Comments on: Palm Signs Perpetual License for Palm OS Garnet

Battery life & lack of 3G aside, it's the best execution yet of a Treo as far as voice/sound/call quality, size/formfactor, and pricing. It's far, far, far from perfect but the 680's bang for the $ factor blows the 700p & the 650 out of the water.

Pilot 1000-->Pilot 5000-->PalmPilot Pro-->IIIe-->Vx-->m505-->T|T-->T|T2-->T|C-->T|T3-->T|T5-->TX-->Treo 700P

Also Access had already lost the rights to royalities for blowing its deadlines for delievering ALP to Palm. Its found money for them.

What I am hope for is first Palm doing a ground up rewrite of Garnet and hopefully getting a purchase option out there for those who might want to up grade. Not likely, but when I dream I like to go first class.

Far more likely, they get the Garnet "look and feel" grafted to a true multitasking OS (Linux) and backward compatability with the "Palm Economy".

"Many men stumble across the truth, but most manage to pick themselves up
and continue as if nothing had happened."
- Winston Churchill

I think assigninig some kind of malice or evil intent to Access based on the bidding war is a bit overboard. It was "business", plain and simple. Access perceived the value of PalmSource (which included the oft forgotten China MobileSoft division) to be what they paid for it. That's it. Nothing more. They may have been terribly mistaken, but companies do not typically make business decisions in an attempt to be "vicious/dirty" to another company. They make decisions based on perceived profitability.

The only people I feel at all sorry for in all of this is the Palm/PalmSource employees and the greater Palm "community" (we the people, if you will).

I'm still waiting for the mythical color HandEra.

Yes, that's probably the understatement of the year!

When you add it all up (er...subtract it all out?) they ended up paying probably 10s of millions of dollars for each Linux developer.

Are salaries that high in Silicon Valley now?

Sheesh.

-- http://www.palminfocenter.com/view_story.asp?ID=8060

You give me $200 million and I'll hire the best contractor programmers in the US at a cool million dollars EACH per year and have you "CMS Plus" before you can shake a stick at it.

Outrageous to the extreme.

Still.

Yeah, they took a huge gamble. But as Gekko said in that links first post, "They're either the dumbest people in the world or the smartest people in the world." Yeah, maybe you will see them on the "top 10 technology acquisition blunders of all time" list in a few years. I'm just throwing out possible explanations for why they did it.

I will say that I do think Access paid too much for PalmSource, but not to the extent I've heard others criticizing. Time will tell.

I'm still waiting for the mythical color HandEra.

The CMS angle is an overhyped mirage that won't pan out. The Chinese market will ALWAYS be closed to foreigners. Japanese Access got suckered by PalmSource and it's devious mo-fo Board of Directors (Gassée & Gang), plain + simple. Due Diligence, indeed. Suckers. Access got bent over with its pants around its ankles and Gassée et. al. didn't even offer a reacharound.

TVoR

Palm saw it as a way to get clear rights to put their modified Garnet PIMS on wince.

Yes, agreed. But in Palm's eye, it apparently isn't so obsolete. What I'm wondering about is, from Palm's perspective, is it about (a) putting the modified Garnet PIMs on WinCE, (b) giving them to an upgrade path to the next "PalmOS" (that Palm is building in-house), (c) making it easier for Palm to keep putting out low-end Palm hardware for the foreseeable future, or (d) some combo of the above? If it is a combo, I'm curious as to "percentages" and what their development plan looks like at this point.

Some corrections:

The bidding war over PSRC wasn't dirty and didn't really involve Palm, who dropped out pretty early.
Agreed, but I would not say "didn't really". That seems to imply that Palm wasn't actually interested. I think they were quite interested in buying back the PalmOS, just not at $300 million, especially considering they probably didn't have a plan to make good use of CMS.

ACCESS did pay too much for PSRC, because they got caught up in the bidding.
Agreed

They weren't buying US Linux developers. They were buying market share in China.
Agreed, but I think it was a little more complex than "either/or". ALPOS (as we have seen it so far) includes Garnet OS. Seems they planned from the beginning to make use of the PalmOS economy.

They never really planned on Palm signing up for ALP.
I'm not yet convinced that was true. I'd love to know if there were any discussions between Palm and Access during the bidding.

They care about China and DoCoMo.
Certainly.

ALP isn't their only revenue source, and sales of their other products plus the DoCoMo investment more than cover the cash flow.

By the way, ACCESS got Palm to pay 44 million for obsolete technology that nobody in the world but Palm wanted. I'd hardly call that Access losing.
Agreed.

(Look at that; I'm feeling quite agreeable today!)

I'm still waiting for the mythical color HandEra.

Access saw this as a way to eek some more revenue out of an obsolete OS.

No kidding.

Palm saw it as a way to get clear rights to put their modified Garnet PIMS on wince.

Nope. It was a way to get clear rights to put PACE on WinCE/Windows Mobile, which will be the way Palm is able to differentiate its products enough to avoid becoming Just Another WindowsMobile Licensee (JAWL). "It's 2 - yes 2!!! - OSes in one!!! Step right up, folks..."

Some corrections:

The bidding war over PSRC wasn't dirty and didn't really involve Palm, who dropped out pretty early.

Motorola might disagree with you there about the "dirty factor", Bubba...

ACCESS did pay too much for PSRC, because they got caught up in the bidding.

They also didn't do due diligence + got suckered by the "1 Billion Chinamen" mirage.

They weren't buying US Linux developers. They were buying market share in China.

They THOUGHT they were buying Chinese marketshare.

They never really planned on Palm signing up for ALP. They care about China and DoCoMo.

Yes, but don't be surprised if DoCoMo lets them wither on the vine.

ALP isn't their only revenue source, and sales of their other products plus the DoCoMo investment more than cover the cash flow.

That cash flow will likely soon turn to a trickle. Access' customers now have alternate sources for the goods previously sourced from Access. Helloooooo Opera...

By the way, ACCESS got Palm to pay 44 million for obsolete technology that nobody in the world but Palm wanted. I'd hardly call that Access losing.

It IS losing if the other properties (including China Mobilesoft) Access acquired in the deal aren't worth $280 million total.

TVoR

Palmsource paid $22 million for CMS so hmmm . . . let me see . . .

I'm going to stick to modified PIMs on win/mob rather PACE on win/mob because of the amount of effort that has been put into the PIMs. PACE would be a big step back at this point.

It is possible that Palm is working with someone like MontaVista or WindRiver to put their PIMs above Linux, but given the state of Linux at this point I would be very surprised indeed if anything ever comes of it.

Moto didn't think the bidding was 'dirty'. They just had a bad case of sour grapes, which the courts will explain to them one of these days.

May You Live in Interesting Times

I'm going to stick to modified PIMs on win/mob rather PACE on win/mob because of the amount of effort that has been put into the PIMs.

Uh, Marty? Have you ever spent any time with the Palm OS SDK? I guess not, because if you had you'd know that getting the Palm OS PIMs to run on WinCE is not a port, it's a complete rewrite. Owning the Garnet source code would have precisely zero value for that project. The only two things Palm can do with what they just bought are:

(1) Make improvements in Garnet so they can keep using it for a while.

(2) Create a Palm OS execution environment that runs on top of another OS (Win CE, Linux, maybe Symbian)

I agree that Win CE is a pretty good candidate for that. Palm could pretty easily do something like Soleus.

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

Is WINCE a better candidate than Linux? If so, why? or why not?

Your #2 is what I had in mind.

They do get #1 out of the deal, but for the practical remaining life of Garnet, they already had that base covered by the existing license. Maintaining Garnet was never something Palm was good at, though.

May You Live in Interesting Times

Is WINCE a better candidate than Linux? If so, why? or why not?

That's a really big question that involves a lot of factors, so I'm not sure that I really have a definitive answer. Linux (as a mobile device OS) is still very young and I don't know how well issues like power management have been worked out by the folks that are doing it. Palm users have very high expectations: they want snappy, multitasking, 3G-enabled smartphones with long battery life and slim form factors. In the short run Palm might be able to deliver something that meets these expectations faster if they were using Win CE as the base OS. If they weren't in such a damned hurry they could probably get what they need out of Linux and avoid becoming completely dependent on Microsoft (which is kind of a scary prospect). Linux is more flexible than Win CE in terms of being able to use it across a broad range of devices with different requirements, and that's a competitive advantage I'd like to think that Palm wants to have. I think they'll need to diversify their products if they are going to keep pace with the demand for personalization which means people want lots of different device categories, not just Treos.

But Win CE is mature, it's probably a lot cheaper to license than Windows Mobile (and has a lot less technical overhead). Intrinsyc is an example of an MS licensee that has developed their own OS on top of Win CE, called Soleus: http://www.windowsfordevices.com/news/NS7100035934.html . It's a feature phone OS, which isn't what Palm wants but it underscores that this must be relatively cheap and easy to do, since the feature phone market is already pretty crowded and commoditized. I wrote back in March that Palm could get a multitasking OS out quick with a Garnet emulator running on Win CE and just a few native Win32 apps--only the ones that really need to multitask: the phone application, browser, email, messaging, music player. Skin those apps to look like the rest, port HotSync to Win32, and the end user wouldn't know that it wasn't all Palm OS. Except for the fact that their email could be downloading and their music playing in the background while they check their calendar or accept a phone call. That might be the quickest way to get a 3G Palm OS phone out that was UMTS-compliant (for the GSM networks). It's a stop-gap solution, though.

Part of why I have been betting that Palm has Linux in mind is because of the statements of insiders that Palm has already been working with Linux for a few years, along with all the hiring of Linux engineers since PalmSource was acquired. If you don't believe that Palm is interested in ALP (and I can't convince myself they are) it's hard to escape the conclusion that they have product plans that involve a Linux-based OS of their own. Long run I still think that's Palm's best bet, but what to do in the meantime? How long is that meantime?

Bottom line: it's a difficult business problem for Palm and it's hard to say right now what the best answer for them is.

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

There are a number of incompatible competing attempts at telephony stacks for linux, none of which are in very good shape.

Two years ago Linux had a reasonable chance of becoming a good OS for the smartphone business, but the events, especially the fragmentation of the commercial developer community, of those two years are certainly put a gloomy look on its prospects; even though there are major cellphone makers who are seriously investing in Linux efforts.

May You Live in Interesting Times

Wow. That's possibly the dumbest thing you've ever posted here, Marty. (And that's saying a lot.)

Palm is now able to use a StyleTap Platform equivalent in their new custom version of Windows Mobile and not worry about getting sued by Access. Coding custom PIM apps for Windows Mobile that recreate the look and feel of PalmOS PIM apps (or even better, good PalmOS PIM replacements like DateBk6) should be a fairly trivial undertaking. A wise Palm would now license (or outright but the companies producing) StyleTap Platform, WisBar, Pocket Informant, PhatNotes, TenGo, and maybe add a couple more companies like Spb Software House and Resco, Ltd. -> in less than 6 months Palm could create a KILLER version of Windows Mobile and end up having the last laugh.

If Palm management has failed to see such an obvious solution, the company DESERVES to die.

TVoR

Linux is a resource hog poorly-suited to smartphones. Furthermore, infighting among Linux geek camps has stalled development. Nokia, PalmSource, etc. have thus far failed in their attempts to (covertly) organize the geek Linux coders and get them to do their bidding in a controlled manner.

TVoR

for the first time in your life you may be on to something here, vampire.

but 320x480 is improbable.

May You Live in Interesting Times

a "true PalmOS" over a linux kernel is a lot like a Lotus-7 body over a humvee frame.

Nicely put. If you don't substantially rework the Palm OS so that you expose the advantages Linux offers (which makes it something very different from Palm OS) then all you've done is recreated Palm OS Garnet with a lot of battery-wasting, processor-bogging system overhead. If Palm is working on making the next-generation Palm OS run on Linux it's going to be quite a bit different from anything the Palm OS faithful have ever seen--even if it is backward compatible.

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

If the third business rumour is correct, we may however see a linux tablet running the Palm PIM's. After all they did look for linux programmers doing exactly that task.
The lifedrive did not do well because of the limitations of Garnet: using linux instead, might provide a better experience. Nokia 770 docet.

--------------------------
Hey Admin: Why do we have to keep two profiles?

* Palm has a perpetual license to Palm Desktop and HotSync Manager, but they did *not* get development rights to these. They agreed to contract with ACCESS on a professional services basis as they have in the past.

* Palm does have a pre-negotiated option to obtain and develop the Palm Desktop source code under the same kind of terms as Garnet OS itself. It will cost them $3M to exercise this option, which expires after 10 yrs.

* Palm has perpetual rights to distribute (and I think develop) the Mac OS version of Palm Desktop, but ACCESS isn't going to give Palm any further support on this: "the Mac OS Desktop is provided "AS IS," and ASA shall have no obligation to re-deliver to Palm the Mac OS Desktop."

* It doesn't look like Palm has the right to develop or redistribute the Garnet Development Environment. No change there.

All I have time to note right now.

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