MobileInfocenter

Comments on: Palm Brings in a New Software Engineer

The New York Times has published a new article on Palm detailing that they have hired a highly regarded software engineer to "respond to the iPhone."

The designer, Paul Mercer, a former Apple computer engineer, began work three weeks ago at Palm on a line of new products, a company spokeswoman said, but she declined to comment further on the project.

Mercer was the lead designer of the Version 7 of the Macintosh finder, the operating system’s graphical desktop. He later founded Pixo, a software tools firm that created a development system used by Apple designers to conceive the first version of the user interface for the iPod.

Return to Story - Permalink

Article Comments

 (44 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

Palm also hired Marc Blank of Chattermail fame

bcombee @ 3/9/2007 12:43:26 AM # Q
(personally, I'm very excited to have Marc on board with Palm engineering... anyway to quote from Marc's posting on the forums at chattermail.com)

"The time has come," the walrus said, "to talk of many things..."

As of this past Monday, I have become an employee of Palm, Inc. and will no longer be actively working on ChatterEmail. This is something of a bittersweet decision for me; ChatterEmail has, after all, been a major part of my life these past few years. On the other hand, I believe that my best chance to make a long-term contribution to the Palm/Treo community is to put my knowledge and experience to work in a larger cause.

RE: Palm also hired Marc Blank of Chattermail fame
mikecane @ 3/9/2007 12:18:25 PM # Q
WTF?

Straighten out the timeline for me so I don't have to call in Doctor Who, OK?

When PalmSource did their Great Linux Now! Purge, where you kicked out then? Or sometime later? When ACCESS bought them?

Then you wound up working on ChatterMail?

And now you will be at Palm?

Why haven't you been updating your blog? Of course I looked!

RE: Palm also hired Marc Blank of Chattermail fame
mikecane @ 3/9/2007 12:24:17 PM # Q
Bloody hell. Re-reading your post, you didn't use quotes, so it looked like YOU joined Palm.

Anyway, where are YOU now, what are YOU doing, and the same Q about the damned blog.

RE: Palm also hired Marc Blank of Chattermail fame
just_little_me @ 3/9/2007 3:14:21 PM # Q
Ben's at Palm - has been for a while from what I recall.


JLM.

RE: Palm also hired Marc Blank of Chattermail fame
Gekko @ 3/9/2007 5:40:25 PM # Q

I think Beersie is on the payroll too as "Chief Propaganda Officer". He took over the role from Kirvin who had the position when it was called "Chief Senior Apologist".



RE: Palm also hired Marc Blank of Chattermail fame
cervezas @ 3/10/2007 4:41:45 PM # Q
Ben left Palmsource for Palm around the time of the original announcement that Palmsource was going to port Cobalt to Linux (2+ yrs ago now). Around the time that it first leaked that Palm was looking at teaming up with Microsoft and Linux vendors on future products.

As for me, I doubt Palm would pay for the weak praise they've gotten from me in recent years. I'll give you this: I'm sure interested to see what will happen this year. In May, in particular.



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

RE: Palm also hired Marc Blank of Chattermail fame
icesalmon @ 3/10/2007 11:33:01 PM # Q
David Beers' opinion is one of the most valuable and informative. In fact, I check back frequently to see if he has posted comments here.
RE: Palm also hired Marc Blank of Chattermail fame
jfinlayson @ 3/20/2007 7:34:02 PM # Q
http://www.chatteremail.com is gone already. That was quick.
RE: Palm also hired Marc Blank of Chattermail fame
jfinlayson @ 3/21/2007 1:36:35 PM # Q
Now it's back, but registered to Palm, Inc.
RE: Palm also hired Marc Blank of Chattermail fame
cervezas @ 3/21/2007 4:17:34 PM # Q
Looks like Palm didn't just hire Marc, they bought Chatteremail. Makes sense since one of Palm's best push email partners (Good Technologies) was acquired away by Motorola.

Doesn't make sense if you're expecting to be acquired by Motorola or Nokia (which already own their own push email solutions).

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

Reply to this comment

myopic move

drw @ 3/9/2007 12:36:02 AM # Q
Paul appears to be going from company to company, riding on his past accomplishments, happy to take money from anybody. And who can blame him for that? Last year Samsung hired him. Excuse me, but iPod is still king of the hill without Mercer.

Palm chunked a million on BeOS and brought their software guys in to help out. We all know what happened to that. When Dallas Cowboy's owner Jerry Jones brought in Bill Parcells, everyone thought Bill would bring glory back to Dallas, but it didn't happen.

You can't manufacture that spark of inspiration. It comes from a bunch of young long hairs sitting around on bean bags like at Palo Alto Research Center.

Look at these new billionaires inventing social networking sites. They aren't pushing 40 years old. They are young and hungry.

Nice try, Palm, but I doubt Paul will stay at Palm for 1 year.

Signed,
T|C devotee (still waiting on a treo without a cell radio, but with wifi-g/n and at least 400 mhz processor and a gig of storage)

---
David

RE: myopic move
SeldomVisitor @ 3/9/2007 6:19:13 AM # Q
Last year he was hired by Samsung? Wow...job hopping is alive and well!

=========

About 27 years ago or so I went to Palo Alto (from Northern Virginia) to take a combined software/hardware training course on a particular graphics/image display (*) for a couple weeks.

It was TOTAL eye-opener.

Those of us who live in mundane suburbia have NO IDEA what it's like to have hundreds, if not thousands, of tech companies arranged adjacent to each other on a gridwork set of wide avenues stretching in all directions.

Where, if you want a new job, you walk across the street to the next company!

Looks like this guy takes advantage of that, eh?

=========

(*) The hardware was a state-of-the-art 512x512 pixel, 12 bits/pixel image/graphics display where pixels could be loaded with values bit by bit or en masse in any x/y direction. There also was a mapping table to translate those pixel values into any other value (the depth of which I forget but it could have been 24 bits).

A whole bunch of TTL logic, essentially.

Very fun device to learn and program at the lowest of levels.

RE: myopic move
mikecane @ 3/11/2007 2:39:07 PM # Q
*slapping head*

Oh sh!t!

THAT Paul Mercer?!!? The one who did Samsung's last MP3 player?!!?

Yep! It be him: http://maku.or.tv/news2006/paul_mercer.jpg

I didn't make the connection til now. Thanks for that.

More info:

http://tinyurl.com/2e8f98
http://tinyurl.com/2759ej

Reply to this comment

Forget it, too late

sinebubble @ 3/9/2007 1:07:11 AM # Q
Three weeks ago?!? Why didn't Palm think to hire someone to refresh their GUI FIVE YEARS AGO!?! What's this guy going to do in response to iPhone in 4 months that Palm was unable to do in the last five years? Resurrect Cobalt?

System 7 lead designer? Oh god, did they really say that?



RE: Forget it, too late
retrospooty @ 3/9/2007 10:04:40 AM # Q
"System 7 lead designer? Oh god, did they really say that?"

No, they didnt say that. they said he was the lead designer of Version 7 of the Macintosh finder.

Either way, an updated UI is in order. I fail to see why they need this guy to do that, but whatever.

Reply to this comment

Putting Mercer's hire in its proper perspective

mikecane @ 3/9/2007 12:20:06 PM # Q
Ryan is falling down on the job. Or else I'm just enough of a b*st*rd to call them on this:

Palm Wants Us To Wait ANOTHER Two Years…
http://mikecane.wordpress.com/2007/03/09/palm-wants-us-to-wait-another-two-years/

RE: Putting Mercer's hire in its proper perspective
ballistic @ 3/11/2007 8:41:46 AM # Q
That's quite a stretch. To suggest that Palm is restarting the clock on all product design and development the day Paul Mercer was hired is silly.

I take Jeff's comments to mean it takes approximately 2 years from the initial concept stages through hardware design, manufacturing, and certifications (carrier and FCC). I'll also remind you that it only took Mercer, Pixo and Apple about 6 months to design the iPod.

Furthermore, Mercer's passion is mobile device user interface design. Don't you think he *might* have some pre-existing work, ideas, concepts, mockups, prototypes, etc that he's bringing to the table, which would all accelerate development at Palm?

I think it's safe to say that the two year clock started a while ago...

RE: Putting Mercer's hire in its proper perspective
ballistic @ 3/11/2007 9:06:59 AM # Q
http://tinyurl.com/2n7h7w

From the last paragraph of the article, it sounds like he's been thinking and working on it for a while, and we can see why he chose to join Palm.

Now that the Z5 is about to reach the market, he is thinking about his next technology target. Mr. Mercer said he was still dreaming about a personal device that offered complete access to information and media.

RE: Putting Mercer's hire in its proper perspective
cervezas @ 3/11/2007 10:11:18 AM # Q
That, and of course it's pretty doubtful that Palm is thinking only one product ahead. Just as a point of reference, Hawkins had plans to turn the Visor into a mobile phone even before the first Handspring Visor hit the store shelves.

David Beers
Pikesoft Mobile Computing
www.pikesoft.com/blog
RE: Putting Mercer's hire in its proper perspective
mikecane @ 3/11/2007 2:43:24 PM # Q
>>>I'll also remind you that it only took Mercer, Pixo and Apple about 6 months to design the iPod.

That didn't require any FCC approval (although I'm certain Jobs was more strict than the FCC ever could be!).

I wasn't impressed at all when I fondled his Samsung MP3 UI. Apparently, neither was the marketplace. It got its a** kicked by the Sandisk Sansa series (which actually doesn't have a very good UI!).

If you think all Mercer will do is slap some makeup on PalmOS, why bother to hire him? PalmOS needs a *lot* more work than, say, the Foo Fighter Facelift. And that will take more than just 6 months. And I frankly don't think Palm has any PalmOS *hardware* coming out this year other than maybe another fugly TX-like device and more Bleh! Treos.

Reply to this comment

palm's product pipeline

nybble @ 3/11/2007 10:52:36 AM # Q
I also read the Hawkin's interview to say that the clock started when they bought back their own OS. Which means, more or less, now. I'm super interested to see what Hawkin's project is - if it isn't a smartphone I wonder what it could be, I hope he does as well as he has in the past for the company.

As for the treo's and Palm thinking one product ahead. I really, at this point, don't give them the benefit of the doubt. It's possible that they think of more than one product in the future, but if they don't actually conceive of those products as being different from each other, I don't know what good it does them. I'm glad they hired the guy, but wonder how much difference he can make - I'll be overjoyed to be pleasantly surprised.

I'm sad that chatteremail has been suspended. Of all apps that's the most key one for me. I hope that Marc can make a difference at Palm.

I love my 650 dearly, but recognize that Palm has squandered a huge industry lead over the past several years. I wrote a little more of my opinions:

http://comments.deasil.com/2007/03/07/for-sale-palm/

RE: palm's product pipeline
cervezas @ 3/11/2007 12:35:43 PM # Q
I don't think Paul's impact will be seen very much in any products Palm releases this year. That's probably just some dumb journalist looking for an angle to attract readers. But I've got good reason to believe that Hawkin's announcement in May isn't going too be "Palm's new product pipeline is going to be empty for the next two years, followed by something really great so please be patient and pay no attention to that iPhone over there."

One thought to consider: Palm has had the source code for Garnet from the very beginning and has been free to put people on the task of building and testing a new Palm OS for years. The buyback was just the deal that allowed them finally to release it in a product. For the many reasons I've discussed before, I seriously doubt that Palm is just now getting started on their new OS.

On a more cautionary note, we should remember that revolutionary new products don't always take the world by storm in their first iteration. The Treo didn't really have an impact until its 3rd rev (the 600).

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

RE: palm's product pipeline
Scott R @ 3/11/2007 1:29:37 PM # Q
David, I'm fairly certain that Palm did not "buy back" Garnet. It is my understanding that they got an unlimited license to use Garnet and more access/rights to the source code. As you stated, they had access to the source code previously, but my understanding is that in the past any licensee could modify the OS but had to give that code back to PalmSource who then could opt to integrate any/all/none of those changes into subsequent OS releases, thereby sharing the licensee's development work with other licensees. I suspect that part of this new agreement allows Palm to change the Garnet source code without needing to share their changes with ACCESS.

http://Tapland.com
- Tapwave Zodiac News, Reviews, & Discussion -
RE: palm's product pipeline
Gekko @ 3/11/2007 2:23:35 PM # Q


Time is fleeting.



RE: palm's product pipeline
nybble @ 3/11/2007 2:36:55 PM # Q
David,

I definitely agree with you that Hawkin's announcement won't be that Palm's got no new treo's for the next two years. :) But I think Hawkin's announcement won't have anything to do with Treo's at all - I think it'll be a product in a different space. That's the impression, at least, that I got from his interview. And I'm excited to see what he brings to the table.

I am, however, much less sanguine about Palm's product line in the short and medium term. Believe me, no one would be happier than me to be proven wrong about this, but I just don't see them turning around anything worthwhile in the next year to two years. I blogged a little bit about what good might come of the two publicized recent hires (http://comments.deasil.com/2007/03/11/chatter-email-no-more/) but I just don't think they've come up with anything good on their own.

RE: palm's product pipeline
joad @ 3/11/2007 4:39:42 PM # Q
If Palm is "thinking 2 years ahead".... then do we surmise that the 700p is the result of looking at the 600 or 650's problems?

I wonder what they did with those 2-3 years leadtime..... Based on what they released (one with less memory than the 600, the other with plenty of obvious bugs and performance issues), the word "thinking" might need to be looked at more closely. Up until now, Palm has been a poster child on how to take a brilliant idea and muck it up so badly that even HP and Motorola can produce garbage that tempts people away.

Hopefully Palm has realized they've hit rock bottom, and need to finally begin releasing some devices with some quality and intelligence again. Hire Handera, the Zodiac designers, the Toyspring guy, and a few other people who have a clue about how to creatively and cleverly make the Palm something that stands out from the crowd and offers considerable value to the purchaser. Otherwise they're just another WinceMob clone company, and they'll get swept aside with the rest of them.

When even *I* have fleeting thoughts of going back to a plain old cell phone instead of paying $650.00 every year or so to beta test their latest crap... things are pretty bad in Palm-land.

RE: palm's product pipeline
PenguinPowered @ 3/12/2007 4:25:32 AM # Q
Two years ago, Palm definitely did [i]not[/i] start to work on anything that might be considered an iphone competitor.

Two years ago, they were working on winmob and trying to figure out what to do about Palmsource and Cobalt, er, PalmSource and the freshly announced Linux project.

One year ago, Palm definitely did [i]not[/i] start to work on anything that might be considered and iphone competior.

One year ago, they were working on winmob and trying to figure out what to do about Palmsource, er, Access, and PalmLinux, er, ALP.

But then, why [i]should[/i] Palm worry about an iphone killer? iphone's aimed at an entirely different market than Palm's. Palm's worried about Crackberries and Nokia or Motorola finaly figuring out a UI.

Maybe Access will buy Palm. ;)


May You Live in Interesting Times

RE: palm's product pipeline
twrock @ 3/12/2007 7:17:18 AM # Q
Maybe Access will buy Palm. ;)

Shhhhh, quite, that isn't supposed to leak just yet.


Thinking about Vista? Think again: http://www.cs.auckland.ac.nz/~pgut001/pubs/vista_cost.txt
Want an alternative? Try this: http://www.ubuntu.com/

RE: palm's product pipeline
cervezas @ 3/12/2007 8:20:23 AM # Q
For the record, I wasn't suggesting that even today Palm is thinking about developing something that you could call an iPhone competitor, which is why I said the "iPhone response" spin on Paul Mercer's hiring is just a not-terribly-bright reporter looking for some spin on a slow news day.

Palm sees mobile computing as both the road and the destination. Meanwhile, Apple has pointedly dropped "Computers" from the company's name and developed a closed-platform phone that targets the consumer electronic segment where it made its iPod successful. While there may be some overlap in features, there is very little in intention. Apple can afford to release an impractical device that doesn't solve any problems and gets by on style. Palm needs to solve real problems and advance its idea that "the future of computing is mobile computing."

ScottR wrote:

David, I'm fairly certain that Palm did not "buy back" Garnet. It is my understanding that they got an unlimited license to use Garnet and more access/rights to the source code.

No. What they bought was an expanded right to develop the Garnet OS itself, to patent the changes, and most importantly to make it part of another operating system. From Palm's press release after the deal:
The new agreement also provides Palm flexibility to use Palm OS Garnet in whole or in part in any Palm product, and together with any other system technologies. The company plans to ensure that applications now compatible with Palm OS Garnet will operate with little or no modification in future Palm products that employ Palm OS Garnet as the company evolves it over time to support Palm's product differentiation strategy.... Palm has secured an expansion of its existing patent license from ACCESS to cover all current and future Palm products, regardless of the underlying operating system.


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

RE: palm's product pipeline
SeldomVisitor @ 3/12/2007 9:26:18 AM # Q
The hiring of some non-executive is not usually grist for the PR mill, yet PALM has now done this twice recently (ya think that "news" came out of thin air?).

I can imagine there are a number of bruised-ego already-existing PALM software guys contemplating their navels right now - and some Engineering management that might be upset with Marketing.

Given the hoopla, there's a lot on Mr. Mercer's shoulders right now - maybe he's pissed, too.

RE: palm's product pipeline
Scott R @ 3/12/2007 9:56:23 AM # Q
cervesas wrote: "No. What they bought was an expanded right to develop the Garnet OS itself, to patent the changes, and most importantly to make it part of another operating system."

I'm not sure what the difference between what you said and what I said is. My point was that Palm does not own the Garnet OS, ACCESS still does. Are you disagreeing with that? If so, I'd like to see some better links to support that.


http://Tapland.com
- Tapwave Zodiac News, Reviews, & Discussion -

RE: palm's product pipeline
ballistic @ 3/12/2007 11:09:37 AM # Q
ScottR wrote
I'm not sure what the difference between what you said and what I said is. My point was that Palm does not own the Garnet OS, ACCESS still does. Are you disagreeing with that? If so, I'd like to see some better links to support that.

At the Merrill Lynch 2007 Communications Conference http://tinyurl.com/2ht3g5 , Palm CFO & SVP Andy Brown said the following:

We purchased a perpetual license to the source code, which allows us to basically do any innovations that we want on top of that. And what that allows us to do is control our own destiny...

....It does allow us to control our own destiny so that we can now specifically customize that operating system for our products and our products only....

...Strategically, it's important. We haven't gotten into the details, but I think the ability to control our own destiny on that platform is strategically important, so we can customize specifically for Palm and Palm Treos, that user experience on that platform.

Sounds like they bought the perpetual rights to Garnet source code, and any changes Palm makes remain proprietary to Palm.

RE: palm's product pipeline
hkklife @ 3/12/2007 11:49:25 AM # Q
To these unlawyer-ly eyes, a perpetual license is NOT the same as owning the IP, though in practice there's not much difference.

What Palm & Access reached is probably the best of both worlds for all parties involved: Palm gets out of paying any more licensing fees, they own whatever new appendages they sew onto FrankenGarnet, and Access can still use "Garnet OS" if/when they feel the need and still license it out to the odd rogue licensee (Janam) that may appear.

Now, if the Access purchase of Palm does indeed come true as some are whispering, then this whole "Palm" trademark acquisition + Garnet perpetual license puchase will be yet ANOTHER shell game played out to brilliant execution by Palm management.

I'm actually much more curious to see what Palm's actual hardware roadmap looks like. I think they are still trying to tread water by trickling out one new model per quarter and trying to bide their time until they are acquired.

In all fairness, do they really need Mercer's expertise *NOW*? I mean, how much could FrankenGarnet be overhauled UI-wise until something becomes horribly broken? I personally thought the UI updates to the 700p & 680 are fine and a step in the right direction. Palm needs to make these uniform across all of Garnet--dialog boxes, menus, prompts etc. and not in just in a handful of updated apps.

And how much will M$ let them tweak WM6?

Perhaps whatever tasks Mercer will be doing are related only to the "next big thing" and/or Palm's own home-brewed Linux solution?

Perhaps saddest of all is that Palm could have SO easily released something like Pat's Treo 800g a year or more ago as a data-centric style of Treo. Its underlying innards could then eventually trickle down to something remotely approaching the iPhone in thinness. Regardless of the iPhone's limitations (and there are many) in its initial version, it's going to be an eye-opener in regards to its display.
320x320 and 240x240 smartphone screens are going to be rendered instantly obsolete in the eyes of most consumers once the iPhone hits the streets, just like how no one wants a 4:3 TV or PC LCD anymore in this day of affordable widescreen LCDs.

Palm's been doing 320x480 LCDs since 2003. It's sad that they have beaten such a hasty retreat from ONE of the things that Garnet still does well.


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

RE: palm's product pipeline
SeldomVisitor @ 3/12/2007 12:08:48 PM # Q
> ...if the Access purchase of Palm does indeed come true as some are whispering...

Docomo, I tell ya, Docomo!

Giggle.

RE: palm's product pipeline
cervezas @ 3/12/2007 1:41:18 PM # Q
I think we're all on the same page.

Palm doesn't own Garnet OS in the sense that they can license it to another company. The other encumbrance on the IP is that they have to comply with an agreement with ACCESS that they will submit any product they release that uses Garnet code to compatibility testing against a pre-agreed test harness, so that Garnet apps are assured to run without modification on either ALP or the OS that Palm develops. ACCESS has to do the same. I'm really curious what that test harness looks like. It's a vital, but publicly invisible part of the agreement which was probably the subject of considerable wrangling as it has implications for how far Palm and ACCESS can run in different directions without losing a limb. This image comes to mind: www.bluebeat.com/i/a/l/l6334.jpg :-)

Palm does own its own work, royalty free and unencumbered by patenting restrictions. With the caveat's just mentioned, the "Palm OS" that they develop with the ACCESS code is their IP in every sense of the word that Palm or its customers cares about. If we agree on that then I think we agree in substance, even if you don't like my choice of the words "bought back" in connection with the Garnet OS.

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

RE: palm's product pipeline
Scott R @ 3/12/2007 2:59:09 PM # Q
Right. But let's look closer at those two restrictions...

1) They can only use their Garnet code on their own devices. If they totally gutted and improved the OS (like improving multitasking support), they could potentially have a mobile OS that was attractive to other licensees, but they would be restricted from licensing it out.

2) How will this forced compatibility work? What if they wanted to totally gut the OS (again, to improve multitasking support) and, in the process, broke backward compatibility. Are they really prevented from using that OS in a device? How does one define "backward compatibility" anyway? There are old apps that won't run on my 700p and year-old apps that work on a 700p but not on a TX. Having said all that...Based on my rusty recollection of one of these press releases, I'm not certain that they *are* required to maintain backward compatibility. Rather, they may only be required to do so if they want to "market" the new device as being Palm OS / Garnet OS-based. So if they *did* totally gut the OS, they might be able to opt out of the backward compatibility testing if they marketed it as an all-new OS.

http://Tapland.com
- Tapwave Zodiac News, Reviews, & Discussion -

RE: palm's product pipeline
nybble @ 3/12/2007 3:06:46 PM # Q
I have to admit that I'm really not all the excited by the repurchasing of Garnet. Is it something that's so awesome we want to keep using it for another 5 years? My understanding, which could be flawed, was that it is never going to support background processing and protected memory. What is the hooplah about Palm being able to continue development of this platform instead of innovating a modern os with all the fixins?

I think also tend to think that the notion that an iPhone that lives up to its hype is not a treo competitor is optimistic. The key app for many treo users is email and sms, if the iPhone has that down, it will start to eat Palm's lunch. I think there's a notion that because it's new and flashy that it is a toy, and I don't get it. The apps I use and most of the people I know are chatter (or snapper), a launcher or two, an rss reader and maybe the occasional game. Apple won't mess up their mail program and they won't mess up the launcher. The web browser they provide will crush blazer with javascript and flash support as well as a wider screen. Throw in easy itunes syncing and stereo bluetooth headsets with a not busted bluetooth implementation and I'm sold. They're good at making software that is good looking and easy to use. People like that.

RE: palm's product pipeline
mikecane @ 3/12/2007 3:30:02 PM # Q
>>>how much could FrankenGarnet be overhauled UI-wise until something becomes horribly broken?

Uh, once Mercer tarts up the UI, EVERYTHING will be broken, won't it? How many apps not currently written to take advantage of the new lipstick and rouge of a Mercerized OS 5.x will work? Isn't that answer NONE?

RE: palm's product pipeline
SeldomVisitor @ 3/12/2007 5:05:14 PM # Q
Just an aside w.r.t. purchasing major rights with PalmOS - this was, as the CFO said in HIS aside, an absolute NO-BRAINER =simply= based off of licensing costs on a per-unit basis compared to this "purchase" - within, I believe ONE YEAR its ROI goes positive.

As such, PALM could THROW AWAY THE WHOLE DAMN THING in a year or two and come away with a HUGE smile on their CFO's face.

RE: palm's product pipeline
ballistic @ 3/12/2007 6:08:56 PM # Q
What he said was that financially, as a CFO, he thought it was a good deal, but that wasn't the main reason:

What we really wanted out of the Palm OS, you know the acquisition of the license, was the ability to get those innovations out sooner. We weren't seeing that from PalmSource, let's just face it, they weren't executing, and it was causing us more pain than it was them. We were paying them 2.5% for every unit that shipped, and yet they're not doing anything. What we like to say that if you look at the UI on the Palm operating system, it's still a great UI. It works seamlessly, it's a great house, but the foundation's a little bit rocky, and what we're going to do initially is shore up that foundation, is really the key.

That doesn't sound like they plan on throwing anything away, but rather Palm plans build on their investment..

RE: palm's product pipeline
cervezas @ 3/12/2007 6:25:00 PM # Q
ScottR wrote:
But let's look closer at those two restrictions...

1) They can only use their Garnet code on their own devices. If they totally gutted and improved the OS (like improving multitasking support), they could potentially have a mobile OS that was attractive to other licensees, but they would be restricted from licensing it out.

...and that business plan worked so well for Palm the last time, didn't it? Honestly, I think Palm's interest in relicensing a new OS is about the same as Apple's interest in licensing OSX to competitors.

2) How will this forced compatibility work? What if they wanted to totally gut the OS (again, to improve multitasking support) and, in the process, broke backward compatibility. Are they really prevented from using that OS in a device? How does one define "backward compatibility" anyway? There are old apps that won't run on my 700p and year-old apps that work on a 700p but not on a TX. Having said all that...Based on my rusty recollection of one of these press releases, I'm not certain that they *are* required to maintain backward compatibility. Rather, they may only be required to do so if they want to "market" the new device as being Palm OS / Garnet OS-based. So if they *did* totally gut the OS, they might be able to opt out of the backward compatibility testing if they marketed it as an all-new OS.

From the press release: "ACCESS and Palm have agreed to continue to measure compatibility against the compatibility test harness in use between the two companies."

Based on that terse statement I'd guess that compatibility is based on the same testing criteria that all the other Garnet licensees have submitted to over the years (Tapwave, Garmin, Janam, GSPDA, maybe even Sony and Acer). Some of those were pretty darn incompatible but still passed the tests (different hi-res APIs, different dynamic input area APIs, special requirements to manage flushing data to non-volatile RAM... Palm had significant incompatibilities just between different models of the same product series.)

I don't think they could gut it to the point where a huge number of applications would be broken, but would they want to? The interesting question, as you point out, is just how far could they go without breaking basic compatibility? There are at least a couple different ways to answer this.

I do think that it would be technically feasible to launch Garnet-compatible applications in their own process and add the capability to switch between running processes. Palm would need to create a new launcher and launcher API, and legacy third party launchers wouldn't multi-task without being rewritten to take advantage of that API. But as long as the old launchers still worked it seems to me a multitasking version of Palm OS could pass compatibility muster. A lot of apps that use undocumented APIs to make up for deficiencies in Garnet would be broken, but everyone knew that could happen anyway and presumably there would be new, fully supported APIs to do many of the things that had to be hacked before. One sticking point might be Global Find. Have to think about that one.... Garnet has an elegant, but rather unique way of implementing Find, which would perform poorly on a system that has to spawn a new task any time it wants to run code inside the application. The old way might still technically "work" (enough to pass compatibility testing) but they would have to come up with a new recommended way to support Find for real.

This approach sounds like what Andy Brown was alluding to when he talked about putting better foundations under the Garnet OS: minimize the number of new APIs and focus on making exisiting applications work in a modern, multi-tasking way.

Remember that Palm OS Cobalt added lots of great graphics capabilities to the Palm OS without breaking Garnet compatibility. It did it by integrating Garnet essentially the same way that ALP does: as a separate runtime environment. That's the other way Palm could get around compatibility roadblocks since the kind of apps that can run natively outside that environment is wide open to Palm from the standpoint of the agreement with ACCESS. But this is also more ambitious since it would require Palm to create an all new SDK to support the next generation applications (remember Protein? And now ACCESS's MAX framework?). Even if they didn't initially make the new APIs public and just used them internally, it's a big project.

As a bit of an aside, here's something to ponder: how hard do you think ACCESS is working on NetFront for Palm OS these days? How important do you think a great browsing experience is to Palm's future product plans? Where are they going to get that great browser? Standards like Linux/GTK+/GStreamer (or even Java) have got to be looking pretty good to Palm these days. Doing something that's very proprietary under the hood (like Cobalt) will make rolling out products even slower than it is for them today. Too much custom development. What does that mean for Palm's approach to building its next-generation OS?

Anyway, I think you can see a couple of different ways that Palm could keep their compatibility agreement with ACCESS and still roll a pretty new "Palm OS" for themselves.

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

RE: palm's product pipeline
cervezas @ 3/12/2007 7:50:01 PM # Q
SeldomVisitor wrote:
Just an aside w.r.t. purchasing major rights with PalmOS - this was, as the CFO said in HIS aside, an absolute NO-BRAINER =simply= based off of licensing costs on a per-unit basis compared to this "purchase" - within, I believe ONE YEAR its ROI goes positive.

Sure. But you don't think ACCESS conceded to Palm that language about using Garnet with different "underlying operating systems" for free, do you? Why would Palm pay for this language if they didn't care about it?

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

RE: palm's product pipeline
freakout @ 3/12/2007 10:19:02 PM # Q
Nybble:
The key app for many treo users is email and sms, if the iPhone has that down, it will start to eat Palm's lunch.

Without real buttons to type your messages on, the iPhone is going to be a terrible messaging machine. The problem with touchscreens instead of buttons is that you can't keep your thumbs resting on the screen, else you're likely going to accidentally press a button that you don't intend to. Instead, you have to keep them hovering above the screen and use your own muscles to bring your thumb back up after a press. This isn't an issue with physical buttons, which have a small degree of resistance and spring back into place after being pushed. It only sounds like a little thing, but for prolonged typing sessions it will be a royal pain in the bum - or thumb, if you like (ho ho!) - super-intelligent predictive text or not.

So, in other words with no stylus and no buttons, thumb-typing Treo-style on the iPhone will be virtually impossible.

Imagine what it's going to be like typing in URL's...

mikecane:
Uh, once Mercer tarts up the UI, EVERYTHING will be broken, won't it? How many apps not currently written to take advantage of the new lipstick and rouge of a Mercerized OS 5.x will work? Isn't that answer NONE?

If Palm went the PalmRevolt route of just replacing the standard PalmOS widgets - scrollbars, buttons and the like - then things could be significantly improved without breaking compatibility. But I'd hope they'd go further than that, and just offer an option where you can set older apps to revert to the "old" PalmOS look.

Tim
I apologise for any and all emoticons that appear in my posts. You may shoot them on sight.
Treo 270 ---> Treo 650 ---> Crimson Treo 680

Reply to this comment

And an old software engineer moves on

SeldomVisitor @ 3/20/2007 4:49:58 PM # Q
John Backus has died - thought that needed mentioning SOMEWHERE.

-- http://www.cnn.com/2007/TECH/03/20/obit.backus.ap/index.html

-- http://en.wikipedia.org/wiki/John_Backus


Reply to this comment
Start a New Comment Thread Top

Account

Register Register | Login Log in
user:
pass: