Trolltech Qtopia 2.1 ROM soon

[_Psycho]

I know its been posted before, but in a "off-topic" post.
http://www.trolltech.com/developer/faqs/qt...tml?cid=22#q110

I was talking to lpotter (Qtopia Community Liaison) the other night about TT plan, and he told me (like the faq) that the Qtopia version 2.1 rom should be out very soon. Thats great news. Im really looking forward to it.

Like 5-6 months ago. I did some Qtopia 1.7 image for my c700, it was really bugged (couples of patches i took from opie to make keyboard work, ts was messed up etc). But i saw the UI and it was a really great improvement. Everything was nicer and cleaner. Im looking a lots forward the 2.1 version (especially since sharp is stuck at 1.5.4 forever). I believe since its directly linked to Qtopia Desktop that Sync will be really better. (Hopefully with Mozilla someday). Of course a lots of peoples swtiched to KDE/PiM already, but who know what TT will release

The ROM will be gcc 2.95.3 to keep compatiblity with older version (so standard sharp ipk will work, guessing the rom is using the same ipkg). I guess they will use the standard 2.4.18 kernel too. Hopefully sashz or others will add the "nice to have" features and support from cacko ROM next

I hope they release the ROM with gcc 3.4.2 for the 20-25x speed improvement later. (even if it break all .ipk compat).

Anyone else looking forward the new Qtopia version ?

[kopsis]

Hopefully sashz or others will add the "nice to have" features and support from cacko ROM next

I hope they release the ROM with gcc 3.4.2 for the 20-25x speed improvement later. (even if it break all .ipk compat).

Anyone else looking forward the new Qtopia version ?

While the newer Qtopia stuff may be nice, I doubt there are any improvements that would allow most people to justify sacrificing Sharp ROM compatibility. Compatibility is one of the main reasons that the Sash/Cacko ROMs have been so popular. I believe (though I may get flamed for saying so) lack of compatibility cost Opie a lot of developer support (and lost Opie app developers a lot of potential users) and I'd hate to see the same happen to Cacko. I hope that the Cacko ROM continues to do what it has done so well in the past ... improve on the Sharp ROM without trying to radically change the fundamental components (kernel and Qt/E - Qtopia) so it doesn't fragment into yet another "niche" ROM.

As for the gcc 3.4.2 speed improvements ... those are for floating point operations only. The FP speed boost is great if you want to run Quake or an AV codec that hasn't been properly "integerized", but 99% of Zaurus apps will see no more than a couple percent speed improvement (if any) from building with gcc 3.4.2 with the alternate FP handling. Just improving Quake is not, IMHO, a compelling reason to break all .ipk compatibility!

[Mickeyl]

I'm also a bit sceptic about that Qtopia ROM since you know my attitude towards fragmentation. Then again, what will the Opie team do? Will they catch up with the Qtopia improvements or improve their own side track? Catching up would be better in terms of compatibility but is probably a huge task and the Opie team does not have the resources to do that.

@kopsis: To many people it's about stagnation or progress. The SharpROM and its clones are hopelessly outdated in many areas, not only the kernel. The most important benefits of Linux over stuff like PalmOS and WinCE are freedom and huge progress. With Sharp stagnating, both advantages start to vanish hence will be the attractiveness of the Linux platform to potential PDA buyers and developers.

Btw., the next steps for OpenZaurus will likely be to use the forthcoming g++ 4.0 and uclibc which both will give a real nice surprise in speed and size improvements.

[lardman]

Then again, what will the Opie team do? Will they catch up with the Qtopia improvements or improve their own side track? Catching up would be better in terms of compatibility but is probably a huge task and the Opie team does not have the resources to do that.

There's little (or nothing) which I can't get from (or would be able to get if I were to add to) OE, so I'm all for Opie 'branching' out.

I can't see that the introduction of a new ROM will suddenly increase commercial (read: no source code available) software production, so it's neither here nor there IMHO.


Si

[lpotter]

Those roms are compatible with applications compiled for the Zaurus, assuming they used the SDK available,
Source code for Qtopia 2.1 will be available.
What you'll get is a stable product that syncs with Windows, Linux and Mac OS.
These roms will come with NO support from Trolltech.

[acpkendo]

@ lpotter

First off, let me say that you guys continue to impress me.  Between opening QT up for the KDE folks (not really for THEM, yatta yatta), and now picking up the ball that Sharp dropped so unceremoniously, Trolltech has always been a class act in my view.  If I ever get around to getting proficient in Python, rest assured that PyQt will be my choice.

Re: Qtopia ROM 2.1. . . I assume when you say

syncs with Windows, Linux and Mac OS

you mean "syncs with Qtopia Desktop on Windows, Linux, and Mac OS."  Will there be any hope for those of us confined to Outlook at work?

[ev1l]

That's very good news loptter. If I understand you correctly, it should be no problem to get the cacko developpers to start improving on 2.1 instead of the standard Sharp ROM, right?
Will we be able to synch up with Outlook/Mozilla? Integration is really a strong selling point whenever I'm involved in PDA implementation projects, so that would be great.
I'd have a serious talk with the KDEPIM/PI guys, they've done a great job replacing the limited integrated PIM apps.

[clivel]

What you'll get is a stable product that syncs with Windows, Linux and Mac OS.

Hi,
Will there be any way for developers to add support for synching 3rd party apps? i.e. something similar to Palm conduits.
Thanks,
Clive

[iamasmith]

@kopsis: To many people it's about stagnation or progress. The SharpROM and its clones are hopelessly outdated in many areas, not only the kernel. The most important benefits of Linux over stuff like PalmOS and WinCE are freedom and huge progress. With Sharp stagnating, both advantages start to vanish hence will be the attractiveness of the Linux platform to potential PDA buyers and developers.

Btw., the next steps for OpenZaurus will likely be to use the forthcoming g++ 4.0 and uclibc which both will give a real nice surprise in speed and size improvements.

Mickey Hi,

Looking forward to taking a look at newer versions of OZ as they come along, I have great hopes for the project, however, your comments on stagnation whilst technically true also ignore the main reason that people continue to use Cacko/Sharp ROMs and really that is compatibility with Zaurus IPKs written for Qtopia.

Prior to getting a Z I had been playing around with OPIE on iPaqs running familiar for a couple of years... was finally totally depondent that HP couldn't release an SD driver for the H3970 that I had (thought it would be a great machine - 48Mb Flash !) due to the vendors of the chipset not wanting an open source driver published (even though the H5450 got one !) and bought an SL-C860.

However, one observation that I had during that time was with every possible update there seemed to be several critical apps that would prove quite dear to me that invariably kept chucking sigsegv's  .

I tried OZ a few weeks ago and the situation had really improved, however, there were still a lot of nasty bits that hadn't been resolved and to be quite frank returning to Cacko 1.21b was a breath of fresh air... the ROM was snappy, I had fast ATIW100 acceleration and I could still tailor it.

Lets face it, in terms of desktop type power the Z hasn't got the grunt so attempting to go that whole extra mile for the fastest FP support etc. has somewhat limited appeal unless you have a real need for it.

If, however, you realise that 2.4.18 kernel/2.95.3 gcc still is a powerful combination in comparison to other PDAs then you can stretch the basic Sharp ROM into new areas without compromising compatibity... i.e. I just discovered that Wikipedia static pages couldn't be processed by the wiki2zaurus project  but the original author for wiki2static.pl was still maintaining... Now I'm a much more seasoned C/C++ programmer than a perl programmer so I built a squashfs module for Cacko 1.21b... problem solved... what takes up 3Gb of space on my desktop can be squashed into 550Mb on my Z and it means I only have to rely on one script maintainer... As I said on that thread "Now can you do THAT on your PocketPC" .

Anyway, I have hopes for the future of OZ and would LOVE to see stable versions of OZ running with 2.6 Kernels but I think that I would be more likely to go GPE than OPIE given that my main motivation in going that route, if I couldn't get 'standard' packages working would be to choose an easier porting path - the X route is obviously going to be better  for this.

No real conclusion to my speach, would really like to see how OZ/OE pans out but at the moment the 'old school' still has a lot of advantages.

- Andy

[lardman]

I disagree with you Andy (iamasmith) ;-).

There's not much that I want which can't be compiled for OE (so no need to be tied to GCC 2.95.3).
The exceptions are:

* a spreadsheet app (opie-sheet needs work, all well and good, at least there's a starting point)
* a Word processor (actually there may be something decent out there now but I've not been looking for a while) - MS Word compat would be nice, but in truth it's easier to work in RTF then do the main stuff on a *big* PC - that said the ability to read MS docs would be useful (Word/Excel).
* Java (not sure whether there's anything on this front)

I can't think of anything else.

GCC 3/4 improves not only the fp speed but the normal code too (or so I believe), and any speed improvement is a good thing. The 2.6 kernel is also supposed to be far better speed-wise.

A couple of other annoyances - some apps can't be compiled with GCC 2.xx, they need 3.xx (Octave and R for example).

I can see why you'd want to stay compatible, but I think that going forward is the only way to go.


Si

[Mickeyl]

lardman pretty much voiced all I could answer. If there was a huge amount of commercial stuff which only runs on SharpROM + Clones, then I would agree that compatibility is a good thing. But as it is, it's really just a couple of Office Style Apps (for which I question the usability on a PDA anyway) and some miscellaneous Java things. Everything else is open source and either already in OE or could be added by someone who is interested and motivated.

I tend to see the bigger picture once and then, which is to attract new users and developers you need to support current software. Sharp in the year 2004 still releasing devices w/ a confirmed outdated and buggy kernel and insisting not to cooperate with mainline is in my eyes a very unfortunate signal. We as the OpenZaurus community try to improve this situation, but our power is limited (as is the number of volunteers). We can't afford to fractionate it even more by trying to progress _and_ staying compatible.

[kopsis]

A couple of other annoyances - some apps can't be compiled with GCC 2.xx, they need 3.xx (Octave and R for example).

In all fairness, there are versions of R that will build just fine under GCC 2.95. No, they're not the "latest and greatest", but that doesn't mean they're not good enough. Good enough seems to be something of a paradox in the Open Source world. It's fundamental to the "release early, release often" philosophy -- you don't hold release until the app is perfect, you throw it out there once it's "good enough". Yet at the same time, a huge amount of effort is invested/wasted in keeping up with the latest kernel and toolchain even though doing so may not have any quantifiable benefits.

Is the 2.6.x kernel superior to 2.4.18? Absolutely! Does that mean any Zaurus running 2.4.18 is hopelessly obsolete and useless? I think not. For most users 2.4.18 (with a very small handful of useful patches) is really good enough. Case in point, the Kopsis web/email server is still running Debain "stable" with a 2.4.27 kernel. Would 2.6.8 give me better performance? Probably. Would it be worth the time/money to upgrade? Nope. Would Apache 2.x give me lots of new features? Yep. Do I need any of them? Nope.

Now, there are plenty of cases where upgrades do make sense. My workstation, for example, always runs the latest released kernel. My point is that for special purpose machines (and for many users the Zaurus falls into that category), progress is wasted effort if what you have is good enough.

Now for some, Sharp ROM compatibility isn't good enough. I respect that and I aplaud the creation of OE to address those people's needs. But to expect those for whom Sharp compatibility is good enough to sacrifice not only compatibility but also word processing, spreadsheets, syncing, video acceleration, etc. just for the sake of "progress" seems unreasonable.

[lardman]

In all fairness, there are versions of R that will build just fine under GCC 2.95. No, they're not the "latest and greatest", but that doesn't mean they're not good enough.

Agreed, there are versions of both R and Octave, but in my case they are not good enough (talking about Octave here) as 2.1.57 is vastly better than 2.0.17 in terms of compatibility with MATLAB.

I agree with you that there's no reason to keep updating willy-nilly for no reason other than that the new version is newer, but I really would like my Z to be faster, to have a more stable kernel, better usb handling, etc.

But to expect those for whom Sharp compatibility is good enough to sacrifice not only compatibility but also word processing, spreadsheets, syncing, video acceleration, etc. just for the sake of "progress" seems unreasonable.

Agreed, so these things have to be improved for OE.

The truth of the matter is that there really isn't much (even the above) keeping me with the old stuff, and I don't want my Z to become extinct.


Si

[cybersphinx]

lardman pretty much voiced all I could answer. If there was a huge amount of commercial stuff which only runs on SharpROM + Clones, then I would agree that compatibility is a good thing. But as it is, it's really just a couple of Office Style Apps (for which I question the usability on a PDA anyway) and some miscellaneous Java things.

... and the Japanese input methods (especially the hand writing recognition). And that's what keeps me (and hordes of Japanese Zaurus users) with the Sharp ROM.

[Mickeyl]

Right. And since we don't have the source of those things, it will stay like that until someone is annoyed(*) enough of the SharpROM to rewrite them from scratch.

(*) Annoyance has shown to be the #1 motivation to get something done in open source land