1SRC Podcast TwentyEight

[Jeff Kirvin]

FullPower extends battery life on T3s (but be careful). PalmOne is deliberately making it difficult to code for their devices. Stand by your developers and let PalmOne know that this behavior WILL NOT STAND! [details (http://www.1src.com/?m=show&id=1089)]

[GadgetGuru]

I think few will take the time to snail mail PalmOne, and that probably wouldn't even get past the letter sorting division... However, if some knows the email of Ed Colligan and the rest of the executives at PalmOne, then I am sure, many of us will email them ASAP.

Haven't heard the podcast yet...still downloading off my slow ISP line. Gotta get broadband soon...

[Jeff Kirvin]

I'm still working on getting Colligan's email address...

[smoothjordan]

Jeff Jeff Jeff,
I'm sadly dissapointed by this podcast. And yet, it shows that your really not a palm apologist. Firstly, this podcast is almost contrary to every other podcast I've heard from you. It sounds like the last thread completely morphed you into a new person. I wish you would have waited until you weren't so mad (even though you did, maybe longer) to do this podcast, because it sounds like you are really pissed at palm and that you are giving up on them.
J-Shep

[archangel]

I think its the best podcast you have ever done. It shows how much you like the OS and how much you want Palmone to make it better so it can continue to survive. It was done in a very fair and even handed way. I'll certainly try and help. Hopefully Palmone will listen. Great job IMO.

[Jeff Kirvin]

SmoothJordan, I call it like I see it. PalmOne isn't doing right by their developers, and I'm going to call them on it. Would you prefer we ignore this?

I am mad. I'm willing to fight for the Palm platform. Are you?

[smoothjordan]

No, I guess you have a valid point. I guess you should be mad, but I don't use that many 3rd party apps, so it's not really a bother for me. But since you do so much for the site, I will write a letter to Palm complaining about this. Every little bit helps!

[Cyker]

FullPower:
This is uber cool! I must admit it's not something I'd use (Paranoid + TH55 owner :p), but as you say, it brings the rather conservative T|3 more in line with almost every other OS5 PDA in terms of battery life ;)
Mobile-Stream are just going from strength to strength with the low-level code hacking! They're starting to enroach on dmitry, Ivan and CliePet's territory! :p :D


Horrific rants:
This is why we love you Jeff! :D


Crap code:
I don't think they are doing it 'deliberately' - The problem (IMHO) is that Palm1's coders just don't have the right skillset: They code like Desktop Application programmers, not embedded device programmers - The desire to keep their docs secret, the sloppy coding - It all reeks of a desktop RAD coder's programming style.
You can get away with being sloppy on the Desktop because the OS will catch you, but when you code like this in the OS, *espescially* core functions, you go to /dev/null...

I'm not against OS mods per se - After all, Sony's biggest contributions to PalmOS were their OS extensions. These did make the Clie line slightly less stable than equivalent Palms, but their additions *were* 'additions to', *not* the 'changing of' any core functions. If a program still used the older functions, they were still there to be used and worked as expected.

Now, Sony are notorious for having desperately bad end-user service, but *even they* set up a developer website *specifically* for the Clie's extensions/mods, and their APIs up for public use (Not all of them, damnit, but some is better than none...), and even had a forum for Clie devs to swap info.

If Palm are going to hack the OS - They NEED to LEARN from what Sony did - Make it open, COMMUNICATE with the developers and DON'T go breaking compatibility left, right and centre just because it seems easier to do so!
If they don't, then for the love of all things PalmOS, leave the OS stuff to PalmSource - Ask THEM to make any mods/extensions if you need them, but let the experienced OS hackers do the core OS stuff!


DIA:
This is odd - There used to be two APIs for this; Sony's which (Allegedly) works right up from the HiRes+ OS4 devices to the TH55, and Palm's, which was only recently introduced (In the T|3 I believe?).
They weren't compatible, but we were used to the whole dual Sony/Palm standards by then.
I had assumed however, that *both* APIs were fairly stable across versions... the report to the contrary is disturbing...


Cobolt:
Totally agree with you - From what I know of it, Cobolt would have been sooooo much more suited to the hardware configuration of the NVFS devices.
As you say, PalmOS was never supposed to have all these layers built onto it - It is a lightweight OS!
This is, in a way, the source of all my whiney rants, and why I think the rumor of the T|5 originally intending to be a Cobolt device has at least a drop of truth.
That said, I personally agree more with what you said in earlier days: PalmOS is better than Cobolt for *us*. But I also agree with you that the agonising cross-migration to Cobolt would be preferable to the hideous flaying of the PalmOS codebase that's currently happening.


'Influencers':
Hehe, sounds like some kinda super secret agents :P
And you forgot mikep! *ducks* (Bahaha! Man, where is mikep these days? I miss his trolling ways :D)
Technically this is marketting - Marketting by proxy and word of mouth.
But you are right; This is possibly *the* most powerful form of marketting.
As I said in another thread, Word of Mouth from the community was the BIGGEST reason Palm managed to get so big in the first place - Even 'tho Psion were the big-boys in those days, Palms just rolled over them because they were attracting 'influencers' (zealots/evanglists I call 'em ;)) like flies to shi...uh... stink ;)


Final Thoughts:
Will making the core of PalmOS Linux allow the enthusiast hackers to make it 'better'? Or will we not be able to change anything at all...?

In my humble and personal, but oft-stated opinion, this whole thing started when Palm split into two companies. See, PalmSource *know* what they are doing. They are seasoned OS hackers and they understand the importance of being Open - Documenting interfaces is one of the most important things that go into OS design. Microsoft do it, Linuxen do it, Sun do it.

Palm1 were mostly comprised of hardware guys, designers, with only a small team left over to integrate the OS into the device. This integration team wouldn't have been any where as experienced as the PalmSource programmers, but like Sony, Tapwave, Handera and Handspring, Palm1 would have found themselves needing to grow their developer team since they no longer had the direct support of the team who are now PalmSource to add their custom mods.

However, because they were used to stuff 'just working', I think they didn't understand the importance of documenting any changes to core code. Worse yet, their previous ties with PalmSource meant they were insulated from the process of modifying core code - They don't seem to understand the amount of respect you HAVE to maintain when even *thinking* about changing something in core code.

It seems like Palm1 looked at what Sony did and thought, "Hey, we can do that!" without actually digging down and seeing how Sony supported all the changes they made or allowed for backwards/alternate compatibility.

Microsoft is the biggest example of this - One of the most impressive things about Windows is that it manages to be very backwards compatible, espescially considering it's age - Microsoft still support a lot of ancient interfaces in their OS code from as far back as Windows 3.0 because they *know* that changing or removing them will potentially have devestating effects elsewhere (Quintessential Butterfly->Tornado scenario).

Contast this with Linux - Linus freely breaks kernel interfaces with almost every revision, which is why closed-source binary drivers are impossible to write for Linux - As soon as you update the kernel they stop working.
For Linux however, this is a Good Thing, because only by providing the Source Code can you guarantee compatibility.
In addition, with Linux, although documentation is relatively poor, the fact that the source code is available to all and sundry means it BECOMES the documentation, and in a way it works out even better because you can SEE with your own eyes what happens.
As an aside, this is why I am slightly hopeful that the PalmOS/Linux initiative will work. (Assuming it is for this real purpose and not just bandwagon-jumping like my skeptical brain reckons ;))
I really think my supposition is not far off the mark - Look at every Palm made before the split: Very stable, robust standardised interfaces. You could synch an m515 with a IIIc's desktop!

After the split, we had a radical change - The T|3!
And it was a good thing, very good. We thought Palm might actually challenge the beloved Clie line! But with the change, I noticed lots of little things, the details, being left out or ignored. Possibly by the inexperienced coders who were once just doing the value-added software and systems integration, and now found themselves fully in charge of full-blown OS customisation!

Your 'campaign', Jeff, is a good idea - If this lack of support really is down to ignorance rather than malice, then maybe there is some hope if we can get them to pay attention.
Heck, if we can get them to add staff to their team, specifically to liase with developers in the way that Microsoft and Sony do/did, then Palm's future software support will almost be assured!

[Surur]

I think its a bit paranoid to think Palm is trying to kill of third party developers. As you said yourself, never attribute to malice what can as easily be explained by incompetence. I suspect behind all the changes P1 made on the cheap is a pile of undocumented spaghetti code. Its not that they don't want to reveal the API, they cant. If Palm does not bring a software platform to the phone networks, what do they bring?

I still wonder why they did not bite the bullet 18 months ago and just transition to Cobalt. Could it really have been worse?

Surur

[jjesusfreak01]

SmoothJordan, I call it like I see it. PalmOne isn't doing right by their developers, and I'm going to call them on it. Would you prefer we ignore this?

I am mad. I'm willing to fight for the Palm platform. Are you?
Im willing to fight, I just don't think I have any stamps. I have not used the USPS since...well I am not sure. Once you get a good email address for use to pester, we can get some of our shady friends to lend us a bot army for a few hours to get Palm. (legal note: Im kidding about the bot army thing)

As for correctly written programs Jeff, regardless of whether or not the developers are writing their programs perfectly, I personally believe every good operating system should be able to handle code that isnt perfect. I understand that the basic code must be correct, but there are different ways to write a program, and any fairly well written program should have a good chance of running as long as it doesnt have major flaws.

To be completely fair about the DIA thing, Sony did it first, I believe, and Palm followed. The problem is that they set a trend. Palm was not going to use Sony's system, and Palm was likely trying to set a standard with their own devices, especially since they were adding rotation.

You should contact Mobile Stream about this. They seem to know alot about reverse engineering DIA and Graffiti systems, and may be able to release some information that Palm will not.

Treo 650: Remember, we know Palm is sitting in the carrier's pocket. They never released the WiFi drivers, though the hacked ones work perfectly, meaning the carriers are to blame.

FrankenGarnet?

Jeff, you really want some PalmOne devices for free, I see! You didnt read my reply on TamsPalm. I said Palm should hire a bunch of people to take new devices and test them, show them around, get on the board, and get paid for it. Besides just having someone look at the boards, they need to give the forum feedback back to Palm.

Good Podcast Jeff


PS: They just need to hold up for four more years. Then, Dmitry graduates college, and becomes Palm's lead programmer, and soon after, Palm will be buying Microsoft.

[Cyker]

Cobolt will be a nightmare - It's a complete paradgim shift towards a more PPC-style way of doing things.

It has to load things off secondary storage into RAM the way a PPC, Mac, or PC does - It doesn't use the run-in-place system of PalmOS. This is why it has a lot of PPC-like features, like multi-taking and the ability to store arbitary files in 'RAM' (Because they really goto the secondary storage FlashRAM).
Good things about it is that it will let us run things off VFS directly without the copy-to-RAM or MSMount-style workarounds, because the copy-to-RAM aspect will be part of the design, not a bodge.

Cobolt is NOT PalmOS compatible in any way. In fact, to run PalmOS programs, they basically hacked out the OS4 PalmOS emulator that is in all OS5 systems and modified it to run on Cobolt.
Running PalmOS apps on Cobolt will be far more akin to using that PalmOS emulator you (Surur) linked to in another thread.

This is why I refer to Cobolt and PalmOS as different things - Cobolt is not PalmOS6, it's completely seperate. The only things they really have in common is the interface and basic coding style, but this is really the best thing about PalmOS ;)

I suspect they had difficulty porting POSE into Cobolt; Things like HotSync passthrough, and general hardware interface and control for things with no API will be difficult.
At the end of the day, a lot of non-trivial PalmOS programs will likely break when run on Cobolt unless they have yet more workarounds coded into them.

This is a fear of migrating to Cobolt 'tho; Apps will really need to be custom written to it to be worth anything, otherwise you end up with apps which have so many hacks and workarounds in them they make up more of the code than the actual program!! :eek: (Windows and Linux programs are a lot like that :p)

If they release a Cobolt device, there is a danger they will end up competing with themselves because I suspect a great deal of existing PalmOS programs, esp. ones that use anything other than standard Forms and APIs (i.e. all games (Excluding things like Puzzle and Giraffe :D), and most applications) will not work properly on Cobolt's POSE.


Edit: Damnit JJF! Type slower! I was replying to Surur!! :p :D
Oh, and ROFL at the P.S. ;) Hehee, we need to get Sharky, CliePet and Dmitry infiltrated into Palm :D

[strider_mt2k]

I think its a bit paranoid to think Palm is trying to kill of third party developers. As you said yourself, never attribute to malice what can as easily be explained by incompetence. I suspect behind all the changes P1 made on the cheap is a pile of undocumented spaghetti code. Its not that they don't want to reveal the API, they cant. If Palm does not bring a software platform to the phone networks, what do they bring?

I still wonder why they did not bite the bullet 18 months ago and just transition to Cobalt. Could it really have been worse?

Surur

I'm with you there.
As Jeff and I have discussed in the past when I said the same thing.
I just don't think they have the smarts for this to be malice.

I seem to remember there was some heat over at P1 over what's-his-name spending too much time on the PDA side and not enough on the smartphone side.

Could all of this be a last-minute rush-out to clear the designers' tables for the next smartphone?

I dunno, and frankly I'm beginning to not care.
These devices won't be what they were, and what they're becoming I don't know if I want.

I'm sorry to see you so upset by this Jeff, but at least progress is being made.
PalmOne has an incredible resource in the form of the developers and community.
They can either profit by this, or be destroyed.

(Yes this is your official Strider-Brand Podcast Star Wars reference, please make a note of it.) :)


What's telling for ME is that I listened to this podcast on my new PSP.
Those funds could EASILY have gone towards a POS handheld upgrade, but unfortunately there STILL is no true upgrade from my TH55.
So, ironically, who got my money again?
Sony.

Go figure.

[Jeff Kirvin]

So what you're saying, Cyker, is that Cobalt works a lot like Garnet (emulating OS4 with PACE) with NVFS (copying apps from storage to RAM).

With one big difference: Cobalt was designed from the ground up to work that way.

[jjesusfreak01]

Edit: Damnit JJF! Type slower! I was replying to Surur!!
Oh, and ROFL at the P.S. Hehee, we need to get Sharky, CliePet and Dmitry infiltrated into Palm Im sorry, I just posted when I finished listening to the Podcast, and had my response written up.

[Surur]

Cyker - You can only go so far by keeping it simple. When you want to do complex things (such as e.g. a dynamic input area vs the old physical one) things get complicated. At some point it gets simpler to start over with more complexity and flexibility.

Surur

[applejosh]

It has to load things off secondary storage into RAM the way a PPC, Mac, or PC does - It doesn't use the run-in-place system of PalmOS.

You mean like copying stuff off a partition to RAM in order to run it? Seems like we're trying to do that now with NVFS, in a slightly hackish manner.

Cobolt is NOT PalmOS compatible in any way. In fact, to run PalmOS programs, they basically hacked out the OS4 PalmOS emulator that is in all OS5 systems and modified it to run on Cobolt.

I thought PalmOS 5 used an emulator (PACE) to run Palm programs on the arm architecture. So how is using an emulator designed/modified for OS6 different than using the emulator deisgned for OS5?

Whatever the case, the Palm companies need to get their collective heads out of their collective derrieres if they want to exist more than 2 years from now.

[Maticek]

So how many have written to P1?

[strider_mt2k]

I registered over at the Palm site, still waiting for an e-mail.
Not that this is related, but I'm locked out of every forum except General over at Brighthand.
I've posted about it (which is the only recourse) but no answer as of yet.

I don't want to drag Clie into this, as I know there is a ridiculous "Ford-Chevy" thing with many Palm users, but I can't help it.

Isn't this very like Sony not releasing what THEY did to change Palm OS for THEIR handhelds?
With Sony it was malice, but we knew that going in. ;)

[Gekko]

The Conversion of Saint Kirvin
Gekko @ 6/15/2005 4:06:46 PM #
http://www.palminfocenter.com/

So now the same guy who stubbornly and continuosly blamed sloppy developer programming and customer user error for the "few and rare" bugs/crashes of the T5 and LD has suddenly had an epiphany? And he FINALLY acknowledges and accepts as his own through revisionism what the so-called "WINGNUTS" have been preaching for the last year - that OS 5.4+ is really a cobbled-together buggy/unstable "FrankenGarnet"?

Congratulations, Kirvin. You're ONE YEAR LATE.

[Cyker]

JJF: No need to apologise, I was just kidding ;) (Hence the smileys :p :D)

So what you're saying, Cyker, is that Cobalt works a lot like Garnet (emulating OS4 with PACE) with NVFS (copying apps from storage to RAM).

With one big difference: Cobalt was designed from the ground up to work that way.

EXACTLY!!!!! :D
Well, functionally anyway. Obviously the two will be very different in back-end code (Behind the scenes, so to speak), but in effect, my PoV is as you so succinctly put it ;)


Cyker - You can only go so far by keeping it simple. When you want to do complex things (such as e.g. a dynamic input area vs the old physical one) things get complicated. At some point it gets simpler to start over with more complexity and flexibility.
One thing Cobolt was supposed to add was a *far* better defined set of interfaces, and a from-the-ground-up modular design.
I don't know what happened to that 'tho, because recent data seems to suggest that Cobolt is even harder to customise than PalmOS!! :confused:

I thought PalmOS 5 used an emulator (PACE) to run Palm programs on the arm architecture. So how is using an emulator designed/modified for OS6 [Cobolt] different than using the emulator designed for OS5?
Ahh! A good question! (I had to re-read that a few times to get your meaning).

I've been thinking about this, and now I'm not so sure.
See, my original thinking was that OS5's PACE works the way WINE does (Code/Function translation - Fast, but less compatible with different systems), and that Cobolt's PACE would be more like UAE or Bochs (Full blown emulation - Slow but more compatible with different systems).

But thinking about it harder, I can't think of a good reason why they can't just continue to do it in a WINE-style way in Cobolt and map PalmOS PACE functions to equivalent Cobolt functions the way they may OS4 functions to ARM functions OS5.

Oh, ooh! Actually I can - Multitasking!
PalmOS is a single-tasking system; Cobolt isn't.
Imagine if you do a WiFi HotSync while downloading something in some hypothetical Cobolt web-browser.

In OS5, PACE has 100% control of the hardware so it could just do it, but in Cobolt it would have to share that connection somehow, either that or it has to suspend all Cobolt tasks and drop to a single-tasking mode when running a PalmOS program, and only switch back to multi-tasking mode once you exit.
This would be the easiest way to do it, but it's kinda sucky...

To maintain the multi-tasking system, they would have to do some heavy re-engineering of PACE to integrate it with Cobolt so that they can somehow share hardware and resources without blowing up...

This is why that PPC PalmOS emulator Surur told us about supports very little aside from basic Forms and display handling at the moment; They will have the same problems the Cobolt teams is facing in managing multiple access to hardware resources...

[Cyker]

I registered over at the Palm site, still waiting for an e-mail.
Not that this is related, but I'm locked out of every forum except General over at Brighthand.
I've posted about it (which is the only recourse) but no answer as of yet.

I don't want to drag Clie into this, as I know there is a ridiculous "Ford-Chevy" thing with many Palm users, but I can't help it.

Isn't this very like Sony not releasing what THEY did to change Palm OS for THEIR handhelds?
With Sony it was malice, but we knew that going in. ;)

It's ironic isn't it? ;)
Although, as I said, Sony DID have developer relations.
Palm's was far better, but when they split into PalmSrc and Palm1, PalmSrc kept their excellent developer relations and Palm1 didn't have any. At All.


The Conversion of Saint Kirvin
Gekko @ 6/15/2005 4:06:46 PM #
http://www.palminfocenter.com/

So now the same guy who stubbornly and continuosly blamed sloppy developer programming and customer user error for the "few and rare" bugs/crashes of the T5 and LD has suddenly had an epiphany? And he FINALLY acknowledges and accepts as his own through revisionism what the so-called "WINGNUTS" have been preaching for the last year - that OS 5.4+ is really a cobbled-together buggy/unstable "FrankenGarnet"?

Congratulations, Kirvin. You're ONE YEAR LATE.

Can we stop picking on Jeff just for having and expressing an opinion please?
This "Ner ner told you so" crap is a waste of zots and electrons.

You're a decent guy Gecko; Don't go all mikep on us... :eek:

Besides, he is apparently some sort of ex-military Texan, and according to British TV this means he has a closet full of guns and drives a tank. :eek: :D (j/k ;))

[applejosh]

Well, I thought the so called multi-tasking in Cobalt was pathetic (by desktop multitasking standards), too, so I don't know if there will be that much of a difference. But I really don't know (nor care that much at this point since a Cobalt device coming to being is growing increasingly unlikely). When they port everything to Linux, then things will probably change greatly.

Frankly, I'd just like to see PalmOne start communicating. It doesn't look like they want to do that (and probably, as suggested before, to hide how bad their code really is). I'd also like them to start fixing the problems that devices have in a more timely fashion. The T5 update (if it does appear in July) will have taken nearly a year. That's just too long, no matter how much of a pain in the rear it is to write and publish it. The longer you go with a buggy OS, the less likely you'll hang on to customers. (Unless you're fortunate to have a virtual monopoly with the initials MS.)

[jjesusfreak01]

Well, I thought the so called multi-tasking in Cobalt was pathetic (by desktop multitasking standards), too, so I don't know if there will be that much of a difference. But I really don't know (nor care that much at this point since a Cobalt device coming to being is growing increasingly unlikely). When they port everything to Linux, then things will probably change greatly.

Frankly, I'd just like to see PalmOne start communicating. It doesn't look like they want to do that (and probably, as suggested before, to hide how bad their code really is). I'd also like them to start fixing the problems that devices have in a more timely fashion. The T5 update (if it does appear in July) will have taken nearly a year. That's just too long, no matter how much of a pain in the rear it is to write and publish it. The longer you go with a buggy OS, the less likely you'll hang on to customers. (Unless you're fortunate to have a virtual monopoly with the initials MS.)
You know the MT is alot better. Were you expecting Cobalt to suddenly have the features of a desktop OS?

As for the update, I agree. I might understand if programming was like a second job for them, but making handheld devices is a full time job for these people.

[Gekko]

p1 in public: "Garnet accomplishes what we need to at this time. We are looking at Cobalt and trying to determine when it makes sense."

p1 in private: "Cobalt is a buggy, unstable piece of garbage. You guys at PSRC better fire Nagel and create something decent or we will have no choice but to switch to WM2005. And in the meantime - now were forced to cobble together FrankenGarnet into a serviceable modern OS. May God help us all."

PSRC in public: "Cobalt is not necessarily betterr than Garnet. It's just different. There are two distinct product lines."

PSRC in private: "We're F***ED."

[Cyker]

Well, I thought the so called multi-tasking in Cobalt was pathetic (by desktop multitasking standards), too, so I don't know if there will be that much of a difference. But I really don't know (nor care that much at this point since a Cobalt device coming to being is growing increasingly unlikely). When they port everything to Linux, then things will probably change greatly.

Frankly, I'd just like to see PalmOne start communicating. It doesn't look like they want to do that (and probably, as suggested before, to hide how bad their code really is). I'd also like them to start fixing the problems that devices have in a more timely fashion. The T5 update (if it does appear in July) will have taken nearly a year. That's just too long, no matter how much of a pain in the rear it is to write and publish it. The longer you go with a buggy OS, the less likely you'll hang on to customers. (Unless you're fortunate to have a virtual monopoly with the initials MS.)

I dunno... On the one hand Cobolt's multitasking is going to have very limited uses (Background network transfers, music playing, calculations?), but if the alternative is PPC-style "Where the f**k did all my CPU and RAM go?!" multi-tasking, I'd much rather stick with Cobolt's...
Come to think of it, you can't even access more than one program's interface on a PPC so it's 'fat-state' multi-tasking is wasted.
Yes, the more I consider it, Cobolt's half/thin-state multitasking is much more suited to a Palmtop... at least in the form I read about...!


I don't know where the Palm roadmap is headed. At the moment it looks like we'll end up with three mostly incompatible Palm operating systems:

PalmOS5
Cobolt
PalmOSx/Linux

With our current Moto68k-based PalmOS being the lowest common denominator that can run on all three.
(I bet ARMlets won't work on Cobolt or Linux, and obviously Cobolt apps won't work on PalmOS or Linux. I bet Linux will introduce YET ANOTHER system that won't work on the other two.)

It's freakin' mess...

[jjesusfreak01]

You are right, ARMlets will likely not work out of the box. At this point, ARMlets are just small pieces of ARM code backboned by 68k. In Cobalt, programmers have to rewrite everything to Protein. OS5 in itself is a FrankenCobalt. It is a ARM OS that runs on ARM processors, but runs 68k programs almost exclusively. It can handle ARM code, but not natively through Palm APIs. Palm running on Linux will essentially be a large emulated OS. Cobalt and PACE apps will work.

More specifically, the PACE on Cobalt will probably not be able to touch ARM code. All ARM for Cobalt will probably have to be written in Protein APIs and such. Correct me if Im wrong, but the PS website mentions nothing about Cobalts PACE handling ARM, and it would be a pain to make it work that way.

For a little more info, you can click here (http://www.palmos.com/dev/support/docs/protein_books/Porting_Apps_to_Cobalt/PACE.html#1004932) for some info on what PACE functions are no longer supported in Cobalt.

[applejosh]

You know the MT is alot better. Were you expecting Cobalt to suddenly have the features of a desktop OS?

How do I know the MT is better? I haven't seen a Cobalt device. What I was trying to impart is that I don't think it will be that different of an emulator than what is currently in OS5. It doesn't matter. As pointed out by others, true multi-tasking on a palmsized device is overkill. The screen can only display one app at a time, so if the app being put into the background will just pop back to its current state (as apps under OS5 are supposed to do from what I understand), it will be enough for me. I really don't need to take up all the resources running multiple apps simultaneously (especially if the hardware vendor is limiting the true RAM in devices anyway).

[Surur]

Because WM does true multi-tasking, one CAN actually have two arbitrary programs displaying simultaneously, and there is software available that will let you split the screen e.g refer to two spreadsheets at the same time, or easy copy multiple pieces of text. A more regularly seen example is multi-tabbed browsing.

Multi-tasking makes things easier for programmers, and makes devices more stable. What else are you going to use all that xscale horsepower for?

Surur

[tungsten t5]

Jeff, I completedly agree with you on this podcast.

[tungsten t5]

Jeff, I do agree with you and becuase of this I'm going to post about this on Tamspalm. If you would like to write something for me to post on Tamspalm please do so. I think the furture of palm is non if something doesn't happen. even though I use and Love My Dell axim x50v (wich will get an upgrade to windows mobile 5.0) Palm was my first choice. I will always be a palm user.

[Surur]

even though I use and Love My Dell axim x50v ... ... I will always be a palm user.

Some Slashdot humour:

It is now official - Gartner has confirmed: Palm OS is dying

Yet another crippling bombshell hit the beleaguered Palm community when recently IDC confirmed that Palm OS accounts for less than 32%percent of all handhelds. Coming on the heels of the latest Canalys survey which plainly states that Palm has lost more market share, this news serves to reinforce what we've known all along. Palm OS is collapsing in complete disarray, as fittingly exemplified by falling dead last [http://www.hhcmag.com/] in the recent HandHeld Computing Road Warrior comprehensive usability test.

You don't need to be a Kreskin [amazingkreskin.com] to predict Palm's future. The hand writing is on the wall: Palm OS faces a bleak future. In fact there won't be any future at all for Palm OS because Palm OS is dying. Things are looking very bad for Palm OS. As many of us are already aware, Palm OS continues to lose market share. Red ink flows like a river of blood. PalmSource is the most endangered of them all, having lost 93% of its core developers.

Let's keep to the facts and look at the numbers.

PalmSource leader Micheal Mace states that there are 275 000 Palm OS developers. How many developers of Windows Mobile are there? Let's see. The number of Palm user versus Palm Developer posts on User Forums is roughly in ratio of 5 to 1. Therefore there are about 275 000/5 = 55000 Palm OS Developers. Palm OS Developer posts on Forum are about half of the volume of Windows Mobile posts. Therefore there are about 7000 Developers of Palm OS. A recent article put PalmOne at about 80 percent of the Palm OS market. Therefore there are (7000+140000+70000)*4 = 868 000 Palm OS users. This is consistent with the number of Palm OS Forum posts.

Due to the troubles at Sunnyvale, California, abysmal sales and so on, Sony went out of the Palm handheld business and the role was taken over by Samsung who sell another troubled OS. Now Samsung Palm Development is also dead, its corpse turned over to yet another charnel house.

All major surveys show that Palm OS has steadily declined in market share. Palm OS is very sick and its long term survival prospects are very dim. If Palm OS is to survive at all it will be among OS hobbyist dabblers. Palm OS continues to decay. Nothing short of a miracle could save it at this point in time. For all practical purposes, Palm OS is dead.

Fact: Palm OS is dead

Surur

[jjesusfreak01]

Jeff, I do agree with you and becuase of this I'm going to post about this on Tamspalm. If you would like to write something for me to post on Tamspalm please do so. I think the furture of palm is non if something doesn't happen. even though I use and Love My Dell axim x50v (wich will get an upgrade to windows mobile 5.0) Palm was my first choice. I will always be a palm user.
Thats why Tam hasnt kicked you off yet!! Good to hear anyway.

[applejosh]

Because WM does true multi-tasking, one CAN actually have two arbitrary programs displaying simultaneously, and there is software available that will let you split the screen e.g refer to two spreadsheets at the same time, or easy copy multiple pieces of text. A more regularly seen example is multi-tabbed browsing.

Multi-tasking makes things easier for programmers, and makes devices more stable. What else are you going to use all that xscale horsepower for?

Surur

I still think multitasking on a handheld device with a 3.5" screen is pushing the bounds of usefulness. I like being able to switch between apps (say to copy and paste), but displaying multiple apps simultaneously on the same tiny screen makes my eyes hurt just thinking about it.

Maybe someday with some sort of bionic eye implant, I could tolerate it. :rolleyes:

Fact: Palm OS is dead

Not yet. Until I can get my iPAQ (or maybe future Dell Axim X50v) to set off my alarms first thing in the morning, a Palm will always have a place. I'm also waiting to see how well the task alarm thing works out in WM5. I'd also like to see a better calendar replacement. I still don't think PI or AF are up to par with DateBk. Just my opinion. I'm sure you'll disagree. :cool:

[tungsten t5]

Thats why Tam hasnt kicked you off yet!! Good to hear anyway.

Yet???? do you know something I don't know?

[jjesusfreak01]

Yet???? do you know something I don't know?
We just thought you were abandoning the POS, due to your PPC rants a while ago.

[Jeff Kirvin]

The Conversion of Saint Kirvin
Gekko @ 6/15/2005 4:06:46 PM #
http://www.palminfocenter.com/

So now the same guy who stubbornly and continuosly blamed sloppy developer programming and customer user error for the "few and rare" bugs/crashes of the T5 and LD has suddenly had an epiphany? And he FINALLY acknowledges and accepts as his own through revisionism what the so-called "WINGNUTS" have been preaching for the last year - that OS 5.4+ is really a cobbled-together buggy/unstable "FrankenGarnet"?

Congratulations, Kirvin. You're ONE YEAR LATE.

And?

I really don't understand these people that act like my taking on this cause is somehow a reversal. I call it like I see it. Always have, always will. New information came into my possession, and I adjusted accordingly. It's what intelligent people do.

Standing by what you previously said in spite of new contradictory information is called being "resolute" and frankly it's just stupid.

[Jeff Kirvin]

I think its a bit paranoid to think Palm is trying to kill of third party developers. As you said yourself, never attribute to malice what can as easily be explained by incompetence. I suspect behind all the changes P1 made on the cheap is a pile of undocumented spaghetti code. Its not that they don't want to reveal the API, they cant. If Palm does not bring a software platform to the phone networks, what do they bring?

I still wonder why they did not bite the bullet 18 months ago and just transition to Cobalt. Could it really have been worse?

As much as it kills me to agree with Surur (this should prove how devoted I am to good ideas!), I have to say that the more I think of this, the better is sounds.

I'm a firm believer in Occam's Razor. Given several possible explanations, the most simplest one is probably true. And the simplest explanation here is that PalmOne can't reveal an API to developers because they no longer have one. What if the reason they tinker with the DIA on each new device is because they have to in order to work around other bugs? Garnet might be so complex and Rube Goldberg that even PalmOne doesn't fully understand what it's doing. This would also explain the presence of bugs that should have been caught in testing. What if they were caught, but no one could figure out how to fix them. "Eh, it's not a show-stopper, ship it anyway."

Oh, I'm not feeling too good about this. Maybe I'll go see Batman again...

[Jeff Kirvin]

Jeff, I do agree with you and becuase of this I'm going to post about this on Tamspalm. If you would like to write something for me to post on Tamspalm please do so. I think the furture of palm is non if something doesn't happen. even though I use and Love My Dell axim x50v (wich will get an upgrade to windows mobile 5.0) Palm was my first choice. I will always be a palm user.

Dude, I saw you on AximSite threating to pitch your X50v. Both platforms have their issues.

[archangel]

Not yet. Until I can get my iPAQ (or maybe future Dell Axim X50v) to set off my alarms first thing in the morning, a Palm will always have a place.

They still haven't fixed the alarm problem?!?

Man, if I can't even rely on the thing to set a proper alarm how can I count on a PPC for anything.

I use Palmary Clock on my Zod as my alarm clock. I absolutely have to rely on my PDA to work as an alarm clock. I have a back up alarm set on my N-Gage, but that one doesn't always wake me up.

[Jeff Kirvin]

They still haven't fixed the alarm problem?!?

Man, if I can't even rely on the thing to set a proper alarm how can I count on a PPC for anything.

I use Palmary Clock on my Zod as my alarm clock. I absolutely have to rely on my PDA to work as an alarm clock. I have a back up alarm set on my N-Gage, but that one doesn't always wake me up.

That was one reason I left the platform.

[archangel]

That was one reason I left the platform.

I'm shocked this wasn't fixed immediately. Say what you will about Palm, but I have never had a bug on any Palm PDA I have owned as inexcusable as that. What minor bugs I had on my Tungsten and T3 where patched very quickly.

An alarm bug that is left unfixed this long is an absolute joke. I thought not allowing you to close the programs easily was stupid, but they look like geniuses compared to letting an alarm bug go in a device designed to keep you on freaking schedule. I'm going to assume that WM5 finally fixes this, but you know what happens when you assume.

[applejosh]

In defense of the PPC platform (which you won't see me do often), there have been quite a few developers releasing/incorporating solutions to the alarm problem. They do help, but still nothing 100% (or close enough). Just for kicks, here's some. Scary Bear software put out a clear notify program which helps (free). Pocket Plus has a feature included with it. Tweaks2k2 has another approach. MemMaid will manually clear out the duplicates (which is thought to be part of the problem, although it's not automatic AFAIK). But yeah, even if I'm in a PPC mode of thinking, I still make sure my Palm device has the up-to-date calendar and tasks with all the alarms.

[tungsten t5]

ok, creepy, jeff first why were you on aximsite? second, how do you know my aximsite screenname? who, what, when, where, why?

and in my defence I was used to palm os and EVERYthing getting deleted on a hard reset but on PPC's only the ram not the rom gets deleted. So that was my fault not the operating systems.

[smoothjordan]

There are 3 reasons why I don't own a PPC anymore, and they have all been discussed already -
1. The alarms
2. "X" doesn't really close a program, it's kind of a pain to go and close all the programs you opened. 3rd party software does this, but why pay? The OS should have this built in.
3. I don't like Pocket Outlook. IMO, The PalmOS standard datebook is more powerful and much more intuitive. Plus, it's much easier to use.

In Pocket PC's defense, feature wise, they are much cheaper to buy then a Palm OS pda. My Tungsten T|5 is very similar to the x50 low, and the low is much cheaper. As much as I love palm, unless Palm Linux is a huge leap from WM 5.0, the Tungsten T|5 will be my last Palm OS handheld. It hurts to say it, but ever since Palm merged with Handspring, and Sony left the market, I haven't seen a HUGE improvement in Palm OS handhelds, while PPC's are progressing rapidly. Damn Palm, can't make me happy! (Even though I really do love my Tungsten T|5!)

[Surur]

There are 3 reasons why I don't own a PPC anymore, and they have all been discussed already -
1. The alarms
2. "X" doesn't really close a program, it's kind of a pain to go and close all the programs you opened. 3rd party software does this, but why pay? The OS should have this built in.
3. I don't like Pocket Outlook. IMO, The PalmOS standard datebook is more powerful and much more intuitive. Plus, it's much easier to use.

1) I thought the alarm issue was sorted in wm2003se? Most people seem to say so.
2) Quality third party software is the backbone of the PPC community. Its true that much of this software is to fix the deficiencies of the OS, but at least MS doesn't get in the way of the software makers, or usurp their market. Its completely analogous to the desktop market. Thats why I think most PPC users DO add third party software, and according to handandgo they are happier with their devices and the software they purchase. PPC''s by themselves is a platform, not a solution.
3) See above. MS supports their super PIM produces. HP is bundling PI with the HP Hx4705. It was on the CD of my Loox 600.

I guess to fully get a PPC up and running to my satisfaction would take about $60-$100 ( PI ($30) Resco explorer ($20) Pocketplus ($10) Pocketbreeze ($10) ) Once you have done this however you have an amazingly capable platform thats also very stable. And you only have to do this once.

You can add enhancements to a pocketpc, but you can not add better programming at any price to e.g. the LifeDrive.

Surur

[smoothjordan]

Valid points Surur, which is why I don't think Palm will be getting anymore money from me for a new device, as PPC's now (especially with the VGA screens!) are going to be much better. Don't get me wrong, I love both OS's, but the devices seem to get in the way. Let's hope that HP starts making Palm OS PDA's (read something about this on Brighthand) as they have gotten the PPC Handheld's perfected with their IPAQ line, especially the hx4700. I would have gotten one except for the high price tag.

[gfunkmagic]

Okay, let me just say that Jeff had some decent points here, but there's alot here I think he's just plain wrong on (many of you will disagree with me). Before I go into why, JV was right about one thing....

>>> IT"S ALL ABOUT THE TREO!!!!!!!!!!!

There was alot of references to 'frankengarnet' in the podcast and speculations on why PalmOne has up to this point refused to release a Cobalt based device. One absolutley absurd theory put forth by Mr. Kirvin here was that PalmOne was some how trying to purposefully make Garnet unstable in order to dissuade users from intalling third party apps on the Treo which would somehow reduce customer service calls and thereby support costs for the carriers! :rolleyes: That is the biggest load of garbage I've ever heard...sorry, but that is completely illogical and insane! That is not to say that I disagree with the point that alot of newest versions of Garnet released by palmOne are buggy and unstable. But the reasoning behind why PalmOne has continued to maintain and push Garnet instead of Cobalt has more to do with its strained relationship with PalmSource more than anything else! In fact, if you really want to blame anybody here, then YOU SHOULD BLAME PALMSOURCE!!

Why you ask? It simple, PalmOne is trying to protect it's smartphone line, namely the Treo. The way PalmOS platform development works, there's always a give and take between PalmSource and the lisencees. Often when PalmSource gives the code to lisencees, the lisencees make customizations and optimzations to make the code work on their devices. When lisencees release the product, they often give those customizations back to PalmSource so that they can be integrated back into the next version of the platform. All you old clie users should be very familiar with this. Sony often made customizations to the palmos which were eventually added and integrated into the next iteration of palmos x.x whatever. That's just the way it has always worked on the palmos platform. In this way, the lisencees have always played an important role in the development and furthering of the platform.

Now the point here is that PalmOne and formerly Handspring have ALOT of proprietary customizations to the PalmOS to make it work on the TREO. For example, 5-way navigation API's and almost the ENTIRE 5.x CDMA software stack/layer has been entirely developed by PalmOne and NOT Palmsource. In fact, alot of the smartphone focused Garnet OS that PalmSource released a while ago was entirely based on code developed by PALMONE and incorporated in to the platform. Now from PalmSource's perspective this is great b/c it has one of it's lisencees doing all the heavy work for it. But what's in it for PalmOne? Developing things like telephony layer on smartphone device is no small matter, and what kind of financial sense does it make for PalmOne to spend all these resources to develop this code only to see it integrated by PalmSource and then give FREE to all the other lisencees like Samsung, Kyocera, Pitech, GSL, etc in the next iteration of PalmOS?!! :confused:

In short, it does not make any sense at all. The reason why PalmOne has not released a Cobalt based device is b/c they want to protect the investments into the proprieatry customizations they have made into Garnet and PalmSoure has so far not lived up to its expectations. For example, if none have you haven't noticed, there is STILL NO CDMA BASED COBALT SMARTPHONE REFERENCE DESIGN released by PalmSource!! :eek: There is a GSM reference design, but no cdma ref design! Why is that? Well b/c afaik PalmSource has still not developed the cdma telephony layer for Cobalt? Now why should PalmOne spend the vast amount of resources and money into developing this code for Cobalt if it will only be absorbed by PalmSource and then given to all the other lisencees for free?!! This is a major stumbling block here that no one seems to acknowledge!

Now you ask, how do I know all this? :confused: Well b/c that was basically the gist of the conversations I had with PalmOne execs last year at the Chicago Treo road show when they were promoting the 650 to IT professionals. They flatly stated that the lack of smartphone support in Cobalt at the time was a problem. Whether it still is a problem today, almost a year later, is another matter and I honestly couldn't tell you.

As to the second point regarding palmos development: Again why isn't the focus here on PalmSource? They are supposed to be the ones providing the IDE aren't they? I'm not a programmer, but it seems to me that it would be a great idea if Palmsource adapted .NET as a development platform? Why can't this happen? It seems to me that alot of the problems with the palmos platform today are directly related to the failure of PalmSource to market Cobalt to lisencees and it's failure to attract new lisencees (not to metion the ones it has lost)...

[Surur]

gfunkmagic:

Why is this the first time I've heard this view expressed? It explains a whole lot of things.

Didn't P1 recently say, as a hardware company, 80% of their engineers were software engineers, not hardware. You are basically saying they are protecting their IP. By holding on too tight however they may be painting themselves in a corner.

In the end they are fighting over valuable customisations to a decrepit OS. They cant expect to still be using the same OS in 3-5 years time (Garnet 5.999?) During the same period WIN CE has had 5 kernel updates already, and even windows the desktop would have had 4 major revisions.

The argument that "If it ain't broke don't fix it" does not really work, because things are broke currently. Maybe part of the solution would just be for Palm to buy Palmsource (although I'm sure they would inherit their licensing commitments)?

Surur

[gfunkmagic]

gfunkmagic:

Why is this the first time I've heard this view expressed? It explains a whole lot of things.


It's been posted widely at Treocentral...sorry


Didn't P1 recently say, as a hardware company, 80% of their engineers were software engineers, not hardware. You are basically saying they are protecting their IP. By holding on too tight however they may be painting themselves in a corner.

In the end they are fighting over valuable customisations to a decrepit OS. They cant expect to still be using the same OS in 3-5 years time (Garnet 5.999?) During the same period WIN CE has had 5 kernel updates already, and even windows the desktop would have had 4 major revisions.

The argument that "If it ain't broke don't fix it" does not really work, because things are broke currently. Maybe part of the solution would just be for Palm to buy Palmsource (although I'm sure they would inherit their licensing commitments)?

Surur


Well, the conversation I had was almost a year ago mind you, and PalmOne has recently renewed it's PalmOS lisence from Palmsource since then. Regarding that renewal, I think a certain part is very important in the language:


Under the terms of the agreement PalmSource will receive minimum royalty payments of $148.5 million, which includes $65 million for calendar years 2007 to 2009 subject to meeting certain development milestones.

Basically, IMO PalmOne has probably been waiting on PalmSource to finish the appropiate smartphone support before it commits to Cobalt. These 'developmental milstones' are probably what's hindering the adoption of Cobalt by PalmOne imo. Hopefully, the signing of the lisencing renewal means that most of these development milestone have currently been realised. Since most handheld development times scales are almost a year or longer, I think it would be deducible that 12-to-18 months ago when devices like the Lifedrive and T5 were in development, such milestones were not there and thus the use of Garnet.

However, I agree that PalmOne can't live off Garnet much longer. Oswin (GSL) is already shopping around their Cobalt smartphone, and once it and other smartphones running Cobalt appear, the Treo line will begin to look extremely dated in comparison. I think that was the whole point of the WM Treo imo. PalmOne may have simply been hedging it's bets with Cobalt, or may simply been giving Palmsoure a signal that the current situation was unacceptable. Who knows... In any event, I do believe there will eventually be a Cobalt device from PalmOne, perhaps even the next gen Treo that should arrive early 2006 imo.

[Gekko]

>but at least MS doesn't get in the way of the software makers, or usurp their market. Its completely analogous to the desktop market. Thats why I think most PPC users DO add third party software, and according to handandgo they are happier with their devices and the software they purchase. PPC''s by themselves is a platform, not a solution.

Surur - i usually agree with you but not this time. This argument of "usurping" 3rd party developers' markets doesn't fly for Palm - and it shouldn't fly for PPC. Both platforms should offer *GREAT* PIM functionality right out of the box - this is basic stuff. The 3rd party people can focus on the niche programs - we should not have to rely on them/purchase BASIC applications just to make the device function correctly - ie alarms, etc. Don't sell me a new car and tell me it comes with rims but no tires.

[Surur]

You're right, but I would still expect much improved functionality from 3rd party software than from the built-in ones. And one cant deny that if MS made a perfect out of the box device much money will be earned by third party programmers.

I'm probably making excuses from MS, but I actually believe MS has an interest in keeping PDA's much less functional that desktops. For one thing,the income from a PDA ($5-20) is much less than the $50-200 they would get from a laptop with office installed. If one person in 20 decides to use a PDA instead of a laptop they are losing money. I think the success of RIM (not palm) in the enterprise market has really given them a kick in the bottom (same as in the browser sphere) as there is a risk of making the desktop irrelevant, or at least much less important, and this is of course their real cash cow.

As usual MS will only really stir if there is competition to spur them on. Due to where Palm started however (constrained devices with little growth potential) they could never really provide it.

Surur

[applejosh]

1) I thought the alarm issue was sorted in wm2003se? Most people seem to say so.
Had an alarm not go off this morning, so I'd say not quite yet.

3) See above. MS supports their super PIM produces. HP is bundling PI with the HP Hx4705. It was on the CD of my Loox 600.
I'm not quite sure what you're saying here, but MS isn't the one supporting PI. HP and other hardware manufacturers may. Although, as pointed out more than few times, MS is better at communicating with the developers, so Alex at WebIS and Amit at SBSH can make stellar products. (OT - Just want to mention that anyone with a PPC should really check out PocketBreeze and that line of stuff. Amit over at SBSH is one of the most communicative developers I've ever seen. He monitors the forums over there and usually posts answers to problems and questions within minutes.)

[archangel]

2) Quality third party software is the backbone of the PPC community.
Oddly enough this is the reason I love the Palm OS so much. My Palm PDAs have always been heavily customized and loaded with third party apps. This is the one area the Palm OS had a heavy advantage in for years. Now the PPC has caught up and Palm needs to do more to help the third party community instead of hurting it.

If Palm can't see that it really will be dark times ahead.

[Jeff Kirvin]

ok, creepy, jeff first why were you on aximsite? second, how do you know my aximsite screenname? who, what, when, where, why?

and in my defence I was used to palm os and EVERYthing getting deleted on a hard reset but on PPC's only the ram not the rom gets deleted. So that was my fault not the operating systems.

I was on Aximsite because I like to do a lot of research before making a purchase.

And it wasn't hard to figure out, Mitch. You use the same sig there as here.

[Jeff Kirvin]

First off, calling me clueless. That's not the way we play here. This ain't PIC.

As to the second point regarding palmos development: Again why isn't the focus here on PalmSource? They are supposed to be the ones providing the IDE aren't they? I'm not a programmer, but it seems to me that it would be a great idea if Palmsource adapted .NET as a development platform? Why can't this happen?

PalmSource tried to license .NET, but Microsoft wouldn't allow it. Talk to them.

[intellidryad]

Palmsource needs a new licensee before their ONLY major licensee, palmOne, decided that it would rather die.

Why is palmOne screwing up again and again? Why are sony clies still unbeatable? palmOne needs to get some people that aren't stupid to think on this.

[gfunkmagic]

First off, calling me clueless. That's not the way we play here. This ain't PIC.


Sorry I hurt your feelings. :rolleyes:


PalmSource tried to license .NET, but Microsoft wouldn't allow it. Talk to them.


Interesting. Do you have a link?

[Jeff Kirvin]

Sorry I hurt your feelings. :rolleyes:

Interesting. Do you have a link?

It ain't about hurt feelings. I'm a writer, I've got really thick skin. It's about civility and tone. We just don't stoop to personal attacks here. If you can't abide by that, by all means go back to the rabble at PIC.

I'll have to dig up the link.

[Jeff Kirvin]

Here it is. http://tinyurl.com/7kj3j

The Palm executive then proceeded to recount some of the ways Microsoft has maneuvered to put Palm at a disadvantage in recent years. Perhaps most striking was his story of Palm's efforts to join in a Microsoft initiative called the Visual Studio Integration Program, which is designed to allow other companies to integrate their own software development tools with Microsoft's Visual Studio, the environment most commercial and corporate programmers use to create Windows software.

Eager to recruit more Windows programmers, Palm first approached Microsoft about joining the program in November 1999. For more than two years, even as officials of the Redmond, Wash., giant continued to declare in public that "anyone can join" the program and there are "no special requirements," they responded to Palm's application with empty promises, unexplained delays and improbable excuses, including -- get this -- the claim that they "lacked the resources" to let Palm in.

When confronted about these contradictions, the Microsoft official in charge said, "Quite frankly, seeing 'New Palm Project' . . . in the (Visual Studio) listing doesn't excite us."

"Why not?" a Palm employee asked.

"Because it isn't Pocket PC," the Microsoft official replied.

Early this year, after it became known that Palm would testify in the trial,

Microsoft suddenly discovered the necessary resources, and the two companies are now working out final arrangements, according to Mace.

But, he concluded, "The process by which we got there was very disturbing. If in order to get access to every Microsoft technology we have to wait two years, catch their employees in falsehoods and testify in a major court case, it will be an enormous hindrance to our business."

There's lots more of interest in Mace's testimony:

-- Microsoft's refusal to make a version of Internet Explorer for the Palm OS, even while offering Palm access to a much more limited browser called Mobile Explorer.

-- Microsoft's unwillingness to license .NET technology -- which Microsoft considers central to its future plans -- to Palm unless the handheldmaker would agree to a long list of burdensome restrictions.

-- The revelation -- not exactly a surprise, but not previously confirmed --

that Microsoft has developed secret networking interfaces for its own Pocket PC products.

[gfunkmagic]

It ain't about hurt feelings. I'm a writer, I've got really thick skin. It's about civility and tone. We just don't stoop to personal attacks here. If you can't abide by that, by all means go back to the rabble at PIC.

I'll have to dig up the link.

Please don't confuse me with Gekko. You made some outrageous speculations and I called you out on it. You should be able to take criticisms and not take things so personally if you want to make your opinions known like that. Furthermore, telling people to "get off" this site is certainly not professional...

Anyway, thanks for the info posted above. Sounds like the quote was taken during the anti-trust litigation period of Microsoft which is a couple years ago now. I wonder what eventualy happenned after that...

[Jeff Kirvin]

(sigh)

Calling someone clueless is an attack. Explaining why you disagree with their conclusions is discussion. See the difference? It was the title of your post that offended, not the contents.

Frankly, I don't give a toasted damn what you think of me. I'm just telling you that we don't tolerate personal attacks on 1SRC.

[Cyker]

I must say your tolerance is most impressive Jeff.
Were I in your position I'd probably be trying to kill them with the week-old pizza that I... really... need to throw out of my fridge... :D

About .Net, well, if Palm/Linux becomes a reality then there is a good probability Mono will get ported over, assuming Microsoft don't manage to kill it while shouting at the top of their lungs how 'Open' they are...
But I'm a native & source-code man, so I tend to avoid things like Java and .Net.

I'm not too worried about Palm's hardware vs PPC; As long as they don't screw over their devs, they'll be fine.
Remember, PPC's have been *FAR* ahead of Palm, technologically, for pretty much forever. Yet still Palm have been totally whipping them in terms of market share until relatively recently, and even then I think they are still marginally ahead.

Why?

Because, as Jeff says, they Just Work.
(At least until recent revisions :p :D)

[tungsten t5]

In microsofts defence, I have never had a problem with my dell x50v. It's worked great no hardware problems, nothin! and they have alredy released 3 updates to improve the device even more. (now wi-fi and bluetooth can be on at the same time since update 1) But ok their was that one problem on aximsite, but that was my fault not microsofts or dells.

Looks like Jeff might get a Dell, x50v maybe? even though he said he would never go back to Windows Mobile.

And yes the alarms have been fixed, and their is free software, one called Magic Button, that allos you to close programs when you tap on an x.

Also I would like to say that having programs run in the backround is nice, say you were surfing the web or playing a game, and all of the sudden you had to write a quick note down, well then you can just go right back to what you were doing.

Also jeff if you get a device with a microphone you can do podcasts on it even through skype.

[Jeff Kirvin]

Never say never.

I'd prefer not to waste memory the way WM multitasks. Palm's method is more efficient.

Will I get a Dell (dude)? I don't know. I'm watching both camps very closely.

[Surur]

Never say never.

I'd prefer not to waste memory the way WM multitasks. Palm's method is more efficient.

Will I get a Dell (dude)? I don't know. I'm watching both camps very closely.

You can have your cake and eat it too. Looks like styletap is trying to position themselves as a platform for migration of POS developers and their software. Also looks like they seriously want to address the conduit issue. Interestingly, in Styletap, when a program crashes mostly it only takes down the styletap environment, without touching the rest of th device. One click on the software again and you are up and running.

Surur

FOR DEVELOPERS
Use FtrGet (sysFtrCreator, sysFtrNumOEMROMVersion, &versionBuildNumber) to obtain the version number of StyleTap Platform that your application is running on. See the Knowledgebase article "Is This App Running on StyleTap™ Platform?"
Use the StyleTap API function GetDesktopIpAddress() in StyleTapPlatform.h to develop your own socket-based synchronization solution. More documentation will be available shortly, or contact StyleTap Support if you are interested.
Your application installation script can copy your .prc and .pdb files to the "ToBeInstalled" subdirectory (e.g. C:\Program Files\StyleTap\ToBeInstalled, if StyleTap was installed in its default location). The next time the StyleTap Application Installer program is run, all databases in this location will be imported into the "Files ready to be installed" list, so they can be installed to the handheld simply by clicking on the "Install to device" button.
VERSION 0.9.050 -- FIRST PREVIEW RELEASE (2005-04-27)
http://www.styletap.com/release_notes.html

[gfunkmagic]

(sigh)

Calling someone clueless is an attack. Explaining why you disagree with their conclusions is discussion. See the difference? It was the title of your post that offended, not the contents.

Frankly, I don't give a toasted damn what you think of me. I'm just telling you that we don't tolerate personal attacks on 1SRC.


As quoted by you:


Gekko's a jilted lover and a borderline stalker. He's got no rationality left. He doesn't want to help, only destroy. He won't be happy until Palm joins Osborne, CPM and the Amiga in the museum of computing dead-ends.

Do we really need that kind of help?

Talk about the pot calling the kettle black... :rolleyes:

I don't recall anyone telling you to get lost when you posted that over at PIC...

Btw, do you actually have any response to the points in my post or what?

[Cyker]

Surur: That is a very cool feature, and one of the advantages of going for a full emulation. It's slower and a lot harder to do, but for stability it's hard to beat ;)
I suspect the way Cobolt will do it will not be a total split, like with styletap, but be more akin to the way OS/2 could run Windows 3.1 applications. This means a rogue app could crash the system, but on the other hand execution speed should be much faster.

I still am wondering how both camps will deal with direct hardware access, and hardware access in general.

It's kinda amusing that PalmOS is enduring so well that people are burning zots to port it, which could have been used to code apps for the platforms instead :p :D


As for you two, stop bickering or we'll put you in detention :p

Oh, and AMIGA LIVES!!! (Damnit! :p :D)

[gfunkmagic]

I'm probably making excuses from MS, but I actually believe MS has an interest in keeping PDA's much less functional that desktops. For one thing,the income from a PDA ($5-20) is much less than the $50-200 they would get from a laptop with office installed. If one person in 20 decides to use a PDA instead of a laptop they are losing money. I think the success of RIM (not palm) in the enterprise market has really given them a kick in the bottom (same as in the browser sphere) as there is a risk of making the desktop irrelevant, or at least much less important, and this is of course their real cash cow.



I agree with this. The technology to fit XP into a pda form factor is already there as various devices testify. It's only the price point that is holding things back, but even that will eventually change...

[gfunkmagic]

You can have your cake and eat it too. Looks like styletap is trying to position themselves as a platform for migration of POS developers and their software. Also looks like they seriously want to address the conduit issue. Interestingly, in Styletap, when a program crashes mostly it only takes down the styletap environment, without touching the rest of th device. One click on the software again and you are up and running.

Surur


http://www.styletap.com/release_notes.html


Now don't you wish that would happen on PalmOS itself? :(

[lajandy]

I really think my supposition is not far off the mark - Look at every Palm made before the split: Very stable, robust standardised interfaces. You could synch an m515 with a IIIc's desktop!

After the split, we had a radical change - The T|3!
And it was a good thing, very good. We thought Palm might actually challenge the beloved Clie line! But with the change, I noticed lots of little things, the details, being left out or ignored. Possibly by the inexperienced coders who were once just doing the value-added software and systems integration, and now found themselves fully in charge of full-blown OS customisation!

Actually, if you have a T3 handy, take a good look at it--the logo says "Palm" and it's the classic blue circle, not the garish red+orange PalmOne logo. The T3 was designed and started production before the split. However, I will agree with you that it is a very good PDA (just try to pry it from my cold dead hands). Although it had its problems initially, after about 6 months on the market Palm (by then PalmOne I think) had ironed out most of the problems (with software, anyway). Of course, now PalmOne seems unwilling to bother fixing problems...

[Jeff Kirvin]

As quoted by you:

(stuff from a thread at PIC)

Talk about the pot calling the kettle black... :rolleyes:

I don't recall anyone telling you to get lost when you posted that over at PIC...

Btw, do you actually have any response to the points in my post or what?

But I didn't post that here, did I? How many ways do you need me to say it? We don't tolerate personal attacks at 1SRC. See? At 1SRC. Ryan has chosen a different tone for his site. Bully for him. And though you didn't post my entire PIC post, it was a response to a personal attack Gekko made on me. I'm done turning the other cheek. You want a war? You got it.

But not here.

As to the response to your post, I'm still thinking. The CDMA thing is interesting, and I'm sure that's a big part of why PalmOne isn't interested in Cobalt, but I doubt it's the whole story.

[Jeff Kirvin]

I agree with this. The technology to fit XP into a pda form factor is already there as various devices testify. It's only the price point that is holding things back, but even that will eventually change...

I've been saying for quite some time that Longhorn-based mini-tablets will eventually replace PDAs as we know them. How that makes me a "Palm apologist," I have no idea.

[jjesusfreak01]

As for you two, stop bickering or we'll put you in detention
Agreed, no more arguing, go to PIC.

[Vampire Lestat]

Good podcast. Jeff is looking at PPC again for 2 reasons: a) Even though he won't say it (or does not realize it yet), I think he is dissappointed with the LD. b) He is right about Palm dicking with developers.

I myself have gone through angry periods when I thought about moving to PPC. In fact, I bought nearly every flagship PPC unit out there. I eventually returned them all and kept using my T5.

After a few months without a PPC, we tend to forget what frustrated us then we want to go back to PPC to see if maybe we personally did not use the unit right or didn't give its chance. Iv'e been through this many times and always keep coming back to Palm.

That having been said, like Jeff, I am keeping all my options available. If Palm does not keep up with steady innovations, PPC will always remain a viable alternative, even as flawed as it may be.

[Jeff Kirvin]

I didn't stick with the PPC for a few months, but a few years. I know exactly what they can and cannot do. Where all the bodies are buried.

I'm not disappointed with the LifeDrive, Lestat. I love the hardware. The software is suspect. I'm beginning to think that each iteration of Garnet from PalmOne is more obfuscated and less stable than the generation before.

I don't know. I'm happy with my T5 (interestingly, after removing all the system patch software, Fitaly, SkinDIA, SkinUI and the rest, I haven't had a single crash). When it comes time to upgrade, I'll look at all the alternatives.

[JAmerican]

Kirvin, your friend at Solomedia is just like me. I'm a .NET programmer and I'm thinking about the new WM5.0 Clamshells coming out. I used an old Axim X3i and it moved quicky and looked great. No WiFi problems when I used it. Streaming did cause the device to lag a lot. I had to reset it multiple times. I'm going to test the new WM devices with streaming videos. I'm convinced that my next device is a Pocket PC. I can't take the problems I have right now with my device.

JAmerican

[Jeff Kirvin]

Kirvin, your friend at Solomedia is just like me. I'm a .NET programmer and I'm thinking about the new WM5.0 Clamshells coming out. I used an old Axim X3i and it moved quicky and looked great. No WiFi problems when I used it. Streaming did cause the device to lag a lot. I had to reset it multiple times. I'm going to test the new WM devices with streaming videos. I'm convinced that my next device is a Pocket PC. I can't take the problems I have right now with my device.

But I thought you loved your UX-50!!! :eek:

[jjesusfreak01]

But I thought you loved your UX-50!!! :eek:
He does, but he also likes the form factor, and he is having wireless problems and such. Since there do not seem to be any clamshell Palms coming out, JAmerican may be changing OSs.

[JAmerican]

He does, but he also likes the form factor, and he is having wireless problems and such. Since there do not seem to be any clamshell Palms coming out, JAmerican may be changing OSs.

Thanks jjf01 for clarifying what I thought was obvious. Clamshell now and forever. I love the look of clamshell/tablet-based devices.

JAmerican

[Jeff Kirvin]

Thanks jjf01 for clarifying what I thought was obvious. Clamshell now and forever. I love the look of clamshell/tablet-based devices.

Have you seen the Flybook? Not all that much bigger than your Clie, but it runs XP Tablet Edition. Full convertable at a tiny size.

[JAmerican]

Looks so cool. Especially the black one. But its too big for me. I need compact. I don't want to have two PCs. That will lead me to store info on both. I don't want to do that. I want to store only media, a few docs on my handhelds as well as apps. Look at this...

Black HTC Universal (http://www.systranbox.com/systran/box?systran_lp=zt_en&systran_id=SystranSoft-en&systran_url=http://www.eprice.com.tw/news/?news_id=2913&systran_f=1118968423)

JAmerican

[JAmerican]

I am not going to spend any more money on my UX. The next step is saving up for the HTC Universal. If it isn't released in the US, I'll find a way to get that or a similar PPC. I need clamshell.

JAmerican

[jjesusfreak01]

I am not going to spend any more money on my UX. The next step is saving up for the HTC Universal. If it isn't released in the US, I'll find a way to get that or a similar PPC. I need clamshell.

JAmerican
Just wait for HTCPet to do a translation.

[JAmerican]

Just wait for HTCPet to do a translation.

LOL. I think its in English already. At least from the pics I saw. I hope it has the WM 5 has the language selection screen.

JAmerican

[Surur]

I'm not disappointed with the LifeDrive, Lestat. I love the hardware. The software is suspect. I'm beginning to think that each iteration of Garnet from PalmOne is more obfuscated and less stable than the generation before.

But surely it was madness in the first place to place the 64MB NVFS on the hard drive, instead of a separate flash chip. It was never going to be good. That was a hardware decision.

Surur

Edit: I just realised this this was also actually a software decision. The obviously just re-used the code from the T5 (where the NVFS was on flash) and substituted a HDD. This reduced the amount of low level coding they had to do (vs supporting another memory chip plus a HDD). Smacks of laziness.

[Thinkertank]

So.. is there anyone that has contacted Palm directly ?? And were their reactions from Palm??

BTW Jeff! I read somewhere that Fitaly works on the LD! ;)
Haven't tested it.. but.. give it a try
http://fitaly.com/board/palmfitaly/posts/4990.html

[Alan G]

I started to read the August issue of PC Today, and I saw an little blurb on Microsoft's Team 99. Here is the entire article:

Team 99 Taking Longhorn To Its Launch

When Microsoft releases the final version of Longhorn late next year, it will have gone through some rigorous testing by Microsoft employees and the members of Team 99, a group of 20 or so bloggers chosen to create a buzz about Microsoft’s newest software.

Robert Scoble, a Microsoft employee and Longhorn evangelist, announced the resurrection of Team 99 in May. The bloggers were given access to Longhorn in return for signing nondisclosure agreements. Team 99 is responsible for working through the program and, according to Scoble, “telling us where we’re screwing up and what we are doing well.”

How did Team 99 come about? Scoble announced in a blog he posted in May that he was searching for bloggers to get the word out about Longhorn. In his words, he explained Longhorn got its name from the bar that is between Whistler and Blackcomb in British Columbia. The 99 comes from the road you drive from Scoble’s house to get up to the Longhorn bar. “So, Team 99 is the team that’ll take us to Longhorn’s launch,” he wrote.

So why isn't palmOne doing this? (Or even Apple now that they are switching to Intel processors? But that is off-topic, so I'll stop here.) As Jeff said in the podcast, palmOne need to get vocal users out there spreading the word about palmOne's devices and software. I really hope that palmOne is suffering for management inaction rather than anything more...corporate. In other words, I hope that palmOne just hasn't seen the light to engaging the user community as oppose to bending to wireless carrier's wishes. (I'll leave that alone because that is a very chilling thought. If that were true, I'd likely leave the platform. And that's a bold statement for me. Honestly.)

As Jeff already knows, I've been thinking about this subject for a log time now. I'm at the point were I'm seriously considering starting my own PUG in Connecticut (Yes, I know that there is already a PUG that meets in Stamford) to carry forward the positive aspects of palmOne, PalmSource, and the platform as a whole. Anyone who wants to join in on the fun, PM meet here at 1src.

Alan G

[Alan G]

By the way, Jeff, I thought that you ended your podcast on an extremely powerful note. I need to get some letterhead printed up over the weekend so I can mail in a letter. Wow. Maybe you should send an audio CD of the podcast over to palmOne also.

Alan G

[JAmerican]

By the way, Jeff, I thought that you ended your podcast on an extremely powerful note. I need to get some letterhead printed up over the weekend so I can mail in a letter. Wow. Maybe you should send an audio CD of the podcast over to palmOne also.

Alan G

That's a great idea.

JAmerican

[Cyker]

But surely it was madness in the first place to place the 64MB NVFS on the hard drive, instead of a separate flash chip. It was never going to be good. That was a hardware decision.

Surur

Edit: I just realised this this was also actually a software decision. The obviously just re-used the code from the T5 (where the NVFS was on flash) and substituted a HDD. This reduced the amount of low level coding they had to do (vs supporting another memory chip plus a HDD). Smacks of laziness.

I think the whole measure was supposed to be cost cutting - You see, Flash is the most expensive of the current mainstream storage technologies, followed by conventional DRAM followed by disk.
If they'd gone with the hybrid route, it would have negated a large portion of the original reason for them doing it.

Of course, it seems a bit odd now given it's price - I'm not sure how it ended up being so expensive when I was expecting it to be around the T|5's price-point!
I must admit, I've been very spoilt by my TH55 and now my Zod2 - The TH55, brand new, was almost half the price of the LD and has more 'tickbox features' (I think it's a better device too, but that's an argument for another thread ;)). And recently my Zod2 was like, a quarter of the LD! (Although I'm probably cheating here since it was at clearance price and not brand new...).
Unfortunately for Palm, I've come to expect a lot more for my pennies these days.


To be honest, I don't think having the 64MB RAM as Flash would have made a significant difference. In fact, I can't think of any difference it would have made! At most you'd have an extra 64MB in the internal VFS area, which is pretty insignificant on a 4GB disk...

The saving in software zots would have been very slight - The Internal Flash/fakeRAM would have been code lifted from the T|5, and the HD would just be mounted as an internal VFS using standard code, along with driver support for the HD interface.

[archangel]

I would have greatly prefered the 4GB drive be used purely as VFS, but I agree it would have added to cost so its doubtful Palmone would even consider this.

[Admiral Cyclops]

FullPower extends battery life on T3s (but be careful). PalmOne is deliberately making it difficult to code for their devices. Stand by your developers and let PalmOne know that this behavior WILL NOT STAND! [details (http://www.1src.com/?m=show&id=1089)]

You wonder about why PalmOne does what they do... The similar question has been asked about every major software company in the world - anyone that uses code from some company has found issues with documentation.

Here's why...

First of all... Documentation takes time as does testing your fault conditions. It takes more time to do this structured (and may I say proper) way of coding than it does to just hash out some code and put it up. Time for a company is money and programmers don't decide their own deadlines. It is very possible that the people who can get ahead at palm one due to upper managment techniques are the ones who don't comment and don't document and just hash it out. I agree its stupid, but its not a plot and its problem created by the upper managment not seeing what it takes to code properly. For them, its an issue for programmer x puts out this much code in this much time, programmer y puts out this much code in THIS much time. Not a question of this is high quality documented code and this isn't.

Second of all... A lot of people who buy palm handhelds don't load it up with apps anymore. My own parents (not computer people at all) who have palm's stick with the built in apps because they don't want to bother with finding another app that does better. They'd rather just make what they have work. This is a common thing that programmers and geeks do. We think that if it doesn't work how we want, get something that does. Non geeks tend to think in terms of does it work for most of what I need, ok, I'll deal with it rather than deal with making the technology work. Non geeks don't care. As long as it fits most of their needs, they'll be happy.

Third of all... Releasing exactly what changed to the public is tricky because they obviously have programming strategies and 'franken garnet' is one of them, even if we have no idea why. They don't want to give out why they did x in garnet and how to make it work as it may reveal strategies they don't want people to know. In the podcast, you asked what they have to gain? Well... Their apps work the best with the DIA meaning you are not likely, if you are a geek, to replace it with something else and then that supports them saying, 'yes our apps aare the best for the device' they don't have to say why they came to that conclusion.

If the users can't make replacement apps work, their apps get used most and they can say its the best. That is the only 'conspiracy' theory I can think of. However, I stick with my first statement... Its a matter of people who don't program controlling what the programmers do and as long as the programmer puts code out that can be demoed for the managment at work they'll get ahead.

Like novice programmers trying to accept text input using cin in C++...

I can write a program saying just:

cin << val;

However, as most of us know without fault checking I can demo this as doing exactly what val expects and have to HOPE that no one enters a value that doesn't fit into val.

If I was demoing that for a managment they may have NO IDEA that I put not fault checking in, all they have an idea of is that I said it works and I got it to them fast.

[Surur]

I think the whole measure was supposed to be cost cutting - You see, Flash is the most expensive of the current mainstream storage technologies, followed by conventional DRAM followed by disk.
If they'd gone with the hybrid route, it would have negated a large portion of the original reason for them doing it.

Well, 128Mbyte NAND is between $7 and $9 (http://www.aice.com.hk/index.aspx). Even if they used NOR, thats still only 2-3 times the price. I dont buy your argument. At the very least it would have increased battery life enormously to 8-10 hours (exceeding the T5 battery life by 25%) vs underperforming the T5.

Surur

[Alan G]

Ok, I'm really hung up on this product evangelism concept. On the chance that someone from palmOne does poke their head in, here is some more info on Microsoft's Team 99.

http://www.eweek.com/article2/0,1759,1791413,00.asp
and
http://radio.weblogs.com/0001011/2005/05/01.html#a9968

Alan G

[Jeff Kirvin]

Microsoft understands how powerful community feedback can be, but Palm never has.

[Jeff Kirvin]

But surely it was madness in the first place to place the 64MB NVFS on the hard drive, instead of a separate flash chip. It was never going to be good. That was a hardware decision.

Surur

Edit: I just realised this this was also actually a software decision. The obviously just re-used the code from the T5 (where the NVFS was on flash) and substituted a HDD. This reduced the amount of low level coding they had to do (vs supporting another memory chip plus a HDD). Smacks of laziness.

In all fairness, it probably also had something to do with complexity. The more components you have in a system, the exponetially more likely something is to break down. Keep it simple.

Jeff (who drives a car with manual windows, manual door locks, no cruise control, no AC and a stock FM radio for this very reason)

[Jeff Kirvin]

So.. is there anyone that has contacted Palm directly ?? And were their reactions from Palm??

BTW Jeff! I read somewhere that Fitaly works on the LD! ;)
Haven't tested it.. but.. give it a try
http://fitaly.com/board/palmfitaly/posts/4990.html

It sort of works. Still has problems when you minimize the DIA.

[Jeff Kirvin]

By the way, Jeff, I thought that you ended your podcast on an extremely powerful note. I need to get some letterhead printed up over the weekend so I can mail in a letter. Wow. Maybe you should send an audio CD of the podcast over to palmOne also.

Thanks. Can you tell that I tested out of Public Speaking in college?

[archangel]

Jeff (who drives a car with manual windows, manual door locks, no cruise control, no AC and a stock FM radio for this very reason)

Wow, I thought I was the only person that does this when they buy a car. I'm definitely the guy that wants no power door locks and windows because they are always the first thing to crap out. Well expect for the AC. That is a must in San Antonio almost year round. I skip the other frills though.

[Jeff Kirvin]

I seem to be doomed to swelter. I grew up in Houston, spent my military career in DC (and if you've never been to DC in the summertime, don't go), and now we're having a heat wave in Denver. Part of my job is driving around in my car (which I chose to have with no AC) and when I get home, one of my two wall AC units has crapped out and only the fan works. Oy.

Well, it's a good incentive to lose weight...

[Jeff Kirvin]

Wow, I thought I was the only person that does this when they buy a car. I'm definitely the guy that wants no power door locks and windows because they are always the first thing to crap out. Well expect for the AC. That is a must in San Antonio almost year round. I skip the other frills though.

I will say this, though. I abuse the heck out of my Neon and largely ignore scheduled maintenance, but that little critter is indestructable. Minimalism has its perks.

In Palm-related news, I'm doing the same thing with my T5. For nearly a week now I've had it stripped down to just the basics. Well, actually I still have nearly 60 applications loaded, but I've gotten rid of almost everything that actually hooks into the OS. No skinning, no pToolset, no dbcachetool, no ZLauncher... The only thing I couldn't bear to get rid of was YAHM/Fontsmoother.

And in that time, no crashes. None. Just like previous versions of Palm OS, Garnet gets flakier the more you hack it.

[Cyker]

Well, 128Mbyte NAND is between $7 and $9 (http://www.aice.com.hk/index.aspx). Even if they used NOR, thats still only 2-3 times the price. I dont buy your argument. At the very least it would have increased battery life enormously to 8-10 hours (exceeding the T5 battery life by 25%) vs underperforming the T5.

Surur
When you consider that disk storage is LESS than $7-9 per GIGABYTE, and then scale everything up to the sort of quantities that Palm make their PDAs in, it makes a lot more sense.

$7-9 may not seem like much, but when you're talking about huge scales it adds up.

I doubt that the Battery Life would have been increased by that much; In fact it would probably get killed even more if the user was writing stuff and/or using the hard disk (You need Flash-burner circuits AND hard disk motors!).

Besides, if they had, you (Or me :p) would probably be complaining about how pointless that extra FlashRAM was when they could have just used a portion of the hard disk ;)

Jeff's point about increased complexity is also a fair point.


What would be neat is a realRAM device, like a TH55 but with 128MB RAM, a CF slot and an MS/SD slot with operating system provisions for panic-dumping (UX-style!) the internal RAM to CF/SD/MS when approaching a low battery condition. And a 1-button restoration post-hard reset.
Or possibly 256MB internal VFS :D

Well, I can dream can't I damnit?? :p

I wonder, if we all made our dream design Palm PDAs and sent them to Palm and Tapwave, whether they'd distill it and make something? :D Or just ignore :(

[archangel]

I tried to do no Zlauncher with my Zod since it seems to cause issues with Audible Player , but I couldn't do it. However, as long as the crashes are rare (which they are on my Zod) I can deal with them. If I'm using hacks and the thing crashes I blame myself and not the device.

I do miss the old days of OS3 and OS4 when devices where super hack friendly and you only had to charge the thing once a week.

[Jeff Kirvin]

I do miss the old days of OS3 and OS4 when devices where super hack friendly and you only had to charge the thing once a week.

Disclaimer: I'm a Gemini with bipolar disorder.

Maybe I'm just the kind of guy that loved to trick out my hardware as much as possible (when I'm not on my minimalist kick), but I had problems on my OS3 Visors. I also had them hacked to death. Sometimes as many as 20 hacks running at the same time. The instability was one of the things (along with lack of color or easy card expansion) that drove me to Pocket PC in 2000. Took me three years to come back to Palm.

[Cyker]

OS3-4 were the golden ages for me.

Devices are getting a bit too unwieldy for my tastes now, but this is a typical trend - The same thing is happening with Windows.

Thats the reality of the market place, and we're on the trailing edge of that bell-curve in terms of PDA tech...

(We're past the Useful point, and into the Adding Mostly Useless Added Value Features To Make You Upgrade)

OTOH, we'll soon be into the realm of literal Palmtop computers (Matlab in the palm of your hand! :p), although I'm not sure whether that's a good thing or not...


Jeff - That is one thing I won't miss from the OS3-4 days - Hack instability... I gotta say 'tho, when I stopped using X-Master and went to YAHM on OS4, it did help a lot. Finding DA-equivalents for a lot of my hacks was a good route too! I wub my T665, even 'tho it's now basically our family's universal remote control :D

[Joel]

Microsoft understands how powerful community feedback can be, but Palm never has.
Yup. Take the LifeDrive for instance. Some sites were never invited for the LifeDrive testing phase. Do you have to beg them to get you to join? I think it should be Palm who should be inviting those sites to join in.

[Joel]

..

In Palm-related news, I'm doing the same thing with my T5. For nearly a week now I've had it stripped down to just the basics. Well, actually I still have nearly 60 applications loaded, but I've gotten rid of almost everything that actually hooks into the OS. No skinning, no pToolset, no dbcachetool, no ZLauncher... The only thing I couldn't bear to get rid of was YAHM/Fontsmoother.

...

This is the current state of my Treo 650. Basic stuff. I get to experience what the Palm OS is all about... same excited feeling I got when I first got my USRobotics Pilot.

[PDAJah]

Jeff Kirvin, your thoughts re: PalmOne/Smartphones/Networks may not be that far from reality. The "3" 3G network, here in the UK, sells Symbian Smartphones but they have installed a "walled garden" so that it is nearly impossible to connect to the Internet (you can only "browse" the "3" web pages) and to install standard UIQ applications - you can only install apps that are approved by the Network!

[jjesusfreak01]

Yup. Take the LifeDrive for instance. Some sites were never invited for the LifeDrive testing phase. Do you have to beg them to get you to join? I think it should be Palm who should be inviting those sites to join in.
Yeah, they should send me one. I was invited to the DevCon, because Tam could not go, but getting a test handheld would be more fun. Of course, going to a conference, as Jeff says about the MS conference, could get you free stuff too. Just not at Palms conferences. I guess they just dont have the funds to throw around free handhelds.

[Jeff Kirvin]

Yeah, they should send me one. I was invited to the DevCon, because Tam could not go, but getting a test handheld would be more fun. Of course, going to a conference, as Jeff says about the MS conference, could get you free stuff too. Just not at Palms conferences. I guess they just dont have the funds to throw around free handhelds.

On the other hand, how can they afford not to?

[Joel]

On the other hand, how can they afford not to?
LOL. Very true... they lost lots of opportunities.

[CrashX]

I have a general comment concerning the podcast. While the content is generally very engaging and informative, the delivery of it is less than stellar. What I'm referring to is Jeff's over use of "um", "ah", "and uh", "so yeah" and those deep breaths. I don't know how else to put it, but it's very distracting and gets annoying. This is true after maybe 3 minutes of listening. It makes me wince. I know you're not a "professional" broadcaster, but I think I stopped listening to the podcast after #3 due to this.

[Cyker]

LOL! Ironically I find them kinda entertaining ;)
I must admit the sometimes-long pauses do confuse me, but the um's and ah's just feel more, I dunno, natural. I find most casts that skip that out, either the guy is talking too fast or they sound like they're a zombie (Or reading a sheet/screen/autocue)..

I know what you mean 'tho, the other podcast I mainly listen to is TWiT and they don't have that, but then there are a lot more people on the cast. One annoyance about that 'tho is how Leo keeps interrupting everyone mid-sentence :p :D

[archangel]

Yeah, I mean obviously most people that do podcasts are not professional radio guys. Its a free service and full of good info so I don't see it as an issue. I've definitely heard much worse podcasts than Jeff's.

[CrashX]

Yeah, I mean obviously most people that do podcasts are not professional radio guys. Its a free service and full of good info so I don't see it as an issue. I've definitely heard much worse podcasts than Jeff's.

I haven't heard any other podcasts. Jeff's is the first I found worth listening to, since I enjoy his articles (http://www.writingonyourpalm.net). Excellent free service, agreed.

Hopefully it comes off as constructive criticism, rather than nitpicking.

- cheers

[Jeff Kirvin]

I'm working on minimizing the ums, but I'd rather not clean them up in post, for the reasons that Cyker mentioned.

[Cyker]

...although this has given me a fantastic idea for the next April Fools joke :D

"Hi, this is the 1src Podcast, I am your host <drop half an octave> DARTH KIRVIN *shhhhh* *whhhhhhhh* [cue Imperial March]"

[Jeff Kirvin]

Actually, my Sith name is Darth Portly, the Wide.

[Cyker]

ROFL! :D


Edit: 2nd attempt to fix mysterious lower-case, after first attempt had no effect...

[Skaah]

Is there an example letter you can send to PalmOne?

[LupeValenz]

Sony come back, we need you!!!!!

[Cyker]

Amen to that!!!