A Big Problem... (deleted) Options

[Laze]

Sash - Mickey was not pointing at anybody but making a general statement.

MickeyL i agree with you with some of the things.

Open source / Linux people have one big problem - leadership and central control but thats also why strange project and solutions popup - in a microsoft world these would be killed. He he

[Glaive]

Unification of pdaX & OZ won't do good to our zauruses. Each of these projects has an idea that appeals to some part of the community. But I don't think that flame wars should emerge from these differences. You guys should collaborate and share the best of these roms.

P.S i wouldn't call any of the roms crappy, so the rule about having one cool thing instead of many crapy things doesn't work here. We got a lot of really cool Z roms!

[Mickeyl]

*sigh* I knew my disclaimer wouldn't be enough to stop people from twisting my words the way they want to. I was thinking about hundreds of window managers, thousands of line editors, gazillions of POP3 mail clients etc...

It's probably not the proper forum for general ranting. *shrug*

[Cresho]

You know what's killing me here? You guys are so good at putting up the rom images but who is making the software? There are just too few people.

For the sharprom, I see tons of people but are well hidden. for the pdaxrom, just 2? and for openembedded, less than 4? for cacko, 1 and he makes his own software and redistributes a few and is also compatible with sharprom.


One of the major reasons that nobody is diving into this is.....................flexibility. kopi does wonders but not good enough. i use it myself.

I'm just saying if this division keeps going on, we seriously are going to loose alot of potential and just vanish!

I don't think we need roms. We do need softwares that can sync, run fast, and actually make us more productive. Somthing to compete vs the cellphones, ipods, and gamboys-which these devices I find to be a joke to begin with.

[Meanie]

You know what's killing me here?  You guys are so good at putting up the rom images but who is making the software?  There are just too few people.

For the sharprom, I only see one person.

for the pdaxrom, just 2?

and for openembedded, less than 4?

for cacko, 1 and he makes his own software and redistributes a few.


One of the major reasons that nobody is diving into this is.....................flexibility.

kopi does wonders but not good enough.  i use it myself.

I'm just saying if this division keeps going on, we seriously are going to loose alot of potential and just vanish!

I agree with Cresho. The ROMs/distros are fine and lots of work have been put into packaging them up, but new application development is lacking. Sure, we can compile and package most linux apps up, but a lot of them were not intended for Z use and some customisation does need to be made to them in order for them to be really useful on the Z. I say the application developers are the one the various distros need to woe and attract, not just the average users. One thing is development tools or lack of. Getting started to write apps on the Z is not easy, especially if you are from the Windows or Java world. I guess most Linux developers will be familiar with the build systems and cross compiller stuff, but for others, like me who don't have a Linux box to install a cross compiler, etc.. this is a big stumbling block and deterrent to develop apps for the Zaurus. Also, when developers write their apps. they want them to be easily ported and run on as many environments as possible. The Zaurus is just scary because it has so many distros that "apppear" to be incompatible with each other and when someone tries to decide whether to write apps for Zaurus and then finds that they have to code for one specific environment and lose a lot of potential users because they are using the other Zaurus distro. Packaging IPK files for Zauri is also so different for the various distros as in the format of the IPK file itself as well as the file locations for all the different distros. Since docs are lacking or way out of date, most potential Z app developers will just give up and continue to write windows apps.
One thing that attracted me to pdaXrom was that it had a preconfigured native gcc compiler that would allow me to quickly compile apps on my Zaurus directly as well as a live CD to boot off a PC to try out the environment and look and feel. It also has a VMWare image too so I can run it inside my Windows session. Sure it has many problems still, but we are talking about choice, this is kinda the type of choice I am looking for, choice of development environment. Don't forget this part of the choice equation. Ok, I finished ranting

[albertr]

IMHO, there's no such thing as perfect software

Everybody has its own tastes and preferences, so having multiply choices is a good thing. If something vanishes, something else will take its place.

-albertr

[CoreDump]

Unification of pdaX & OZ won't do good to our zauruses. Each of these projects has an idea that appeals to some part of the community. But I don't think that flame wars should emerge from these differences. You guys should collaborate and share the best of these roms.

P.S i wouldn't call any of the roms crappy, so the rule about having one cool thing instead of many crapy things doesn't work here. We got a lot of really cool Z roms!

The "unification" isn't about the ROM itself or the installed apps, just the method used to create said ROM. If pdaX decided to use OE from today it would still be the exact same ROM from a user POV.

With the difference that the user then could use the rather large and quite complete OZ feeds

[sashz]

If pdaX decided to use OE from today it would still be the exact same ROM from a user POV.

With the difference that the user then could use the rather large and quite complete OZ feeds 

I agree with about large complete OZ feed, just small note - 90% of this feed is unworkable because OE guys just add but not testing it - need for speed

From another side - if OE/OZ come to pdaXrom they will get soft-vfp, iwmmxt, "hacky optimization" etc - stuff which they want get in their future system, but which we use already about year.

[cycle_55]

If pdaX decided to use OE from today it would still be the exact same ROM from a user POV.

With the difference that the user then could use the rather large and quite complete OZ feeds 

I agree with about large complete OZ feed, just small note - 90% of this feed is unworkable because OE guys just add but not testing it - need for speed

From another side - if OE/OZ come to pdaXrom they will get soft-vfp, iwmmxt, "hacky optimization" etc - stuff which they want get in their future system, but which we use already about year.

Hi sashz

Even though I don't totally understand what is going on, i say do what feels right. Carry on the good work and don't apologize.

cycle_55

[lardman]

Couple of minor corrections:

For the sharprom, I see tons of people but are well hidden. for the pdaxrom, just 2? and for openembedded, less than 4? for cacko, 1 and he makes his own software and redistributes a few and is also compatible with sharprom.

I just asked on IRC and after a quick look at the commit logs: we have about 46 active developers, of whom 10 are very active, 20 are active and the rest commit infrequently. The numbers here are also slightly skewed, because many of the active group are also commiting patches, etc., from the bugtracker - so in fact we have more developers than these figures would indicate.

Now I should note that not all of these people are working on the Zaurus/ipaq/simpad. A rough guess is that 40% do palmtop work, 40% do nslu, and 20% do misc other devices. There are of course also a number of commercial companies doing work, but I've no idea of numbers, etc.

To summarise: more than 4 doing Zaurus/palmtop work

I agree with about large complete OZ feed, just small note - 90% of this feed is unworkable because OE guys just add but not testing it - need for speed

I'd be surprised if some percentage of the available packages didn't have some issues, though I'd be very suprised if it were actually 90% (I'm just trying to think of a way of working it out, without having to install and test every package myself.) Even if only 10% did work, that would give us 10% * ~6000 packages ~= 600 packages, which is still a fair few .

I'll do some thinking about how to work out the exact number if anyone's interested.


Si

[adf]

(To skip back a little)
The "one way to do things" "one best product" kinda reminds me (though I have no personal experience ) of soviet auto manufacture. As I recall, the lada wasn't really a hit, it merely prospered for lack of competition.

[karlto]

(To skip back a little)
The "one way to do things" "one best product" kinda reminds me (though I have no personal experience ) of soviet auto manufacture.  As I recall, the lada wasn't really a hit, it merely prospered for lack of competition.

Lada? My Zaurus is a Cadillac!
(especially with pdaXrom installed)

[sashz]

I agree with about large complete OZ feed, just small note - 90% of this feed is unworkable because OE guys just add but not testing it - need for speed

I'd be surprised if some percentage of the available packages didn't have some issues, though I'd be very suprised if it were actually 90% (I'm just trying to think of a way of working it out, without having to install and test every package myself.) Even if only 10% did work, that would give us 10% * ~6000 packages ~= 600 packages, which is still a fair few .

I'll do some thinking about how to work out the exact number if anyone's interested.
Si

lil more fire - i found in OE package directory only 1325 applications, where 6000 ? Maybe you count split applications into few packages, locale packages, man and documentation ?

[clofland]

...but I'd gladly give up chosing between 10 crappy things when I could just take the one really cool thing out there.

What is crap and what is cool? Who willl judge?

Good point. I do not think Sash was being nasty here guys! He was just making the point that that is part of the problem.

Mickeyl would rather have 1 nice IMAP client, than 10 crappy ones, but Sash was pointing out that the problem is that everyone has a different idea of what the 1 non-crappy IMAP client is.

It is a problem. It isn't anybody's fault, it is just the way things are.

Remember, be nice, and give everyone the benefit of the doubt here. Text is a poor way to communicate, especially when we are from different countries and don't all use English in the same way.

[clofland]

Here is my thought on all of this. Am I crazy???

If someone REALLY thinks that pdaXrom should be moved over to the OE system, then it seems to me, since all of the source is there in the open, they should just start doing it. I mean, why not one more splinter? Take pdaXrom, port it to OE, make it work, prove to all of us that it works better, and if it really does do everything we want, then we'll come over to your camp.

Personally, I don't want to see Sash stop his current direction. He is pushing toward a production release a of a system that a lot of us can (and already do) make daily use of on a PDA that is starting to show its age. (Mine is anyway. )

Sash does not want to use OE, and there really is no reason to push him to. pdaXrom has this problem you see. It works for us. It does what we want it to. So we really do not want it to change. We don't want to slow down on the development process in order to make infrastructure changes either. Honestly, in five years, most of us will have a new PDA, so delaying development of our working platform for infrastructure changes is not very appealing when what we have works so well for us.

As for OE. I think Mickey's goal is to have a system that we can be using on our new PDA's in five years too. I just wish that some of the OE guys could sit back and let that be their vindication, rather poking at us pdaXrom users now.

(NOTE: "works for us" == "works for us", not "is better" and not "yours does not work", just "works for us")

(NOTE2: I am not implying that pdaXrom has no future, just that most of the pdaXrom USERS are focused on the present. This can be seen by the fact that so many of us flash the latest RC or BETA as soon as it arrives and then proceed to use it daily in production.)