MobileInfocenter

PalmSource Sale Won't Affect Licensees

A representative for ACCESS recently told News.com that the company's acquisition of PalmSource would not have an immediate impact on the Palm OS or any of the licensees. The company also says work will continue on Palm Linux and that Cobalt may still see a release.

The statement in the article reads:

"The existing deployments of Palm OS, including Palm OS Garnet, will continuously be supported," an Access representative said. "The purpose of the agreement is to maximize the synergy between the two companies. We really don't expect any changes in the relationships that we have with licensees."

The rep also goes on to say that ACCESS wants to finish creating the Linux Palm operating system, and may revisit Palm OS Cobalt.

Article Comments

 (121 comments)

The following comments are owned by whoever posted them. PalmInfocenter is not responsible for them in any way.
Please Login or register here to add your comments.

Start a new Comment Down View Full Comment Thread

Access: You can trust us™.

The_Voice_of_Reason @ 9/12/2005 5:32:43 PM # Q
How many times has it happened where the management changes in your favorite restaurant and they promise everything will be the same? Only it just never seems to work out that way...

While NetFrontLinux has the potential to be a killer platform, after enduring YEARS of empty Cobalt promises, I'll believe it when I see it.


http://www.palminfocenter.com/comment_view.asp?ID=8060#111922


At this point, it's looking like with hacked-up, kludgy Garnet being a lame duck OS and PalmLinux/NetFrontLinux being at least 2 years away, Windows Mobile has just been given the market share outright and will consolidate their position in the coming months. "Just wait." seems to be the sad refrain from the PalmSource Apologists. Guess what? We're tired of waiting. This train ain't coming, Bubba.



TVoR

------------------------
Sony CLIE UX100: 128 MB real RAM, OLED screen. All the PDA anyone really ever wanted.
------------------------

The Palm eCONomy = Communism™

The Great Palm Swindle: http://www.palminfocenter.com/comment_view.asp?ID=7864#108038

NetFrontLinux - the next major cellphone OS?: http://www.palminfocenter.com/comment_view.asp?ID=8060#111823

RE: Access: You can trust us™.
cervezas @ 9/12/2005 5:54:29 PM # Q
How many times has it happened where the management changes in your favorite restaurant and they promise everything will be the same [but it doesn't] work out that way?

At least as often as the management changing and everything being better.

Still, I have my concerns. Not that Access can't be trusted or that they are stating intentions that are more wishful than known-deliverable: I expect they've done plenty of due diligence and more than likely know that they won't gain anything from not being straight about things like this. My concern is whether the business cultures, development cultures, and "culture" cultures of the two companies will be able to merge successfully and that the attempt to integrate them won't be pushed too fast. Difficulties in integrating disparate cultures is the kind of thing that kills many a merger that looked good on paper because it's an effect that's hard to quantify in advance.

As a people, the Japanese tend to have a stronger sense of "us" (Japanese) and "them" (foreigners) than Americans do. In business this often means that if you're not Japanese you're not going to be around in a leadership role for any longer than absolutely necessary. Not that you don't expect leadership changes in the target company after an acquisition, or even that I have some strong opinion that PalmSource has the best people at the helm now (although I definitely think they're better than what we've seen in recent years). But I'm concerned that Access could try to change too many horses mid-race and create more confusion and distraction than the important development effort underway can sustain.

I'd be interested to know more about the US division of Access that I believe PSI is actually merging with and know some things like:

- How culturally open are they?
- How similar is the composition of the company as far as the proportion of management vs. engineers?
- How many of the Access engineers here in the US are the ones who have actually worked closely with Palm and PalmSource in the past?

One good thing: the fact that mobile systems design is a new business for Access and that they are continuing (or perhaps even ramping up) their own browser development makes it less likely that there will be a lot of redundancy at the engineering level that would produce big rounds of layoffs or distracting turf wars.

So far, it seems like the statements Access is making are designed to signal investors that Access understands the need to be as non-disruptive as possible at this critical juncture. That's going to be hard, but the very fact that they are saying these things and realizing that this would be a concern of investors is a good sign, IMO. That understanding isn't in itself sufficient to make the merger successful, but it's absolutely necessary.



David Beers
Pikesoft Mobile Computing
www.pikesoft.com
Software Everywhere blog
www.pikesoft.com/blog

RE: Access: You can trust us™.
AdamaDBrown @ 9/12/2005 6:01:47 PM # Q
How many times has it happened where the management changes in your favorite restaurant and they promise everything will be the same? Only it just never seems to work out that way...

While that's true, we can still hope. Some amount of change could be a good thing, cleaning out some of the people who are just useless. I doubt that the acquisition will really change anything about Cobalt, but it may help actually bring about a new OS eventually, rather than a Cobalt redux.

RE: Access: You can trust us™.
sr4 @ 9/12/2005 6:09:26 PM # Q

Well, Ive learned some interesting information from out belligerent Access Engineers.

1) NVFS is being dropped.
2) Cobalt type processes are being dropped.

This implies to me that less than I thought is being carried over to POL. Its a whole new platform (like styletap on WM) and I wish them the best of luck. The less mistakes the carry over from POS 5/6, the better.

Surur

RE: Access: You can trust us™.
AdamaDBrown @ 9/12/2005 6:17:26 PM # Q
Is it flash storage that's being dropped, or just the implementation? After all, flash storage is actually a very good idea, if it's implemented right. It just got buggered up the first time.

RE: Access: You can trust us™.
cervezas @ 9/12/2005 6:30:27 PM # Q
1) NVFS is being dropped.
2) Cobalt type processes are being dropped.

Linux does the file system differently and I'm pretty sure that includes non-volatile file systems. And Linux does multi-tasking differently in terms of how processes are created to handle background threads. None of this is the result of the acquisition, it is a result of the decision by PalmSource to move to Linux last December.

Incidently, for all the griping here about how background threads in Cobalt share a single process and therefore isn't "true" multitasking, there actually is a pretty reasonable justification for this approach and it's not at all clear to me that the Linux way is the better way.

[Note: the following is a paraphrase of stuff that Dianne Hackborn has explained elsewhere] Basically the difference between multithreading in Linux and Cobalt is a trade-off between robustness and performance. Multitasking with ARM Linux (or Windows Mobile for that matter) involves more of a performance hit than in Cobalt because Linux's model of creating a separate process for every thread (including background threads) prevents it from using a much faster method of context-switching (juggling threads) that is used by the Cobalt kernel.

On the other hand, as has rightly been noted, running multiple backgound threads in a single process does mean that if one thread crashes the process it takes the others down. It does not take the main (launched) application down--it has its own process--and it wouldn't even take down a legacy application that was launched since PACE also has its own process. But if you were trying to do multiple things in the background like downloading a document from the Internet and playing an MP3 while reading your email in the "foreground" it would be possible for a problem in the MP3 player to terminate your download. I'm not sure how often people would want to do two background tasks while performing a third in the foreground, but this limitation could occur in that circumstance. In exchange for the performance hit, Linux avoids this scenario since every thread has its own process.

System development, like almost anything else in the world, is an exercise in making good compromises with limited device resources. So the question is, what do you care about most? Do you want to be sure that when you multitask three or more applications simultaneously there's no chance of a bug in one of them crashing any of the others? Or would you rather have your download proceed speedily even though you're listening to an MP3 while it happens? Or maybe have a device that doesn't require such a hot processor to provide good multitasking performance and therefore has longer battery life?

The fact that the Cobalt engineers chose to spend a bit of robustness to buy some better performance is a sign that they are lazy or bad engineers. Good engineers embrace tradeoffs like this and seek to find the compromise that delivers the best overall user experience. A lot of times its a judgment call.


David Beers
Pikesoft Mobile Computing
www.pikesoft.com
Software Everywhere blog
www.pikesoft.com/blog

RE: Access: You can trust us™.
sr4 @ 9/12/2005 6:31:02 PM # Q

They are sticking with flash, but dropping the NVFS implementation. The are moving to a traditional Linux flash-optimized file-system. I wonder if they will also implement a proper file system also (directories etc). I would think the Linux subsystems would be expecting it, so I'm sure they will implement it.

Surur

RE: Access: You can trust us™.
neuron @ 9/12/2005 6:50:32 PM # Q
NVFS doesn't serve any good so far, but act like a buildin disk. Using Resco explorer, it's even treated as buildin disk A. If they remove NVFS, it won't do an bad to POS. Treo 650 can be faster if they don't use NVFS.

RE: Access: You can trust us™.
hkklife @ 9/12/2005 6:54:56 PM # Q
EVERYTHING can be faster if they remove NVFS.

Let's REALLY hope for a standard, at least moderately robust, out-of-the-box-ready backup app as part of the *OS*.

I'd wager that a proper file system and subsequent app associations to handle those files would be a no-brainer as well.

Surur, did you did up any additional tidbits past what you've already posted?

Cobalt-type process being dropped=no chance of revisiting Cobalt in any way, shape, or form as well?



RE: Access: You can trust us™.
sr4 @ 9/12/2005 6:59:21 PM # Q

Nothing else really (except that PSRC engineers want you to pay to know their road map). This is all about POS on Linux, not Cobalt. With POL about two years away, its not impossible that they will release and interim OS such as cobalt. I think with every OS change however they lose some more developers and a whole lot of legacy apps. POL will be nothing like POS 5 internally, and probably even the UI will be different.

Surur

RE: Access: You can trust us™.
cervezas @ 9/12/2005 7:18:44 PM # Q
Surer wrote:
its not impossible that they will release and interim OS such as cobalt.

I'm not sure if you meant that they'll release another version of Cobalt (1.2, for example) or if you are just being loose with your words and meaning to say "release Cobalt devices."

If you meant the latter, it's not only not impossible, but I've been assured as recently as a six weeks ago by PalmSource that multiple Cobalt devices are in the works. You can believe that this is a lie, but you can't deny that it's been publicly stated recently and with full acknowledgement about how much the earlier promises of Cobalt devices have come back to burn them, by the way.

I must say, I find the statements that Access has an interest in looking at Cobalt to be... surprising. In my case that's not a technical dis against Cobalt (too early to say much about it); I just hope we can keep our eye on the Linux ball right now.



David Beers
Pikesoft Mobile Computing
www.pikesoft.com
Software Everywhere blog
www.pikesoft.com/blog

RE: Access: You can trust us™.
cervezas @ 9/12/2005 7:35:10 PM # Q
I wrote:
The fact that the Cobalt engineers chose to spend a bit of robustness to buy some better performance is a sign that they are lazy or bad engineers. Good engineers embrace tradeoffs like this and seek to find the compromise that delivers the best overall user experience. A lot of times its a judgment call.

Obviously, from the context of what I wrote that should have been "The fact that the Cobalt engineers chose to spend a bit of robustness to buy some better performance is not a sign that they are lazy or bad engineers."

Sigh.

David Beers
Pikesoft Mobile Computing
www.pikesoft.com
Software Everywhere blog
www.pikesoft.com/blog

RE: Access: You can trust us™.
sr4 @ 9/12/2005 7:52:27 PM # Q

By release I mean drop POS 5 and push POS 6. How many platforms can they really support concurrently?

I think if there is one thing thats clear, its that over the years optimizations for performance at the expense of robustness always comes back to bite you in the future. I'm sure as a software engineer you can think of many many examples.

Surur

GSPDA Cobalt Phone
Admin @ 9/12/2005 7:53:38 PM # Q
GSPDA did state at the PalmSource DevCon this year that they promised a Cobalt Smartphone for the US market by the fourth quarter.

-Ryan

RE: Access: You can trust us™.
neuron @ 9/12/2005 7:56:52 PM # Q
GSPDA is too small company to promise anything. Cobalt may be appeared, but considering the upcoming Plinux is only one or two years away, it won't be widely used. It may not be survived longer than developers fixing their softwares compatiability problem.

But who knows. Garnet should be retired two years ago.

Beersy: Tell everyone how bad that OSWIN Cobalt phone was
The_Voice_of_Reason @ 9/12/2005 8:02:22 PM # Q
Be honest, Beersy. We both know that thing was a piece of steaming feces.

------------------------
Sony CLIE UX100: 128 MB real RAM, OLED screen. All the PDA anyone really ever wanted.
------------------------

The Palm eCONomy = Communism™

The Great Palm Swindle: http://www.palminfocenter.com/comment_view.asp?ID=7864#108038

NetFrontLinux - the next major cellphone OS?: http://www.palminfocenter.com/comment_view.asp?ID=8060#111823

RE: Voice-Of-Dumness is paid by m$ to write this cr4p
Dr Opinion @ 9/12/2005 8:16:50 PM # Q
Be quiet, shill. :)

------
"People who like M$ products tend to be insecure crowd-following newbies lacking in experience and imagination."
RE: Access: You can trust us™.
PenguinPowered @ 9/12/2005 10:15:41 PM # Q
> They are moving to a traditional Linux flash-optimized file-
> system.

Half correct. The flash file systems are flash-optimized. They're not 'traditional'.

> I wonder if they will also implement a proper file system
> also (directories etc).

Yes.

You don't have to wonder, though. The file systems we're using are available as open source. You can even compile them yourself and see how they currently perform.



Marty Fouts
Lead Linux Lizard
PalmSource Core Technology Department

RE: Access: You can trust us™.
PenguinPowered @ 9/12/2005 10:19:05 PM # Q
> I'd wager that a proper file system and subsequent app
> associations to handle those files would be a no-brainer as
> well.

Proper file system is a no brainer. Backup app is a different kettle of fish. I have no idea what the licensees or carriers will want to do in that area.

> Surur, did you did up any additional tidbits past what you've
> already posted?

He's just citing me and Lefty from another thread. And we're just quoting stuff that was made known at DevCon. I also gave the names of the file systems that we plan to use fo the different kinds of flash.

> Cobalt-type process being dropped=no chance of revisiting
> Cobalt in any way, shape, or form as well?

The process thing is sort of a red herring. Processes aren't being dropped, so much as applications are each getting their own process -- It is Linux, after all.

Applications that were multithreaded before will still be multithreaded. But there's no magic way to make singlethreaded applications magically be multithreaded. That's up to the application writer.

Marty Fouts
Lead Linux Lizard
PalmSource Core Technology Department

RE: Access: You can trust us™.
AdamaDBrown @ 9/13/2005 3:55:43 AM # Q
David Beers said:
I've been assured as recently as a six weeks ago by PalmSource that multiple Cobalt devices are in the works.

Ow. My eyes just rolled all the way up into my head, and I can now see my brain. Good thing I'm a touch typist.

Actually, as insane as this sounds, particularly given the history of these kinds of statements, there exists a small possibility of there being a kernel of truth involved here, maybe. Over the last few months, I've been hearing reports about a Treo running Cobalt. Unlike the WM version, there have been no photos, but the information has even been coming from people I would consider reliable.

While I'm hedging my bets, I wouldn't be entirely surprised if this fall brought us two new Treos, running Windows Mobile 5 and Cobalt on nearly identical hardware.

Ryan said:
GSPDA did state at the PalmSource DevCon this year that they promised a Cobalt Smartphone for the US market by the fourth quarter.

As I recall, wasn't GSPDA having trouble getting a carrier in the US to pick up their phones? I know that they originally had scheduled some Q3 releases of the M68 and M98 family that didn't pan out.

D. O., if all you're not going to contribute any actual information or discussion to the thread, as opposed to launching unprovoked personal attacks on others, would you mind being silent?

RE: Access: You can trust us™.
LiveFaith @ 9/13/2005 11:12:10 AM # Q
LG has already shown a prototype that definitely looks like Cobalt. No way it's PL yet. Not sure if it has a sliding keyboard (aka i730) or not, but the OS is for sure.

http://www.churchoflivingfaith.com/images/lg600i.jpg

Cingular has already spilled the beans on the Samsung i733 too, so it's already passed the carrier barrier. That is going to be one sweet Cobalt phone if it is like the WM version already out.

http://www.churchoflivingfaith.com/images/samsungi733.JPG

This is going to be an exciting year! :-)

Pat Horne; www.churchoflivingfaith.com

RE: Access: You can trust us™.
The_Voice_of_Reason @ 9/14/2005 12:18:37 AM # Q
I've been assured as recently as a six weeks ago by PalmSource that multiple Cobalt devices are in the works.

Beersy, that was believable in 2003; acceptable in 2004; but just plain laughable in September, 2005. How long do you think people are going to fall for that dog and pony Cobalt act? It's amlost embarassing hearing people like you talk about Cobalt as if it any manufacturer would be stupid enough to release a device with it installed.

Here's an idea: until a Cobalt device is actually RELEASED in the U.S., how about we all accept that Cobalt was an utter failure that has already been thrown into the dumpster behind PalmSource HQ and will never - I repeat NEVER - be seen on anything other than that piece of feces Oswin Cobalt phone. Deal? Good.

And while you're at it, please stop trying to give Ryan advice on how to run his site. People like you (and just_little_me, and RhinoSteve, etc) should be rounded up and severely ****** up the ***. Repeatedly.

TVoR

------------------------
Sony CLIE UX100: 128 MB real RAM, OLED screen. All the PDA anyone really ever wanted.
------------------------

The Palm eCONomy = Communism™

The Great Palm Swindle: http://www.palminfocenter.com/comment_view.asp?ID=7864#108038

NetFrontLinux - the next major cellphone OS?: http://www.palminfocenter.com/comment_view.asp?ID=8060#111823

RE: Access: You can trust us™.
spit @ 9/14/2005 1:59:53 AM # Q
From a user perspective, I don't care what file system POL uses but I what I would like to see is POL to be able to run TextMaker for Zaurus.

I'm fed up with the current Palm word processors out there all of which provide single click syncs only for Windows PC users running MS Word on desktop.

TextMaker running on POL would reshape the general thinking of other Palm word processor developers for not having the courage to bundle their own desktop word processor so we don't have to cough up another several hundred bucks for MS Office after we purchase a Palm.

RE: Access: You can trust us™.
cervezas @ 9/14/2005 9:17:25 AM # Q
People like you (and just_little_me, and RhinoSteve, etc) should be rounded up and severely ****** up the ***. Repeatedly.

Keep it up, TVoR! Ryan's the one you are hurting, not me. And he's the one who puts the roof over your head here.


David Beers
Pikesoft Mobile Computing
www.pikesoft.com
Software Everywhere blog
www.pikesoft.com/blog

RE: Access: You can trust us™.
cervezas @ 9/14/2005 9:20:56 AM # Q
TextMaker running on POL would reshape the general thinking of other Palm word processor developers for not having the courage to bundle their own desktop word processor so we don't have to cough up another several hundred bucks for MS Office after we purchase a Palm.

Interesting idea. What I'd love to see is for PalmSource to build a deeply extensible plug-in API into Palm Desktop and then open the source. As an Eclipse fanatic I would love to see it built on the Eclipse framework, but I realize that asking people to install a Java environment is probably impractical. Still, if you know about Eclipse that gives you the general idea: create a radically open platform for people to add plugins like word processors and email clients so 3rd party developers can be as active on the desktop side as they have been on the handheld.

David Beers
Pikesoft Mobile Computing
www.pikesoft.com
Software Everywhere blog
www.pikesoft.com/blog

RE: Access: You can trust us™.
spit @ 9/14/2005 9:41:43 AM # Q
Palm building a deeply extensible plug-in API into Palm Desktop would be great. They should even consider going further by open sourcing and releasing Palm Desktop under GPL. Why not? Palmsource gives away Palm Desktop for free already.

This would really tend to stimulate it's uptake into various open source projects such like OpenOffice. Mr. Softy couldn't follow suit because he wouldn't even consider open sourcing Outlook, and no one in the open source community trusts him any way.

It would also be a great way for Palm to enter the Linux forums with a bang and stimulate developer support for POL.

RE: Access: You can trust us™.
cervezas @ 9/14/2005 10:47:30 AM # Q
It would also be a great way for Palm to enter the Linux forums with a bang and stimulate developer support for POL.

Exactly, open source the whole Palm Desktop. Nothing to lose, everything to gain.

RE: Access: You can trust us™.
The_Voice_of_Reason @ 9/14/2005 9:24:01 PM # Q
>>>People like you (and just_little_me, and RhinoSteve, etc)
>>>should be rounded up and severely kissed up the arm. Repeatedly.

Keep it up, TVoR! Ryan's the one you are hurting, not me. And he's the one who puts the roof over your head here.


I wonder why the words "kissed" and "arm" were censored. Odd.

TVoR



------------------------
Sony CLIE UX100: 128 MB real RAM, OLED screen. All the PDA anyone really ever wanted.
------------------------

The Palm eCONomy = Communism™

The Great Palm Swindle: http://www.palminfocenter.com/comment_view.asp?ID=7864#108038

NetFrontLinux - the next major cellphone OS?: http://www.palminfocenter.com/comment_view.asp?ID=8060#111823

RE: Access: You can trust us™.
spit @ 9/14/2005 11:33:54 PM # Q
I concur with you David. Check out this news article from today:

http://www.ictworld.co.za/EditorialEdit.asp?EditorialID=24361

'Open source community embracing Novell's openSuSE project'

Had Palmsource’s BOD made the right decisions, this could easily have read “Open source community embracing Palmsource’s openPalmDesktop project and POL”.

Palmsource / Access needs to realize that open source developers are not stupid. If you scratch their back, then they will scratch yours. You can’t have a one way deal and just show up at the Linux party expecting a warm open armed embrace.

RE: Access: You can trust us™.
PenguinPowered @ 9/14/2005 11:51:01 PM # Q
Well, given that our technical representative (that's me, for now) to the open source community has been involved in open source projects since long before they were called "open source", I'd bet that we are aware of what it takes to interact effectively with that community.

OpenSuSE, by the way, is a "me too" response to Fedora Core.

As I've mentioned here already, we're involved in the open source community and fully expect to have to earn our credibility with that group. We're active in Yaffs, Omap730 support, and Eclipse, and have several other open source projects ramping up.

Neither RedHat nor Novell are in the same position we're in, with respect to business models. They sell to end users, while we sell to licensees who adopt and adapt, for instance. They make their money off of supporting those users, and enterprise users will pay well for support. We're not an end user support company, end user support is our licensees business. As far as open source business models are concerned, we're closer to Nokia's. I wish we were at a point where we had something like their relationship with Scratchbox and Maemo, but there's only so much you can do with a small team (WRT open source, that would be, again, me, for now) in six months.


Marty Fouts
Lead Linux Lizard
PalmSource Core Technology Department

RE: Access: You can trust us™.
spit @ 9/15/2005 2:08:51 AM # Q
Neither RedHat nor Novell are in the same position we're in, with respect to business models. They sell to end users, while we sell to licensees who adopt and adapt, for instance. They make their money off of supporting those users, and enterprise users will pay well for support. We're not an end user support company, end user support is our licensees business.

Hi Marty,

All you say is no doubt true. But who owns and develops Palm Desktop? It’s you guys. Not the licensees. You can’t expect them to do the job of promoting extensibility of Palm Desktop with the open source community. Only you can do that.

And doing so promotes genuine interest from the open source community in future development of Palm Desktop and extension of its conduits into open source projects like OpenOffice, K Office, Chandler, etc.

This, in turn, creates better products for us end users than can be found with competing products like PPC and this drives your licensee’s revenues over those of PPC licensees and this should also be your goal as an operating system vendor.

Working with Eclipse, etc. is fine but why not accelerate the process to a running start by open sourcing Palm Desktop? As David pointed out, you have nothing to lose but potentially everything to gain.

Quote from the OpenSuSE article:

... The community is also directly contributing to the openSuSE.org wiki, creating more than 100 additional Web pages and almost doubling the site's content since its launch.

Increasingly, questions are being asked and answered by this community, fostering a dialogue that will help improve the quality of both the project and the distribution. ...

By open sourcing Palm Desktop, you not provide a focus for open source community to rapidly build on, but it may also reduce your own overhead expenses for support costs as the open source crowd are often eager to answer questions about their projects to users who run into problems.

RE: Access: You can trust us™.
spit @ 9/15/2005 2:13:32 AM # Q
may also reduce your own overhead expenses for support costs as the open source crowd are often eager to answer questions about their projects to users who run into problems.

Sorry, I was referring to support costs to POL developers, not end users.

RE: Access: You can trust us™.
spit @ 9/15/2005 2:17:23 AM # Q
Sorry, I was referring to support costs to POL developers, not end users.

Of course, open sourcing Palm Desktop would also likely result in lower support costs to your licensees who deal with end users.

Although it doesn't reflect directly on your bottom line, it does trickle down to you eventually.

If one company marketing PPC devices ends up with higher support costs than another marketing POL devices, then this impacts eventually on your platforms competitiveness.

Your licensees could then focus more on development and marketing and be more willing to pay higher licensing fees in the future.

RE: Access: You can trust us™.
PenguinPowered @ 9/15/2005 2:42:03 AM # Q
Why we don't open PalmDesktop is an interesting question.

Unfortunately, I'm a Linux Lizard, and don't have any visibility into what the thinking on that is.

I do know that we have some other potential open source projects that we haven't yet decided on, simply because the decision makers have been busy with other things.

I will ask what plans we have with regard to opening PalmDesktop and share whatever answer I can.

Personally, I see you point, and would agree that it is something we should do unless there's a very significant reason not to.

But when it comes to stuff "above the line", ie, outside of the kernel and core utilities, I have about as much influence as you do on corporate decision making.

Some comments, in general, but interacting with the open source community. There is much potential value in doing so, but it does not come without a cost. If nothing else, it takes time to audit your own sources and make sure you hold clear title to them so that you can release them under an open source license.

A good example from my area is that if we have an SDIO driver for PalmOS, there's no way I can convert it to Linux and release it, because it is encumbered by other people's licenses.

A less obvious one, but very important in my area is that when we work closely with a licensee on an implementation on their device, they will hold title to any drivers or kernel modifications for that device; so I can't release that code, even if they will eventually do so. It's theirs, and they release or not on their schedule.



Marty Fouts
Lead Linux Lizard
PalmSource Core Technology Department

RE: Access: You can trust us™.
spit @ 9/15/2005 3:15:20 AM # Q
Some comments, in general, but interacting with the open source community. There is much potential value in doing so, but it does not come without a cost. If nothing else, it takes time to audit your own sources and make sure you hold clear title to them so that you can release them under an open source license.

A good example from my area is that if we have an SDIO driver for PalmOS, there's no way I can convert it to Linux and release it, because it is encumbered by other people's licenses.

A less obvious one, but very important in my area is that when we work closely with a licensee on an implementation on their device, they will hold title to any drivers or kernel modifications for that device; so I can't release that code, even if they will eventually do so. It's theirs, and they release or not on their schedule.

I see your points here. Perhaps this is also why Sun decided to release OpenOffice.org in parallel with, rather than as a replacement for StarOffice.

Yet OpenOffice.org has taken on a life of its own as you know, especially with small businesses and government offices who are placing more emphasis now on open source products to improve cost efficiency.

Perhaps a similar parallel release of OpenPalmDesktop.org alongside Palmsource's own closed edition would allow users to pick which product best suits their needs.

For example, some users may prefer a version of OpenPalmDesktop.org that's integrated with OpenOffice or one that intefaces with Chandler or K-Office, while others may choose to stick with Palmsource's own Desktop that comes with the CD in the box.

Having a variety of desktop solutions for your products like this would likely be a *major* consideration when users try and decide between POL devices and PPC & other competitors.

RE: Access: You can trust us™.
The_Voice_of_Reason @ 9/15/2005 4:21:10 AM # Q
I see your points here. Perhaps this is also why Sun decided to release OpenOffice.org in parallel with, rather than as a replacement for StarOffice.

Yet OpenOffice.org has taken on a life of its own as you know, especially with small businesses and government offices who are placing more emphasis now on open source products to improve cost efficiency.

Perhaps a similar parallel release of OpenPalmDesktop.org alongside Palmsource's own closed edition would allow users to pick which product best suits their needs.

For example, some users may prefer a version of OpenPalmDesktop.org that's integrated with OpenOffice or one that intefaces with Chandler or K-Office, while others may choose to stick with Palmsource's own Desktop that comes with the CD in the box.

Having a variety of desktop solutions for your products like this would likely be a *major* consideration when users try and decide between POL devices and PPC & other competitors.


Here's a better idea: PalmSource focuses their limited energy on quickly creating + releasing the best possible next-generation mobile OS and leaves extraneous bits to 3rd party developers.

The average Joe either uses Outlook or is satisfied with the current Palm Desktop. Wasting precious resources reinventing the wheel (especially when IntelliSync, PocketMirror and The Missing Sync could be licensed and the rest left up to individual users) makes no sense.

PalmSource needs to learn how to PRIORITIZE. Be engineers don't know the meaning of the word. Your suggestionsare not practical for a company with limited resources and a very tight time schedule.

TVoR


------------------------
Sony CLIE UX100: 128 MB real RAM, OLED screen. All the PDA anyone really ever wanted.
------------------------

The Palm eCONomy = Communism™

The Great Palm Swindle: http://www.palminfocenter.com/comment_view.asp?ID=7864#108038

NetFrontLinux - the next major cellphone OS?: http://www.palminfocenter.com/comment_view.asp?ID=8060#111823

RE: Access: You can trust us™.
cervezas @ 9/15/2005 8:56:39 AM # Q
PenguinPowered wrote:
I will ask what plans we have with regard to opening PalmDesktop and share whatever answer I can.

Thank you for inquiring about this, Marty. I'm very interested. Do you happen to know who at PalmSource would be the best person to channel Palm Desktop questions/suggestions?

Understand that I think we all realize that you guys are probably pretty focused on device software right now rather than the desktop. That's as it should be. But if a process was put in place to start auditing and opening the Palm Desktop and HotSync code, even just a piece at a time, I think this would be seen with considerable enthusiasm by the developer community. Certainly this member of the developer community!

Since PalmSource is getting increasingly focused on wireless devices and users are seeing their Palm devices less as PC accessories and more as free-standing mobile computers that might never get synced with a desktop, I'm guessing that there are internal discussions about how much bandwidth PalmSource should be dedicating to the desktop software. It seems to me that the best argument for opening Palm Desktop is the same reasoning that led PalmSource to go to an open source kernel: you have limited resources and need to deploy them toward the things that make the biggest difference for the platform. Palm Desktop is important to many people but probably shouldn't be seen as part of your core development effort when the open source developer community could lift a lot of that responsibility off your shoulders.


David Beers
Pikesoft Mobile Computing
www.pikesoft.com
Software Everywhere blog
www.pikesoft.com/blog

RE: Access: You can trust us™.
cervezas @ 9/15/2005 9:37:20 AM # Q
One more thought on Palm Desktop:

Wouldn't it be nice (for PalmSource, not just for users) if Palm Desktop could be ported to a cross-platform native GUI toolkit like wxWindows? And there was was only one code base to maintain that would support Windows, Mac OS X, Linux variants, etc? (Ok, the "one code base" thing is probably an exaggeration, but you could probably get close, right?)


David Beers
Pikesoft Mobile Computing
www.pikesoft.com
Software Everywhere blog
www.pikesoft.com/blog

RE: Access: You can trust us™.
PenguinPowered @ 9/15/2005 9:46:56 PM # Q
I checked this afternoon, and my first guess is correct. We've made no decision one way or another about open-sourcing Palm Desktop. I have no idea when it might filter up through the pipeline to that review.

I'm sorry, I don't know who is responsible for Palm Desktop.

The problem with a parallel "OpenDesktop" along side Palm Desktop is that someone would have to write it, and we just don't have the resources now.

Marty


Marty Fouts
Lead Linux Lizard
PalmSource Core Technology Department

RE: Access: You can trust us™.
spit @ 9/15/2005 11:24:24 PM # Q
The problem with a parallel "OpenDesktop" along side Palm Desktop is that someone would have to write it, and we just don't have the resources now.

Is there really a need to write a parallel OpenDesktop if you word the licensing so that it can be resold similar to OpenOffice’s terms?

Essentially, you could just release Palm Desktop as is and then continue on in your own merry way (as Sun did with StarOffice) while the open source community takes over in their own directions from that point on.

Reply to this comment
Start a New Comment Thread Top View Full Comment Thread
Achtung! Only the first 50 comments are displayed within the article.
    Click here for the full story discussion page...