Does Palm Inc. Need ALP?

[Ed Hansberry]

<a href="http://brighthand.com/article/Palm_Must_Commit_to_Linux">http://brighthand.com/article/Palm_Must_Commit_to_Linux</a><br /><br />Just over a month ago, PalmSource's new owner, ACCESS, <a href="http://www.pocketpcthoughts.com/forums/viewtopic.php?t=46460">announced the Access Linux Platform</a>, or ALP, as the next generation of the venerable PalmOS. Ed Hardy at Brighthand is convinced that PalmSource needs Palm, Inc. to survive, and seems equally convinced that Palm, Inc. needs ALP to survive. Current PalmOS devices run PalmOS 5, which is little more than PalmOS4 running in PACE, the Palm Application Compatibility Environment. It was meant to be a stepping stone allowing developers get used to the new ARM architecture, not to be in the market for four or five years.<br /><br />ALP will theoretically bring "desperately needed features like concurrent multitasking, allowing it to run multiple applications simultaneously." <a href="http://en.wikipedia.org/wiki/O_RLY">O RLY?</a> :? For years we've been told that either we don't need multitasking or have been told Palm does do multitasking, so long as you redefine the term multitasking to suit your needs. I could have told you four years ago PalmOS needed multitasking. Oh, wait, <a href="http://www.pocketpcthoughts.com/forums/viewtopic.php?t=5971">I did</a>. Happy to have you on board Mr. Hardy! :)<br /><br /><i>"It seems obvious to me that Palm, Inc. and the other licensees need to keep up with the times. Or they should if they still want to be in business in a few years. Palm, Inc. must give up its devotion to Palm OS Garnet. No matter how many tweaks it adds to this operating system, it is just too far behind its competitors."</i><br /><br />I couldn't agree more, but I disagree with Mr. Hardy's line of thinking. He is convinced that ALP is the next logical step. Now hold on, because I don't want to shock some of you, but I am convinced the next logical step is selling quite well right now at your local Verizon store, the Palm Treo 700w, a Windows Mobile 5 device. It supports a killer push email system (or will when the MSFP ROM update is available), has Microsoft Office support, supports the latest wireless broadband technologies and is supported by one company that excels at hardware design and has a skilled sales force getting the device into the hands of carriers, and is supported by another company that is busy at work on the next version of Windows Mobile, ensuring it will work with Office 12, Exchange 12 and communication technologies that are still in early testing phases. Why would Palm, Inc. want to add what amounts to a third operating system to its stable of products, an operating system that has no firm availability date, no developer kit available yet and could easily be 12-18 months away from release? Day one, that product would continue to run many PalmOS 4 applications in yet another compatibility environment. Developers were burned once by switching resources to make PalmOS 6 applications for an operating system that never saw the light of day. I suspect many will sit by and wait until products start shipping before committing time to them. So the end user is stuck with a new device and no great software for it.<br /><br />ACCESS definitely needs Palm Inc. if it wants to carry the PalmOS namesake forward and try to restore it to the glory days of 2000, but Palm, Inc. does not need ACCESS or ALP. And will ALP really be a true PalmOS successor as many envision, or will it be more of a Symbian type OS that allows carriers to customize, and allows some degree of tweaking by end users, but not really a full blown consumer oriented OS like Windows Mobile and PalmOS 5? I personally am betting ALP will be a huge disappointment for PalmOS fans. Not because of some technical limitation, but because of how ACCESS will ultimately target it. I don't know if ACCESS gives a flip about having a seller like Palm, Inc. that will sell products with the OS directly to consumers. I suspect they are more interested in working with carriers to make custom devices that become unique to the carrier and tend to have higher profit margins and customer retention.

[JvanEkris]

Great article Ed!

I see your point from a technical viewpoint, but i guess Palm still has a base of (puting it politically correct) dedicated followers who have some dislike of Microsoft. See them as the traditional Apple user. As i (and probably you as well) have seen, these people can be quite passionate about 'their' platform, even as it has it's limitations.

ALP could try to keep connected to these people, play with the "traditionaly good Palm product, from the people who brought you OS 5 before" vibe a bit. It could be a niche, but it kept companies like Apple alive for a long time....

Jaap

[cervezas]

Palm may not need ALP, but it's not for the reason you give (http://www.pikesoft.com/blog/index.php?itemid=57), Ed.

[Ed Hansberry]

Palm may not need ALP, but it's not for the reason you give (http://www.pikesoft.com/blog/index.php?itemid=57), Ed.

i've heard this line of fantasizing before. I think there is zero chanch Palm will roll their own OS. they never have before. PalmOS1-4 were based on the Kadak provided kernel and Garnet is just more of thr same on ARM. Palm cannot compete with MS, Wind River, Nokia/Symbian on platform development.they should focus on what they do well-device design and OS/software customizations. anything else is a repeat of 2002-2006.

[surur]

Palm may not need ALP, but it's not for the reason you give (http://www.pikesoft.com/blog/index.php?itemid=57), Ed.

David, while I agree that Palm is cooking up something in their skunkworks (they would be a pretty poor company if they were not) your justification is flawed. They do not pay a per device license fee for Garnet, so the license for Garnet is a sunk cost. It would be pretty stupid to waste their development resources on a low cost device such as the Lowrider. Secondarily, Hollywood, focused on Europe, would be a pretty poor place to try out a PalmLinux, as it would have all the disadvantages of PalmOS (poor traction in Europe) and Linux (poor, fragmented handheld software platform) in one. Hollywood is almost definitely WM5, like the XDA Exec, XDA mini, Eten M600, HP 6915, Loox T810, Mitac Mio 701, gigabyte g-smart, imate Jamin, HTC startrek, T-mobile SDA etc etc, in other words it would join a well supported software platform instead of being stranded on its own.

Palm's own Linux platform will only arrive in devices in 2007 the earliest, and face a very competitive environment, even internally at Palm Inc, and the next versions of windows mobile and Symbian.

Surur

[alese]

I definatelly agree with Ed.
Palm does not need ALP or Access to survive or be sucessfull and very possibly Access does not need Palm or even want Palm in his strategy with ALP. They will target Phones only, I think that there will be no ALP based PDAs like Zire, Tungsten or LD...

Also why would Palm build it's own OS. Even if it's Linux based it's still a huge thing to develop and especially support a full blown OS. Today even "small embeded" OS is actually big and complicated project that requires a lot of resources, something Palm does not have.
Also considering that they would have to build a platform pratically from nothing (running old SW in emulator is not really usefull), that they'll need to support this platform (probably without other licencees - who would want to licence an OS from the company you compete with) and finally they will have to develop this new platform and keep it competitive for years to come - it's just too much.

Why fragment their precious resources, compete with M$ again and get beaten again?

[wshwe]

The most likely scenario IMHO is Palm switching completely to WM and Symbian then eventually to Windows Vista. Palm will avoid depending entirely on 1 operating system unless that OS has overwhelming market share. Vista should have that overwhelming market presence.

[cervezas]

alese wrote:
Palm does not need ALP or Access to survive or be sucessfull and very possibly Access does not need Palm or even want Palm in his strategy with ALP. They will target Phones only, I think that there will be no ALP based PDAs like Zire, Tungsten or LD...

I agree that ACCESS will probably do just fine without Palm. I'm pretty sure that they had a good idea where Palm would stand with regard to their plans before they initiated the acquisition. While I don't believe we know for sure yet that Palm isn't interested in ALP, it does seem like the writing is on the wall.

Also why would Palm build it's own OS. Even if it's Linux based it's still a huge thing to develop and especially support a full blown OS. Today even "small embeded" OS is actually big and complicated project that requires a lot of resources, something Palm does not have.

Of course the answer is, they wouldn't build their own OS. They'd partner with a embedded Linux vendor like MontaVista, TrollTech, or Wind River and build upon an already established mobile Linux platform. This is still a pretty big undertaking, as I'm not aware that any of these platforms currently bears a lot of similarity to the Palm OS, but by at least one reliable account (http://news.com.com/PalmOne+ponders+OS+options/2100-1045_3-5438347.html?part=rss&amp;tag=5438347&amp;subj=news.1045.20) Palm started working on this even before PalmSource announced their Linux move in 2004, and they began hiring Linux engineers in earnest in last Fall right after losing the bid to acquire PalmSource. That in itself seems rather telling.

Also, understand that Palm doesn't necessarily need a "full blown" Linux platform coming out of the gate. Initially, they could expose only the Garnet API for third party applications and reserve the native Linux interface for a few of their own ROM applications--browser, email, messaging, phone, and audio player, for example. I explain why this makes sense as a short-term strategy here (http://www.pikesoft.com/blog/index.php?itemid=59), but the short version is that this could quickly get Palm OS Garnet compliant with the UMTS 3G standard and greatly improve the multi-tasking abilities of the platform, while saving the operators from a lot of the costly support headaches they get with the Windows Mobile smartphones.

I don't know that this is what Palm is trying to do with their current Linux strategy, but they are definitely doing some kind of Linux product right now.

[cervezas]

They do not pay a per device license fee for Garnet, so the license for Garnet is a sunk cost.
If you know your economics then you know that sunk costs are water under the bridge, not a basis for forward looking decisions. But the Linux platform I have in mind will run Palm apps in a Palm OS emulator much as current Garnet Treos do. No per device fee, so why not?

It would be pretty stupid to waste their development resources on a low cost device such as the Lowrider.

Stupid like it was for Nokia to develop the low-cost S60 platform on top of the Symbian OS? Now, I don't expect Lowrider to give Nokia a serious run for their S60 money, but the point is that Palm really wants to extend it's Treo run down into that mid-priced smartphone territory. To do that they need to reduce the BOM and differentiate enough that they minimize the cannibalization of their higher-margin 700 phones. They may very well figure Linux will help them do that long term. Realistically, it won't help much at the start, but as the platform matured it could help them grab and hold onto a piece of this very large market.

Secondarily, Hollywood, focused on Europe, would be a pretty poor place to try out a PalmLinux, as it would have all the disadvantages of PalmOS (poor traction in Europe) and Linux (poor, fragmented handheld software platform) in one. Hollywood is almost definitely WM5

Your reasoning may be fairly good here (aside from the fragmentation BS) and you probably are right about Hollywood. But the European wireless operators (France Telecom/Orange in particular) are not happy with the current smartphone situation (particularly WM's high support costs) and are putting money behind Linux right now. If Hollywood isn't Palm's attempt to break open this market then another Linux-based phone may be not long after.

[cervezas]

ACCESS definitely needs Palm Inc. if it wants to carry the PalmOS namesake forward and try to restore it to the glory days of 2000....

I don't know if ACCESS gives a flip about having a seller like Palm, Inc. that will sell products with the OS directly to consumers. I suspect they are more interested in working with carriers to make custom devices that become unique to the carrier and tend to have higher profit margins and customer retention.
So which is it, Ed? Does ACCESS need Palm or do they not give a flip?

I'll go with "they don't give a flip." Ok, they might give half a flip, but ALP doesn't really seem to be designed with Palm in mind and I think ACCESS was going after bigger fish anyway. I think Palm knew this too after ACCESS won the bidding war for Palm, which is why they ramped up their hiring of Linux talent around that time.

I'd give only about a 25% chance that Palm will go completely over to Windows Mobile for the Treo line. And almost no chance at all that they will release future non-phone products on Windows Mobile. The company has been built since the split-off of PalmSource to provide greater choice than their competitors: choice of push email solutions, choice of enterprise sync solutions, choice of operating system. That's been their modus operandi for some years and I expect it to continue to inform their platform decisions going forward.

By the way, you should check your facts about Palm's business model: most of their product is now sold through the wireless operators and that will probably only increase over time. I suspect that one of the reasons for their quiet move to Linux is the encouragement they are getting from operators like France Telecom and Orange to do so.

[Ed Hansberry]

ACCESS definitely needs Palm Inc. if it wants to carry the PalmOS namesake forward and try to restore it to the glory days of 2000....

I don't know if ACCESS gives a flip about having a seller like Palm, Inc. that will sell products with the OS directly to consumers. I suspect they are more interested in working with carriers to make custom devices that become unique to the carrier and tend to have higher profit margins and customer retention.
So which is it, Ed? Does ACCESS need Palm or do they not give a flip?
Beersy, what is the word after "ACCESS definitely nees Palm Inc.?" I'll give you a hint. It is a conjunction, and it leads you into the rest of the sentence you left out in your question, which renders your question invalid at best.

[cervezas]

ACCESS definitely needs Palm Inc. if it wants to carry the PalmOS namesake forward and try to restore it to the glory days of 2000....

I don't know if ACCESS gives a flip about having a seller like Palm, Inc. that will sell products with the OS directly to consumers. I suspect they are more interested in working with carriers to make custom devices that become unique to the carrier and tend to have higher profit margins and customer retention.
So which is it, Ed? Does ACCESS need Palm or do they not give a flip?
Beersy, what is the word after "ACCESS definitely nees Palm Inc.?" I'll give you a hint. It is a conjunction, and it leads you into the rest of the sentence you left out in your question, which renders your question invalid at best.

Okey dokey. So I guess that's your back-ass way of saying you agree with me and that your hypothetical proposition that ACCESS might want to "carry the PalmOS namesake forward" wasn't worth the electrons you used to post it?