live in a world, where I can't play Bejeweled with my Palm. Okay, it's not my favourite game, but others'.)
2. Sync/Backup
I have a Treo650 now, and I don't sync. I am happy with the weekly backup to card.
What about a small client which runs on the PC/Mac, and can receive the backup requests initiated from the phone? This small stuff should listen on your internet server as well, for a small fee (or for free: "We know that this gadget is the most important toy of you, dear customer, so we provide you not just reliable hardware but professional backup solution. We know, that your next phone will be a Palm too, so we're in continous partnership, and that's why we provide it for no extra fee. It's a kind of marketing cost for your next phone buying.").
Also, if something make apart professionals from amateurs, it's the backup solution. Professionals take care of their data. Let turn Palm customers to Professionals.
Thanks!
Free development tools too, of course.
Anyway, I hope this phone makes it here, to Canada. I wouldn't want to drive south and have to unlock it. This is an exciting product but unavailability would mean at least 1 less developer.
// wp-includes/feed-rss.php
<description><?php the_content_rss('', 0, '', get_option('rss_excerpt_length')) ?></description>
with...
<description><![CDATA[ <?php the_content(); ?> ]]></description>
The latter solution will give you full HTML formating in your feed.
I would want to be able to choose my delivery method. I am not sold on the "app store" concept, I would like to know that I can distribute my apps to my users without an intermediary party. I like the model from my Palm V and Tungsten days: I found apps via the web and installed them myself. I'm not against an app store, but I want to install other apps if I so choose.
2. Sandbox functionality for 3rd party apps.
3. Code security verification or something...
4. Desktop development IDE like Visual Studio/eclipse (!)
5. Application uninstaller.
6. Animation Framework/Small Physics engine
8. Online sync with/like Live Mesh Platform
9. Sync todo and memo with some website like rememberthemilk
10. firmware update/software update from Palm Desktop instead of connecting directly to internet.
11. Security Center application which has settings to say, block all internet calls..etc.,
12. Firewall alerts
1) An SDK, as soon as you can manage it
2) An emulator
3) The ability to intercept low-level commands like gestures that currently trigger Palm-built apps.
Like others, I'm based outside the U.S. (in my case, Canada) so access to an unlocked phone for a reasonable price at launch would be very beneficial.
For the ecommerce store, the ability to:
- Invite people to opt in/out of mailing lists (in my case, one for free upgrade notifications, and another for new product / bundling announcements).
- The ability to bundle, offer discounts, and to call a remote server to GENERATE MY OWN REGISTRATION CODES so I can offer free trials and not be locked in to whatever (certain-to-be-hacked) "protection" Palm may have planned.
Really, the ideal would be to make the Palm app store optional. I think because you're on the receiving end of so many support calls, it gives your executives a distorted view of how stable/unstable the majority of existing Palm devices tend to be. It really depends on the kinds of applications one installs, and I don't think the rest of us should be penalized by forcing us to use an app store of your making.
To speak bluntly, I don't want to subsidize Palm to the tune of double-digit percentages of my sales. Your existing partnership with PalmGear is usurious; it's over 50% of revenue.
Code signing is fine; again, I hope you allow for the fact that many of us have Comodo or other code-signing certificates we use for our Windows installers or conduit DLLs. Having to purchase a separate one for use in your store would be redundant and probably more expensive than it needs to be.
In general, keep the platform truly open and don't try to fight us as we help make the Pre a bestseller. We all want to see this succeed.
Since it's already running on Android and Windows Mobile anyways, by not allowing it to run you guys will seriously miss out on the best virtual machine EVER. This would fix one of the biggest criticisms of Web OS which is that Mojo doesn't do much for developing excellent games on mobile. With flash, all kinds of existing games could be ported and new ones will surely arise. Don't be fools like Apple, get flash to run on it, flash 10 is on its way to mobile (especially android and winmo) in 2009, don't miss out!
1) NDA - learn from Apple's mistakes: you want developers to talk to one another, to create their own resources, to leverage each other's experiences. If your NDA is like Apple's was, then I as a developer couldn't even tell anyone I was developing on the iPhone. Don't make the same mistake please.
2) Distribution - be flexible. Learn from Apple's mistakes: if you try and lock things down all you will do is alienate a large portion of your developers, who will go develop for some other platform. If you must have an app store, go ahead, but it should be one of many mechanisms for getting apps onto the device, not forced upon the user.
3) Concurrency model - learn from Apple's mistakes: if you want to limit background processing then there are many ways to do so without adopting a 'only one app running at a time, unless it's an Apple app' policy, which just makes things frustrating for the user and hamstrings the developers. We have many resource allocation/management mechanisms from back in the good old days of cycles being a precious resource - use one or more of them.
4) Iterating the public beta SDK - learn from Apple's mistakes: trying to guess what the hell would be broken every morning when you got into the office because the latest SDK drop arrived unannounced with an unannounced set of often major and poorly documented changes is just a stupid waste of time. Publish ahead of time when the next version will be released, and what API changes and bug fixes it will have in it, at least a week ahead of time, preferably two - development teams for projects any larger than trivial size need time to plan around changes to the SDK and not waste precious development resource, e.g. recoding stuff that we never would have done had we known that feature was being removed in the next SDK iteration.
5) Community - do it right. Employ people that have created successful developer platform communities in the past, learn from their experiences. Learn from Apple's mistakes: they had a great Mac developer community with driven passionate developers. Somehow they managed to avoid utilising any of their successful experiences of the Mac developer community when starting the iPhone programme, much to the developers' confusion and annoyance.
Off to go write a book now. Good luck, I'd love to see Pre succeed.
2) Rich API. Full access to the platform's data and events: timers, PIM data (including PIM item modification events), calls, messages, location - shouldn't be limited compared to, say, Windows Mobile, Android or PalmOS.
3) Services: ability to create modules that run in the background, maybe without UI, survive restarts and low memory conditions (e.g., restarted by the system after being dumped out of memory), and are accessible from other applications/services via some protocols (like DLL, COM, CORBA, RMI, etc.)
4) Code and data persistence - no imposed dependence from the availability of a broadband (or any) connection. In other words, an application that doesn't require connectivity should be able to be started and work offline.
5) International availability of the devices.
1) Ability to install programs how I want to, without having to go through the App Store. Useful for Beta Testing or installing a program I found while browsing the internet and don't feel like hunting down in the app store.
2) Provide a sensible way of ordering applications and a solid search feature. I don't really care if I have to tap through to an advanced screen to get it, I just want it to be there. I want to be able to search by company, title, tags, price (user-defineable ranges please!), ratings, etc. And order them by similar.
3) Provide the ability for developers to sell bundles of content for an application, and for them to offer upgrade bundles as well, ideally with automated detection of what the user has installed already, and only offer them valid upgrade options. As a hard-and-fast example, take PlecoDict (http://www.pleco.com).
They offer Chinese/English dictionaries (and related) software. They currently have 8 dictionaries (I think) available for sale. The total price is over US$200 for all of them and not everyone wants to pay that much (and not everyone needs or wants them all), so they currently offer bundles. Buy 2 basic dictionaries for cheap, then you can upgrade and buy the more advanced/expensive ones later on. Or buy the more advanced ones first then buy the others later if you miss them. In each case, the price varies for the package and for the upgrades. More importantly, if you buy bundle B1 (including dictionaries D1, D2), you then pay a different price to get dictionaries D3, D4 than you would have if you had purchased bundle B2, including Dictionaries D4, D5, so it can get a little complex, but is also necessary.
4) Provide an easy way to do trials of applications but don't lock developers in. Again, taking Pleco as an example, some of their contracts may not allow them to make the full dictionary available for free, even if it's as part of a time-limited trial, so they instead reduce the functionality of the program (quite significantly) but let you use it for an unlimited amount of time.
5) Allow developers to re-use the App Store for a different purpose. Take eReader as an example - currently they have a huge number of books on sale, and I would love to be able to search through them all on the device and purchase them then and there. This would require the app store to be customizable and either pointed to a different database or to a different server or whatever. You would obviously need to allow developers to add extra search fields, like ISBN, Author, Publisher, Year, etc. It would make it quicker for developers to finish their program, and probably result in it being more full-featured.
6) Provide a 'Company' page accessible through the App store that lists any and all applications they've made available on it. Also allow for Open Source developers to have their own page, ideally with an optional 'Donate' button on it (some people would prefer money went to actual charities rather than themselves).
7) Provide a seperate category for OSS software, but mix it in amongst the regular results as well (while allowing people to exclude free apps) so that people who might not know what is or why they should investigate that category can still find potential solutions.
That's all I've got for distribution ideas at the moment. In terms of what else I want, there's the following:
8) Unicode support. UTF-8/16 is essential. If you haven't already got it supported then it's going to be hell to add it later...
9) An unlocked GSM model of the phone available for international developers, ideally running quad-band 3G (850/900/1900/2100). Why Quad-band 3G? Because not everyone lives in a country with EDGE support, and I can't imagine using this phone with only GPRS support. It's one thing to develop an app using an emulator, but quite another to actually use it in the real world and I suspect most international developers will struggle to make usable programs until they've actually got a phone they can carry around with them and see how their app performs in day-to-day use. I'm down in New Zealand, and I don't expect my carrier to be carrying this phone for some time after the GSM launch. While it's Vodafone, and if Vodafone UK picks it up I expect <acronym>VFNZ</acronym> will, it won't be a simultaneous launch, and may be several months behind the VFUK launch. It will also be more expensive, and I don't particularly want a carrier-branded phone.
This is more important than it might be with other phones because it's marketed pretty heavily as an 'always connected' phone, and importing the Sprint CDMA version won't be an awful lot of use as there aren't many Wifi hotspots where I live, and none are free. I imagine that's true for other developers as well.
10) An SDK that runs on Linux. I understand that you're basing this off Eclipse. Eclipse runs just fine on Linux, and the phone itself is Linux at the core, so I'm really hoping that I'll be able to compile and develop applications on Linux. I don't want to have to use WINE, I want native Linux support. I can deal with it if I don't (Windows in a VM), but I'll be sad if I have to do it that way.
In terms of what others have been asking for, I'm not terribly worried about code encryption/obfuscation (and will be turning that off if it's present), nor do I care about low-level development access. I would like that both of these things are present (I would like to see Pleco be able to be ported, and that requires that both low-level development access and code encryption/obfuscation are present. The first purely for speed, the second because of contractual obligations. Same for eReader, really), but I don't expect to make use of them myself.
Anyway, thanks for reading. I'm looking forward to this phone, and I hope it's out soon.
- If possible, a phone simulator on SDK.
- Show a lot of examples for learning
Please consider allowing us users to have access to these small, 'home-brewed' apps that have always made the Palm platform unique, open, and versatile.
- Let developers upload apps without having to wait for Palm to check them like Apple does. It's very frustrating having to wait for Apple when you have a fix that you need to get release ASAP.
- Release the SDK as soon as possible.
- Don't be like Apple and take complete control of the store. Google's model is much better. Give developers freedom.
- Provide devices without having to sign for a Sprint contract/plan (like Google does)
Keep up the good work. From what I've seen the Pre looks like a real winner.
This bluetooth keyboard capability is also present in Nokia phones.
I have a few questions to ask right here which is the idea(thinking) of majority chinese users.
[b]1. [/b]Can new Webos support any Asian language,such as Simplified Chinese、traditional Chinese or Japanese?by which way to make it come true? original unicode?
[b]2. [/b] I konw u guys have already pulled out of china market,if the new webos support chinese, when u guys would come back to china as the #1 sellphone market of the world? form 01-01-09,china already to release 3G licenses,include CDMA2000 and WCDMA, I think the network is not a problem.
[b]3.[/b] The new Webos can't run any oldOS prc,maybe the only way is use simulator.Do u guys have any plan to make one which simulator can support chinese and let us run old app-old PIM and read personal information?
1) A central app store is quite welcome to ensure ease of distribution for low cost applications. Apple does it mostly right, the only improvement on their model would be to provide both a vetted area and a non-checked area, so that approval processes would not need to delay the release. This would also allow bug fixes to be released quickly.
As for the payment models that should be supported for applications, single app, bundling, upgrades and money-back are the only important ones that need support (trials are the same as money back).
2) There needs to be a way to distribute applications outside of the app store, especially for in house applications. For these, there should be a way to sign phones as company phones, and provide them with a link an in house app store. This would also enable applications to be locked down to in house phones, if so desired. (And on the other side, provide a company white list of approved apps for use on the phone)
A further use of company certified phones could be to support volume deals on applications, a company could use their in house software to buy licenses in bulk, and the normal app store could then use the company certificate to distribute these applications and provide the license management.
For testing applications, direct download would be the easiest way to support it.
3) For updates, there should be some kind of update service that checks the download location of each application for updates automatically. For this you might need to define the format of an update file to provide the update information, to be maintained by the application provider. This update check should be able to do automatic updates.
4) All applications should be signed, so that specific signatures could be blocked from the phone. There would be an opt-in palm blacklist, and you could block them yourself on either the phone or the in house store. There would be two classes of signatures, one free without any checking, and one where you have to pay a small fee, but where the contact details are verified. This is essentially security by auditing, you do not prevent people doing harm, but you know who they are to punish them later if they misbehave.
5) Since applications cannot be checked thoroughly in an cost efficient manner, there should be three access levels to the functions of the phone:
green: cannot do much harm, has web access, can add/change information to synergy (I imagine synergy to provide a join of multiple tables, and applications can add/maintain their own table to be included in the shared synergy view), have their private database, can exchange information only via cut and paste, and have access to rough location information only (at the county level).
yellow: more dangerous, gain access to detailed location information, can read all shared databases, but not modify these
red: access to everything there is available on the phone.
In addition, an application should be able to permanently gain read or read/modify access to a database by asking the security manager for permission. The permanent permission is key here, you should be bothered by the security only once during installation.
1. For us to test Pre Apps on the computer.
2. For us to run old apps on the Pre.
These are my only prerequisies to buying a Pre
www.windowsmobile.com/MobileDeviceManager
- security/password policies, file/device encryption
- remote management
- OTA deployment
- application dependency management
- allow/deny only certain applications etc.
It would be wise, Palm and developers alike, not to underestimate the WebOS platform for internal enterprise solutions. This is potentially a huge market being dominated by Microsoft at the moment, but as we all know, domination doesn't necessarily mean superiority. Using web based technologies is a smart move, since this can potentially enable even our group of web developers to easily extend their expertise to a mobile platform.
Sorry, struggling to stay on point here, but a question was raised earlier regarding database support. Please could this be answered, as it would obviously make a big impact on the viability of Pre in the Enterprise.
Back to deployment... An App Catalog is not a bad idea, but app distribution should definitely not be limited to just that model. That having been said, lack of regulation or control could potentially lead to a whole bunch of nasty apps breaking my Pre. If I want to download and install an app, I'd like to know if there's a degree of risk involved. This is where security certificates issued by Palm to approved apps can prove helpful. That can give developers the opportunity to develop a partnership with Palm and be rewarded for writing good apps. But, if you still want to write bad apps, or couldn't care less about that program, no one will stop you from writing apps... but your users will be notified that your app is not certified.
Just a thought to throw in the pool.
Now... where's that SDK!?
** If apps are html, can they be written in php? If the pre supported php which in turn had access to the filesystem, I would think there is not much an app could not do.
** Also some means of source obfuscation is required to securely distribute apps.
** Even if apps can be written in php, which reduces source code exposure to "view source" in the browser, is there a way to make sure users can't view the php source in some sort of file viewer on their Pre? I can't see a developer spending a lot of time and $ writing an app if it can only be distributed as plain text without any protections. For instance, a user could easily comment out the trial period or copy protection portions of the javascript, php or html.
2. We need to see what can be done with Mojo to decide whether we are going to support palm pre or not.
3. Don't emulate the legacy plam software on palm pre , this will affect new software sale, that eventually affect the altitude of programmer to support palm pre or not.
I'd rather see the SDK before I even worry about app distribution. Allowing the use of native code would make the SDK much more useful and could solve, for some people, the point raised in #2 (yes, I am one of those people). Provide a alternate SDK that would allow developer's to chose the best tool for the job or allow the development of alternative tools/languages. It should not be as straitjacketed as Apple or Android platforms (what remained nice about PalmOS).
Well OK:
1) allow installation of code directly on the phone: do not require it to be loaded off a server.
2) Make it clear at installation that software is not "certified" and the risks involved.
3) Make it possible for software developers to use the technology of your app store on or from their web site for both limited distribution (e.g. internal enterprise, specific customers, trials, etc.) and retail distribution.
4) make optional, the use of your app store.
2) I'd like to see a Twitterfeed for new apps. That would be helpful while at the desktop.
3) Have a web-accessible version of the store. This is a huge shortcoming of both Apple and Google. Again, this is great when at the desktop (or even on the road, using a netbook!).
1. Update the public site to explain how the multiple Calendar/Email services work. What I want to know is can I have an Exchange (Work: Email, Contacts, Email), POP Email, Imap Email, A desktop app that syncs my home Outlook calendar and contacts but keeps them all in their own playground. With the ability to select what i want to overla[p and to add items and pick from a drop down where to save them.
2. The ability to download and install personaly written apps. The ability to self sign or have no signature and let them run once I have click on the am I sure box. (Would be great for internal apps for companies who support this device)
The SDK should be a fully functional simulator including the ability to dial phoen # and have the phone go thru the motions in making a call (Even though it is just going thru the motions)
1) Ability to at least open and view Word, Excel, Open office versions also.
2) Ability to view PDF (As an extra feture would be advanced support for PDF forms processing)
www dot parrot dot org
The perl comminuty is eager to do stuff on new platforms(i include myself)
A "official" palm post on perlporters could do the trick. Besides, is right on time, being that Parrot (the VM that runs Java, .NET, Python, PHP. Perl, Perl6, Basic, and many more) is about to be released to 1.0 in march.
Giving the WebOS that kind of power will open the gates to a HUGE number of apps, still keeping the advantage of reusing the skills that are more popular, and adding more (A lot of people knows BASIC, a lot more PHP, etc etc)
I would also like to re-re-re-stress that Palm needs to start Day Zero full on developer relations; I cannot begin to go into the horrific nightmare some other platform is causing, on MANY levels, and the seeming outright contempt that is being shown to developers...a lot of which the general public has no clue about (yet).
We need more information, and preferably, before March. We need tools, guys.
segment app store: people self-organize into tribes so cater to more than one 'Palm tribe': 'experimental' apps, vetted apps, 'private' (enterprise/organization) apps, etc.
As an user in China, I care more about internationalization of WebOS. A very important reason that Palm is not as popular in East Asia (China, Japan, Korea) as in North America is bad support for Double Byte Character Set (DBCS) support. Even latest unreleased OS6 never mentioned build-in DBCS-support. We have to process & display based on 3rd party software. Althrough iPhone is not officially release in China, its build-in local language interface & display support lead to its popular in China (In my office, 5 out 30 staffs use iPhone, that is more than 15% market share. It's amazing considering about 600 million+ cell phone user, twice number of US population, in China!).
1. Build-in support for unicode is bottomline for a success mobile OS in East Asia. Extended support for other encoding system (such as GB2312, GBK, BIG5,JS-???) will be perfect!
2. Good develop enviroment for input method programing is also very important. Besides high-efficient PIM & rich software resource, an excellent input method with perfect keybord support (like MacroHard & HandEase) is a key reason for choosing treo according feedback from treo forum. But based on existing information, programer can't develop an input method with only HTML, CSS & Javascript.
3. Local installtion for non-Sprint (or rumored Vodafone) user.
Also, it seems as though people abominate cloud computing (myself included). Dummy terminals in internet cafés and in huge organizations/corporations or banks are really the only places for such technologies. Just think of cheap people like me who pay for per kilobye cent usage of the internet on their phones (which makes downloading a megabyte's worth of data cost 10 dollars); I have a pretty good idea where this HTML/CSS/Javascript thing is going.
I also think that the built-in trial functionality would be a very good idea, as Anderson (poster #3) describes.
Previous Palm devices used to support device locking (JackFlash, JackSprat, FlashPro, etc.), but the new architecture has closed out the user who wants to modify their device (say, to start up with a particular program, or have a program always running, even if the device is reset to (now (assuming these things are possible)) factory settings) to fill a particular need (but really, this could be for ANY purpose). What I am getting at is that (referring to the Centro/Treo x) programs like Warden Security, TealLock, or SecureX cannot be put built into the operating system's image on a device except by a Palm insider! It would make the devices much less desirable by thieves (and more desirable to customers) to make device encryption >and< locking standard on all Palm Prē devices, not to mention give us peace of mind (this means using a good encryption algorithm, like Twofish, and not a decrepit one, like DES). I hope you guys take this into consideration for FUTURE devices, because it still pisses me off (although I admire Palm) every time I look at my handheld.
Don't promise anything too early, either; take as long as is needed to implement these features (webOS, the developer SDK, etc.). If it takes until later this year, or even until next year, it's okay. It's alright to hurry sometimes, but not to rush.
I also think that the built-in trial functionality would be a very good idea, as Anderson (poster #3) describes.
2. A contest like "Android Developer Challenge" will be great to attract developers in huge number and starts with a bang.
3. Most of the developers can't market the products themselves, so appstore going to be great help. Restricted installation ( allowing installation only thru store kinda ) will eliminate piracy completely.
Let's see a bit more bite to match up to the bark!
We're ready to start building apps. What's the hold up? A reply would be nice.
Also, you really need to think about how searching will work for languages like Chinese. "But it's easy," I hear you say, "just match character for character!". Well, no. Not quite.
Take Chinese for example. If someone has a rare character in their name, it could take 30 seconds or a minute or more to find it in the IME. That's quite a long time if all you want to do is call them. What is far preferable in that case is to be able to enter the pinyin for their name - instead of (or in addition to) the characters - and have it come up with everyone whose name matches those characters. It's likely this will produce a number of results, but it's also possible to know multiple people with the same full name in English, though less common.
Same for Japanese and possibly Korean as well (and any other languages I've not mentioned). And let it be customizable! Provide your own default implementation, sure (in fact, please do!), but let others install their own. This would, for example, make it possible for people to map Characters to Cantonese pronunciation rather than Mandarin if they're in Hong Kong or Guangzhou or wherever.
This should also be applied to all core apps, and to 3rd part apps where applicable and possible. It would not be cool to have this functionality in the contacts listing but not in the calendar.
There's been a few people on here requesting Chinese support, and there's likely to be a lot more who'll buy the phone if it was available in China (for a reasonable price).
Though I'm likely to be buying the unlocked GSM version, assuming there's no major problems in the CDMA version and the SDK is sufficiently open and flexible. Oh, and assuming you're going to be selling an unlocked GSM version in a timely manner.
Finally, where's the next blog post? It's nice to have this one, but making a single big post and staying silent isn't what a lot of us were hoping for...
Release whatever you have for the SDK TODAY, not tomorrow, not next week. Even if it's handwritten notes from the Palm development team... we need to know so we can make plans to support this platform or not. Really. And if that is not possible... maybe you don't have the handwritten notes yet... at least narrow the release date down... 30 days? 60 days? Come on guys... you KNOW how this works!
- Applications, once created, should be stored in a single zip file. The Palm should be able to recognize these zip files as programs (through an XML file stored inside) and automatically give the user options to install, upgrade, or remove.
- While having an official app store is important, the Pre should offer the same install options no matter how it receives a program: through a 3rd-party site, e-mail, or otherwise.
- Applications should be stored on the Pre in their own subdirectories, so that deleting ALL of an app is as simple as deleting the folder (or running an uninstall wizard that does the same thing in the background).
I’ve only made one program in Adobe AIR, but I love the way AIR recognizes an application, installs it, and offers to put links on the Desktop / Start Menu for you. Palm could do the same with their cards system.
1- Possibility to install applications directly from the user's PC (or at least the possibility to have our own store). Our applications are not for the wide public but rather for large corporations. They should not appear on a public store.
2- Possibility for an application to communicate directly with the user PC (thru the USB cable, not over the air) or if not possible, the possibility for the application to write to the mass storage drive that is visible to the PC.
3- Code hiding is imperative. We can't have our source code visible. It must be protected somehow. Commercial applications need to be protected. If me code is visible, you can count me out.
4- A good simulator/emulator for debugging.
5- An SDK available ASAP (even a beta, to see what’s coming) or at least an API function list or help file.
6- An official developer forum (web based and/or NNTP based)
Thanks
7- A unique device id (serial number)
All this multitask multi app thing is great, but we NEED a way to close ALL or MULTIPLE apps at once.
After some normal activity I guess many apps will be open. For order freaks like me, that can be a problem. And there will always be the suscipion that active process will use juice and the need for the users to be reassured and that they can close the apps.
Now if 15 apps are open that's really not pratical
Give us some gesture or shortcut to :
- select multiple cards that we can throw away at once
- kill all apps at once
Thanks!!
I can't imagine myslef constantly throwing cards away which I'll do..
IYCC has no problem in principle with proprietary software (though we believe that proprietary code will soon go the way of the fax machine). However, as a developer, I don't want to depend by mistake on applications that are proprietary. Further, the more I use already written open source code, the smaller footprint my applications will have and the fewer bugs I'll have to fix. And of course, as a user myself, I want the freedom to use Free and Open Source Software to the greatest extent possible.
We have some "gotta haves", though, in order to be in a position to support the WebOS platform.
First, in order to develop the quality of seamless user experience that we deliver, we need the OS, including kernel, libraries, drivers, languages, and toolkits to remain open source to the maximum extent possible. There's really no better method of ensuring my applications integrate seamlessly than by looking at the source code of the software with which my applications interact.
Second, interoperability with Debian packages, using aptitude (or a GUI frontend to it) system to install .deb packages would *significantly* reduce our costs and speed development time. Debian's package management system is by far the best in the industry. Using the well-tested APT system will significantly reduce our costs, and provide a huge codebase of already-written apps and libraries to draw from. The RPM system is also very good, and its codebase is broad and stable, but nothing beats Debian's package management.
Third, applications that do not meet the Open Source Definition should be in a separate repository from those that do, and the user should be able to turn off the proprietary repository if the user wants so to do.
IYCC looks forward to developing for the Palm WebOS, and I personally am looking forward to being one of the first to buy a PalmPre!
Happy Trails,
Loye Young
Isaac & Young Computer Company
Laredo, Texas
http://start.iycc.net
http://www.iycc.net
Happy Trails,
Loye Young
1. The suggestion that apps be loadable outside a "store" is essential.
2. The suggestion to allow for apps to go through a "certification" process is a good one, and warning users that non-certified apps could cause damage is okay. Asking developers to pay something for the work involved to gain that certification is okay too.
3. allow for use of a memory expansion slot in devices.
One of the single best apps that my wife and I still morn the death of ... is DUAL DATE. The ability to quickly share our calendars over lunch with one beam was just so simple and .. well GREAT. Please bring back the ability for calendar sharing.
rick
This is not the place nor the platform for OSS Evangelism.
I believe there are more than a couple of Linux-based mobile platforms that in some way or another meet the "gotta haves" you posited; and from the looks of the consumer adoption, market penetration, general buzz and all measurable standards...the fact that they are OSS-y doesn't seem to be doing them any good.
As a (former) long-time advocate of OSS in general and Linux, specifically, at the end of the day, I can say that for commercial ventures aimed at end users in an environment where *quality and User Experience* trumps *quasi-religious* ideology, Palm would find itself in the same position as these other platforms if they came out waving a GNU-Style OSS banner.
Besides: only the Linux that it runs on is likely OSS; it is being used as an embedded os/HAL. The real action happens in the API's running on top of them, and I am willing to bet that Palm has no intention of free-for-all source access to that stuff...
There are many great things on the internet, but there are many others that I don't want coming through my phone. Can you please add the ability to add even a simple level of parental controls to a webOS-enabled phone?
This is a huge deal in my family. RIM supports a similar kind of control in Blackberry devices, but I believe that Palm is better for many reasons.
We have not subscribed to internet access in our Palm phones because there is no filter. It would be wonderful to be able to purchase a Palm Pre with restrictions to allow internet access on only certain applications (such as Google Maps or yellow page searches) and to restrict the addresses available in the built-in Webkit browser to safe sites we trust.
I use Opera on my PC, and it lets you restrict access using a urlfilter.ini file. I am not sure if Webkit is designed to support a similar file, but all we need is to be able to configure a URL filter file and password protect it.
I believe that this functionality would open the door to a lot of conscientious parents who are concerned about allowing unrestricted internet access on their families' cell phones.
Having said that:
we need a way to make our private business apps private, to get them ONLY with a special code, or to specially authorized PRE's.
Recurring billing would be nice.
auto-version update would be nice.
as 11000 other people have said, html/css/javascript is all well and good, but I'm not putting proprietary business methods on a phone as source code. Javascript obfuscation if needed has to be industrial strength.
0) Less is more. The less Palm itself is crucial to distribution, the better. The fewer issues have to be solved with or in distribution methods, the better.
For instance, security. If applications are sandboxed well enough that they can't harm or spy on anything, that means fewer worries about installing something. Fewer road blocks and steps in both users' and developers' faces.
Apps shouldn't be able to even read other apps' data unless the user has knowingly (i.e. with a clear UI) set up a specific channel. Example: a GPS app should not be able to read my address book (even though that's obviously useful) unless I said so. The user should be able to allow all sorts of interconnections between apps, and between apps and things on the internet, usb drive, computer thru usb, computer thru bluetooth, etc.
If you can securely verify that an update comes from the author of the original program, then updates should be no hassle even though the user has given the software permission to touch things outside its sandbox.
1) Make a MIME type for "Palm Pre app". Clicking a Pre app link on any site means you are going through the install process. User should be able to "download only," and maybe install later, but this should be a menu item or option you turn on rather than in the face of average users.
2) Figure out how to allow Amazon to sell Pre apps.
3) Think about content (as opposed to application) installation. Seems like a lot of good content should be viewable with the built-in webbish software (including PDFs).
I'm with others here: the more standard formats you can handle, the less distribution hassle. I hate Flash, but as long as I can turn it off, or get it to quit in less than a second, I far prefer that to some indirect transcoding scheme. Figure out how to prevent plugins from stealing the CPU. Like I said: if you can handle the problem outside of distribution, less distribution problems. To put it another way: where web browsing works, distribution is solved.
Some content requires players or plugins. On the one hand, allowing the user to buy or download some content bundled with the right player is nice, but effectively forcing content developers to compile the content into a copy of the player is sad. More like a prerequisites thing or bundle in the sales sense. Prerequisites are good for things that depend on plugins, too. Also helpful for uninstalling ("You are uninstalling the only app that uses the RealPlayer plugin. Do you want to uninstall RealPlayer, too?" (yes))
Also, Apple got into the situation of censoring content based on their app-vetting process. Don't try to censor content on your own store. An "I'm 18" category is fine.
Btw, another kind of bundle is collections of files that go with something, like the directory full of css and jpg files that goes with a web page, or the Apple "packages" which are also just directories of files. Sounds like something you would have down already, just saying.
4) The phone should belong to the user, not the developers.
In the following I am not saying users should need to muck about with settings files, but that the user should have the ultimate power.
That includes letting the user say no to updates, do manual downgrades, look at and change config files, prevent the app from contacting the developer's (or any) site. If the developer doesn't want the app to work in a true sandbox, the app can put up a dialog saying what it wants the user to enable. But the dialog where the user actually makes the choice must be controlled by the OS.
You might want a way for the user to give permission for the app to keep some data "hidden" from the user, but the user should be able to locate and delete any "hidden" files.
Hiding data from the owner of the phone is essentially DRM. It's impossible without special tamper-proof hardware, nearly impossible with. Don't get into the morass of trying to provide an impossible service to worried developers. That is, you can make a stab at hiding, but don't try to guarantee to developers that this is secure, and do not get involved in DMCA fights with hackers or pirates over DRM, third-party apps, or content.
Same with code obfuscation. It's fine to make a gesture at it although leaving it up to the developers themselves is better. There is no such thing as industrial strength obfuscation or DRM unless you mean the flailing strength of dinosaur industries. Don't waste effort on the DRM at the expense of regular developers who just want to deliver useful stuff.
The user needs the ability to delete all of an app's user data and config info and force it to reconfigure from scratch. (I don't mean muck with internal code files and necessarily expect the app to keep working!)
5) Obviously sync apps and content as well as data. Allow installing stuff that was downloaded (or produced) on the PC as part of the sync process.
6) Install from (and sync with) USB storage. (Might want to have content, apps, or apps' data selectively reside on a USB drive as main location (like with SD on a Treo). However, that requires syncing the USB drive with the PC. I know, this is drifting away from distribution.)
Please release the SDK before the phone!
2) An method to put in variable discounts depending on previous software purchased. If someone purchased my three most expensive applications, maybe I would want them to get a big discount on another one, or give away the cheapest one for free for being a great customer.
3) At the very least, a way to put variations of a product on one page instead of taking up multiple pages. Say I had a program with features ABCD. I know that most people will only want three of those features, but which one depends on the customer. It would be nice to have the ability to let them check mark which ones they want, and show the adjusted price, or at least see all the options in one page for an easy comparison. Same goes for a much simpler regular, deluxe, and premium editions of one product.
3.Good text and voice functionality.
4.Repetitive alert capability
5.Applications:
a. Epocrates
Patientkeeper (future consideration)
6.Battery life 12h+ at usage rates of text/voice
7.Handsfree capability? Voice dialing? (Colorado House Bill 1094)
8.Cloud backup & sync options? Encryption options for the cloud? (HIPAA compliance)
From what I've read, #3, #4, and #5a are covered.
Thanks,
Kyle Thompson, M.D.
South Denver Anesthesiology, P.C.
USR Pilot 5000->3Com PalmPilot Pro->Palm IIIx->Palm 515->Palm TT3->Treo 700p->HTC Touch Pro
Would make the community much more vibrant and allow for more "serious" applications to pop. I can't think of you winning with just the webOS. Doing it later rather than sooner might be too late.
Palm's WebOS is an open source operating system based on the Linux kernel. Using an open source operating system such as WebOS does not *require* that you provide the source code for your proprietary application. If users want to pay you for a cool application and they don't care about the source code, the users should be free to make that choice.
Proprietary apps need a tollbooth for downloads, while open source applications don't. For this reason, proprietary and open source applications need to be kept in separate repositories, but that separation is more important to developers than to users.
A well designed user interface can make the existence of various repository seamless. Various applications have already been written to allow users to add, modify, and remove repositories. Such applications are typically found in a "system settings" menu that most users don't bother to mess with because the defaults are usually sane.
To Kai (post 165):
If you read my earlier post carefully, you will see that I am far from a religious zealot when it comes to open source. My issues are practical and economic.
You will note that the title of this thread is "Application Distribution". My comments are addressed to the nuts-and-bolts of distributing software (of various licenses) on top of an open source operating system.
IYCC has a fair amount of expertise on these issues because we already distribute both open source and proprietary software as a part of the IYCC Linux distribution. We know well the issues surrounding a mixed proprietary and open source software stack, and we make our living serving OEMs and ODMs who want their customers to have a seamless experience.
IYCC sees Palm's WebOS as an opportunity to make mobile computing available not just to the techno-enthusiasts (ala iPhone) but to mainstream users who want their gadgets to "Just Work".
Happy Trails,
Loye Young
Isaac & Young Computer Company
Laredo, Texas
http://start.iycc.net
http://www.iycc.net
I am wondering if there is a pre-launch application development program that interested developers can participate in. We recently lauched a local search application(Superpages) for iPhone and would love to get our hands on Pre's sdk as soon as possible. Let me know.
Thanks,
Uddhav
Our team is now aware of your interest in the SDK. We are currently in private prerelease, and if you haven't done so already, please be sure to subscribe to the RSS feed on the blog to keep up to speed on any news we may have about the SDK
1) SD card
2) ability to run Palm apps
Touchscreen with ability to switch between stylus or finger, at convenience of the user.
Palm OS (Garnet) full interoperability (and it was one of your "election days" promises, Palm, don't forget it)
Ability to admit native code aplications, instead of only surface-level .html and JS.
SDIO slot, with SDHC support (as in 32GB, i.e.)
GPS and WiFi (for the case you were thinking of ditching them)
The best online market resource, IMO, is a lean client app, or a simple browser bookmark. So feel free to use my opinion to endorse the death of the AddIt client and the death of the buggy MyPalm client. If much of this is now about getting online, all we need is a bookmark.
Apps installation: as of my experience, best practice is a clean download that can then be executed, but clean download goes first. That may do for e-commerce and upgrades.
Software security: OK, better leave it for Symantec or Trend Micro. Things are getting this far, anyway.
Come on, Palm people, you did it before. DO YOUR MAGIC!!!
There are many college students that are tired of hauling laptops around. Many would opt for a Palm Pilot if there was a keyboard that actually worked. Even though I have had bad luck with them, I would quickly buy another if it was truly functional.
Is there anyone listening to me?!!!