ACCESS and PalmSource Announce the ACCESS Linux Platform

[Ed Hansberry]

http://www.palmsource.com/press/2006/021406_accesslinuxplatform.html

"ACCESS Co., Ltd., and its wholly owned subsidiary, PalmSource, Inc., today announced the ACCESS Linux Platform (ALP), the latest evolution of Palm OS® for Linux. The ACCESS Linux Platform is designed to be an integrated, open and flexible Linux-based platform tailored for smartphones and mobile devices."

Last summer, ACCESS acquired PalmSource, the maker of PalmOS, which was spun off from Palm, Inc. several years earlier. In late 2004, PalmSource itself purchased China MobileSoft. The replacement to the current PalmOS 5 version, PalmOS 6, or Cobalt, was faltering even then having no device makers interested in bringing a device to market. With the acquisition of China MobileSoft, it was assumed a Linux version of the PalmOS would follow. Sure enough, a few months later, PalmSource halted all non-Linux development, essentially declaring PalmOS, as we know it, was dead. In September, ACCESS announced it was buying PalmSource and in October, said that PalmOS was nearing the end of its life - which is what I said back in June. ACCESS had no interest in PalmOS itself, just the look and feel, their customer base and the tools necessary to make today's PalmOS apps work on a full blown Linux version at some point in the future.

The announcement today confirms all of this and gives us a peek at what is next. App compatibility will probably mean all "properly written" 68K apps can run unchanged. That means emulator. No one cares about Protein (Cobalt APIs) anymore so there isn't anything to support. As for the apps that take Garnet (PalmOS5) to the limit, it is unknown. The press release doesn't speak to that directly. Most OS5 apps are really 68K apps anyway that occasionally make calls to the APIs outside of the current Palm Application Compatibility Environment, or PACE.

The SDK is supposed to be available by the end of this year. This is probably a beta SDK. For comparison, Microsoft typically releases a beta SDK 3-9 months before a Windows Mobile OS ships to device manufacturers, and then there is another 3-6 months before devices start flowing freely. Given this is a totally new OS, ACCESS may be trying to get the SDK out even earlier to get developers involved in some degree of testing as well as having ample time to get their apps ready for a launch. It is definitely safe to say it will be the end of 2007 before devices come out and more likely, mid-2008.

[Ed Hansberry]

http://www.palmsource.com/press/2006/021406_accesslinuxplatform.html

"ACCESS Co., Ltd., and its wholly owned subsidiary, PalmSource, Inc., today announced the ACCESS Linux Platform (ALP), the latest evolution of Palm OS® for Linux. The ACCESS Linux Platform is designed to be an integrated, open and flexible Linux-based platform tailored for smartphones and mobile devices."

Last summer, ACCESS acquired PalmSource, the maker of PalmOS, which was spun off from Palm, Inc. several years earlier. In late 2004, PalmSource itself purchased China MobileSoft. The replacement to the current PalmOS 5 version, PalmOS 6, or Cobalt, was faltering even then having no device makers interested in bringing a device to market. With the acquisition of China MobileSoft, it was assumed a Linux version of the PalmOS would follow. Sure enough, a few months later, PalmSource halted all non-Linux development, essentially declaring PalmOS, as we know it, was dead. In September, ACCESS announced it was buying PalmSource and in October, said that PalmOS was nearing the end of its life - which is what I said back in June. ACCESS had no interest in PalmOS itself, just the look and feel, their customer base and the tools necessary to make today's PalmOS apps work on a full blown Linux version at some point in the future.

The announcement today confirms all of this and gives us a peek at what is next. App compatibility will probably mean all "properly written" 68K apps can run unchanged. That means emulator. No one cares about Protein (Cobalt APIs) anymore so there isn't anything to support. As for the apps that take Garnet (PalmOS5) to the limit, it is unknown. The press release doesn't speak to that directly. Most OS5 apps are really 68K apps anyway that occasionally make calls to the APIs outside of the current Palm Application Compatibility Environment, or PACE.

The SDK is supposed to be available by the end of this year. This is probably a beta SDK. For comparison, Microsoft typically releases a beta SDK 3-9 months before a Windows Mobile OS ships to device manufacturers, and then there is another 3-6 months before devices start flowing freely. Given this is a totally new OS, ACCESS may be trying to get the SDK out even earlier to get developers involved in some degree of testing as well as having ample time to get their apps ready for a launch. It is definitely safe to say it will be the end of 2007 before devices come out and more likely, mid-2008.

[Foo Fighter]

The SDK won't ship until the end of this year, which means we won't see ALPOS-based hardware until late 2007/early 2008. And that means we're stuck with Garnet for almost another two years...

[Tim Rapson]

By the time a PDA with this OS is out Dell will be selling X52vs with WinMob 2005 Second Edition.
I am most likely to get an X51v this Spring or Summer. Already, my most used program (FITALY) has given up on Palm OS as every time they introduce a new model TextAware has to hack a way to make FITALY work because Palm has broken the rules for their own (or formerly their own) OS.
Sadly, even when this new Plinux OS is ready it will be 6 months to a year before a product ships and then I would have to wait another 6 month to a year to see if the OS is going to work and another year for my basic software to be updated to use the new features. Sorry, I am just not going ot wait 2-3 years for something I could get with a X51v now.
I give up.

[Janak Parekh]

Originally Posted by Tim Rapson

I give up.

It is indeed a shame that PalmOS is falling apart so soon. I'm not surprised the old architecture doesn't scale; I am surprised that the post-Garnet effort was coordinated so poorly, and I think the lack of competition is probably a loss for end-users.

--janak