We recently took part in an short interview about the Palm OS platform as part of the "Developers Speak Out About Palm OS" run by Kent Pribbernow at PocketFactory.com.
It’s no secret the Palm platform, as we know it, is in dire straits. PalmOS Garnet has survived long past its shelf life, with no suitable replacement in sight for at least the next few years. And PalmSource having been acquired by Access, who intends to replace the OS (Garnet) with a Linux based framework, and Palm licensing bits and pieces (as well as the whole) of Garnet code, it seems the future of the platform is up for grabs. (source: pocketfactory.com)
PalmOS of the future will turn out to be; something developed by Palm, or something entirely different from Access. Who knows.
While just about any devoted PalmOS fan has a passionate opinion on the matter, one voice that gets lost in the chorus is that of the developer. How does this shaky drama impact third party developers and their product roadmaps? I interviewed a few well known application developers to gain their thoughts about the current situation and their views on the future.
Howard Tomlinson, CEO of mobile game developer Astraware gives his thoughts…
“Where is Palm OS®? Is it viable in the long term for development? Will Astraware shift towards Windows Mobile® and other platforms?
“From a pure development standpoint, I think it is unlikely that Palm OS® will go beyond Garnet. Having said that, Garnet still has plenty of life left - it does the job of providing a PDA/Smartphone OS pretty well, and I expect Palm will keep it going with various bits of update for a couple of years, since there’ll be a continuing market for devices using it. I don’t see Palm OS alone as a growing market though. As a developer we can expect to be able to sell products for an operating system for some time after the last device using it is produced, and I expect we will still be making Palm OS sales in 2010 - after all, we’re still selling some titles for OS4 users. I don’t think we could make a viable business by supporting only Palm OS though, we’ve known that for some time. Until now we’ve really just focussed on two platforms – Palm OS and Windows Mobile®, with the intention of supporting both of those well.
“I know it is often repeated, but the “Zen of Palm” is a useful concept, whether it is Palm OS, or any other platform that Palm Inc. uses. Simply put, making the things that the user needs to do most often as easy to do as possible. Palm’s threaded messaging for SMS is a great example of making things easy, and their user interface choices for Windows Mobile show the same principle using a different OS. Although Garnet itself is likely the end of the Palm OS line, I’m hoping - and expecting - that it won’t be the end of the Zen.
“Linux development makes things interesting for us. We’ve been restructuring our development to become less reliant on a particular platform for several years, and recent work to make our games transfer well to mobile Linux should mean we’re able to support the upcoming ACCESS Linux Platform quite quickly. PalmSource saw their number of licensees for Palm OS drop, but I expect that a number of existing and new hardware companies will pick up on the Linux platform.
“I think Astraware is in a pretty good position - we’re just large enough to be able to devote a couple of people towards supporting new technologies and adding them to our infrastructure, while not so large that we can’t make business changes to redirect effort quickly where it makes sense. Plus, with our catalog of games written towards a generic handheld platform, being able to bring popular titles to new platforms quickly is one of our strengths. With the more recent versions of Symbian becoming easier for us to develop for, and with the work to make our systems more flexible, adding Symbian to our list of standard supported platforms is possible and something I’m hoping we will accomplish over the next year.
“It is always hard to predict the future, but since I have to try to look at the possible futures several years ahead, I do try to aim our business so that we can take advantage of platforms becoming strong, whilst not being too reliant on any one platform that reaches its end. Our shift to add Windows Mobile was a smart one a few years back - from where we are now, it seems like supporting mobile Linux is a smart thing for us to do right now. I am certainly looking forward to seeing what ACCESS bring out!”
Adriano Chiaretta, COO of iambic, Inc, developer of productivity software for PalmOS, had this to say…
- Is Iambic concerned about the future of the platform?
We are closely monitoring the developments around Palm OS as well as the upcoming Access Linux Platform, and yes, we are a bit concerned. At the same time we are confident that both Access and the acquired PalmSource are well aware of developers’ concerns and they are doing their best to evolve the platform in everyone’s best interest.
- Does your company see Palm OS as a “dead” or dying platform?
We don’t see Palm OS as a “dead” or dying platform. It continues to be adopted on new Licensees’ handhelds and smartphones (see the Treo 680 for example). Nonetheless, it clearly suffers from competition coming from more flexible mobile solutions-like Windows Mobile.
- How confident are you in the Access Linux Platform and the direction that company is taking the platform?
We are confident that Access is working closely with prospect Licensees in order to keep backward application compatibility and at the same time allow for a more flexible and modern operating system. Until we’ll see Licensees actually adopting the ALP, it is a bit early to really draw any conclusions.
- Will you port your products onto ALP?
At the moment we haven’t taken a definitive decision on this. Also, because thanks to the fact that ALP will have a compatibility layer allowing to run classic “68k” applications, the need for porting our titles to native ALP might not be too strong, at least in the initial phases of ALP deployment.
- How is the uncertain future of the platform affecting your product roadmap?
The declining diffusion of the Palm OS platform is introducing a serious need to cover as many mobile OS platforms as possible, and this certainly affects our roadmap.
- Do you plan to place less emphasis on Palm OS development going forward?
If at all possible, we’ll keep the same emphasis on Palm OS as we have so far. Despite the platform being not as strong as years ago, it still has a more than significant share of the mobile market, or at least of consumers looking for software for their devices.
- Will you move to Windows Mobile, etc.
Windows Mobile is definitely in our scope of expansion. Over the last few months, we started releasing titles for the Windows Mobile Pocket PC and Smartphone platforms (TravelTalk, EzQuote, LoanAnalyzer to mention a few). In the upcoming weeks the brand new Agendus for Windows Mobile will be available in public beta as well.
C.E. Stuart Dewar from Pimlico Software, developer of the acclaimed DateBk6 for PalmOS, gives a no holds barred viewpoint…
Right now Palm OS development has fractured badly - on the one hand, PalmSource wasted a huge number of man years developing an operating system that no one has adopted (OS/6 Cobalt), while on the other hand, PalmOne (I still use this name to emphasize its separate identify from PalmSource) started in on their own software development with new PIM apps and databases and NVFS memory support. Unfortunately, these projects were ill-conceived from the outset as they resulted in a significant loss of stability - one of the hallmarks of Palm OS handhelds. Many loyal palm users gave up when they updated to devices that were not nearly as stable as the ones they had been using.
Worse still, PalmOne does not appear to view the Palm OS as a platform to run third party applications. Palm Tech support regularly blames third party applications for known bugs in their OS - something that leaves developers extremely frustrated. PalmSource had an excellent reputation for working with developers on an open basis - even releasing the source code for the top level routines in the Palm OS. PalmOne views their software development as proprietary and it’s been a struggle to get them to release internal documentation on their products. A lot of the work that I did on DateBk involved painful reverse engineering of what they were doing in order to provide on-going support for my product.
The basic concept behind what Access is supposedly looking at is very attractive - PACE was a good idea that was excecuted almost flawlessly - an emulator that was rock solid and provided a completely seamless transition for Palm Handhelds as they moved to a totally different CPU architecture. A version of PACE running on top of Linux would provide an extremely attractive architecture for a handheld, especially if developers could tap PACE for Palm OS compatibility while also having direct access to Linux.
Such a device would truly be jump-started by the wealth of Palm OS software that’s been written.
But timing is important - OS/6 Cobalt was “delivered” at the end of 2004. We are now at the end of 2006 with nothing firm in sight. It’s not clear what Palmone is going to do in the future as they will need a more sophisticated operating system to move forward and compete with Windows Mobile Devices.
It’s doubtful they have the expertise or resources at this time to develop a new operating system and if Palm abandons OS/Garnet and just decides to use Windows Mobile, it’ll probably kill the Palm platform.
Maybe we will see some new handhelds appear in 2007 with the PACE/Linux architecture, or maybe Access will just decide to pull the plug on the entire project. They might decide to just do a straight Linux handheld - after all many Palm developers have or are working on converting their applications to run on Windows handhelds and they could equally convert them to run on a Linux device as well without any compatibility layer. I read a lot about ALP (Access Linux Platform), but then again I spent two years reading about OS/6 Cobalt as well. I need to see a physical device in hand with this platform.
Finally, we can also thank Microsoft’s FUD for part of the problem. I really got annoyed reading report after report about how Microsoft was quickly catching up with Palm and passing them in sales, with all the early reports conveniently ignoring the huge rising percentage of SmartPhones being sold by Palm. When total sales of PDA’s and SmartPhones were factored in, an entirely different picture emerged, but Microsoft managed to convince many industry pundits that the figures reporting on non-connected PDA’s were relevant and that fueled self-fulfilling prophecies. If everyone decides that the palm OS is likely to go away and plans accordingly, then it will, regardless of what Access, PalmSource or PalmOne do. I think that “Palm”
(collectively) needs to not only establish a road plan that gives both developers and end-users the confidence that there is something to look forward to in the near future, but also deliver real devices quickly.
|
Sphere related content 
iambic (116) - 
The support staff backend is much streamlined and will allow us to cut down response times considerably. Our goal as always is to take care of all the inquiries within 24 hours, and we are confident the new system will allow us to do so most if not all weekdays.