Quality Team For OZ

[Anonymous]

Was just reading in /. about the proposed Quality team for KDE.  Maybe OZ too should adapt that...

[Mickeyl]

This is a great idea. Hey we would really like to, but at the moment no one is working on OZ so who is supposed to fix the bugs the Q/A team reports?  :?

[tamarian]

I thought OZ was discontinued?

[Mickeyl]

Who told you that? OpenZaurus as a linux distribution is not discontinued - it never was. The OpenZaurus build system is discontinued. OpenZaurus development is halted because of working on the higher priority task \'build system migration\'. The next planned OZ release will be OZ 3.5.1 made out of the OpenEmbedded buildroot which is far more flexible and capable compared to our old system.

The build system migration is nearly finished now but since most of the  work was done by nearly two people ( see http://openembedded.bkbits.net:8080/packag...?nav=index.html ) we\'re pretty exhausted and motivation to start doing the now necessary integration work (integrate the device specifics to make OpenEmbedded capable of doing OpenZaurus builds) is low... frankly, rather not existing.

If the workload doesn\'t get distributed, it isn\'t fun. And if it isn\'t fun, it\'s not worth doing it. This is the Open Source credo isn\'t it?

If a bunch of people would start to do that integration work we could get into release mode in a few weeks... however since that didn\'t happen in the last months it could also take years to get into release mode again.

The community finally has to realize that two people are not enough to do such a huge task. I have realized it and I will no longer risk my profession and my health only for this hobby. Well, I\'m getting tired preaching the ole\' sermon again and again... read the OpenZaurus website, the OOO newsletters and openzaurus-devel@lists.sf.net and see how often we called for helping hands... without success.

[Anonymous]

Who told you that? OpenZaurus as a linux distribution is not discontinued - it never was. The OpenZaurus build system is discontinued. OpenZaurus development is halted because of working on the higher priority task \'build system migration\'. The next planned OZ release will be OZ 3.5.1 made out of the OpenEmbedded buildroot which is far more flexible and capable compared to our old system.

The build system migration is nearly finished now but since all work was done by two people ( see http://openembedded.bkbits.net:8080/packag...?nav=index.html ) we\'re pretty exhausted and motivation to start doing the now necessary integration work (integrate the device specifics to make OpenEmbedded capable of doing OpenZaurus builds) is low... frankly, rather not existing.

If a bunch of people would start to do that integration work we could get into release mode in a few weeks... however since that didn\'t happen in the last months it could also take years to get into release mode again.

The community finally has to realize that two people are not enough to do such a huge task. I have realized it and I will no longer risk my profession and my health only for this hobby. Well, I\'m getting tired preaching the ole\' sermon again and again... read the OpenZaurus website and the OOO newsletters and see how often we called for helping hands... without success.

Hi Mickey,

Sorry, I misunderstood previous announcements to mean the team switched to a different project (open embedded) from scratch.  At least that\'s what the home page of OZ implied. Finding news about the project (s) is hidden among several project mailing lists, and every time I visit handhelds.org, I can\'t tell who\'s doing what, and what project belongs to what hardware.

And I see your point, yes, 2 is not enough for such work load.

I have seen you make previous requests for help, and I imagine most here and on the old dev board have read them too.  So, my guess is that there is awareness that there is a need for help, but maybe not enough awareness about what kind of help, what\'s involved, and who\'s qualified.

I\'ve tried before to install OZ, just to see if I can understand it, and try to see how it works, unfortunately, I\'m a new comer to Z and only have a C760, and all attempt at installing OZ unstable have failed, it won\'t even boot. So to this day, I have not used OZ, don\'t know what\'s in it, and how different it is to the other ROM\'s around.

[tamarian]

That was me, my cookies logged me out

[bluey]

Was just reading in /. about the proposed Quality team for KDE.  Maybe OZ too should adapt that...

There is no need of more people to find bugs, but to solve them.

I have seen you make previous requests for help, and I imagine most here and on the old dev board have read them too. So, my guess is that there is awareness that there is a need for help, but maybe not enough awareness about what kind of help, what\'s involved, and who\'s qualified.

This could be the reason to make people give up, or not even try.     Most people could help, probably, but doesn\'t have the time or knowledge to start, and ends up thinking they can\'t help. Is there any way to have a list of tasks which need to be done so people know what they can do, and understand the task so they can know how to start, what is necessary to work on it (cross compile env, just linux, etc), and stuff like that? If enough people started to work on the little tasks, Mickey and Kergoth could rest, just needing to guide the others on their tasks. Is this too much utopian?  

[Anonymous]

This is my take on OZ development.  There needs to be comprehensive documentation on how to get started.  A command by command step on how/where to download source, what command to build and what to get started would be of great help.  

Most people get turned off by the fact they have no clue (myself included) where to start.  Instead of writing a 1 page doc, if actual commands are given to
download
build
Test run the environment

That would go  a long way in enlisting new members.

Also, for each app, have a list of small todos so newbies can take a shot at it.  Also should have good doc on each app head as to how to build it, what package is needed to run it on a particular h/w, what needs to be fixed, what priorty etc.

The problem with OZ site is it is hardly updated and is skimpy on the details.  

No offense meant Mickey, I greatly appreciate what you guys are doing.  You guys are the only reason I\'m still sticking to Z.  But this is what I think is honestly missing.  It is not that people are dumb or anything, I know if we really put our mind to it, we can figure out how to get the source, do the test build and get started, but if all this info is already available, it makes it easy.

I guess it is just inertia!

[Anonymous]

I would agree with that.  Making it easier to apply one\'s interest and knowledge to oe and oz development would help get a lot of those who want to help involved.  I\'m sure a lot of those want to be devs really do feel uncomfortable and overwhelmed by the whole process.  From what I hear, setting up a build environment is itself enough of a task.

[lardman]

I agree that it\'s difficult to get started. I\'m quite happy to write docs as I go along (as I\'ve only just started).

Where should I place them?


Si

[Mickeyl]

For now I suggest using the openembedded.org Wiki. We can then have people review these and go through an iterative enhancement process before we add them to the official OpenEmbedded manuals.

[bluey]

That would be great, since I believe Mickey and Kergoth don\'t have the time to update the existing documentation.

Maybe a section here on the ZUG docs, or on the OE wiki? What would be better?

Do you think there would be a need for a separate section for OE here?

[lardman]

OE wiki\'s fine, I was just loath to tread on people\'s toes.

What do you reckon then, make a link from the GettingStarted page, or just stick it in there with it?


Si

[Richjn]

O.K
I will join with larman and add my bits of knowledge as I fumble my way through the learning curve. I have held back from posting stuff (other than a small version change) as I didn\'t feel comfortable posting info that may be wrong / mislead others but overall I think its worth the risk to try and give this great project a helping hand and give others a well earned rest. Maybe Mickey and Kergoth can remove the worst of the incorrect stuff. (mine probably)

Richard

[Mickeyl]

Cool - appreciating this. I\'d say just modify the GettingStarted page and add additional pages as you like to break it in smaller and possibly better comprehensable chunks - your take on it is as much worth as ours. That\'s what I like in the WikiWiki system. If something is vague or even incorrect, it is not set in stone - we can just improve the documentation without having to deal with access rights, versioning or whatever complicated procedures - this is spontaneous collaboration.