organizing blender2indigo development+collaboration attempt

[Big Fan]

edited to remove all

[u3dreal]

Hi all,

well i'm a bit sad nobody mentioned me although i did alot of work
for the lastest 0.6 version of the exporter. Especially a better Gui.
And some changes to make the export phasical correct.
Anybody can see my intention by taking a look at the
max2indigo exporter .. my intention is to make it not
just a toy but a tool to "work" with..

But after all i just started with the blender exporter but i wouldn't see myself as the creator. At least i announced and tried to organize
my coding with the others. And i'm still in contact with zuegs.
Where's leope after all.

I see zuegs has already begun with the idea we came up with.
The material editor... and i hope we can implemnet it into
the 0.7 version.

I already have most of the code ready for
0.7 test2 .. For sss it's possible to enter the values form the list
ono posted right from inside Blender.

But i want to wait to make some real implemenation
when i got the whole picture of what 0.7 final will look like.

So i don't see the need to have a "work" version for 0.7 as many
changes will come for sure until we reach final.

I would see myself somewhere between group 3 and 4 and with time
and improving my skills i will hopefully make it to group 3

I would be happy to be part of the 0.7 final exporter team.
But thats up to you to decide.

Anyway i still have my max2indigo script and i'm so happy i didn't make
it puplic as the same would have happend there. There will be
an update with many improvements for 0.6 final.
Yes also the fix radiance requested..
0.7 final will have to wait. but then again it will be production ready.
at least i hope....

The idea of a restricted developer forum might also be a good solution.
For Kerkythea this works perfect and ppl that become worthy to be a developer are much more dedicated as they are treated different
due to there contribution.....

cheers to all may the force be with you.

u3dreal

[Big Fan]

edited to remove all

[Jananton]

To me it seems unlikely that we will make use of the renderer api or nodes in the forseeable future.
The difficulty is that blenders gui doesnt really lend itself to tapping into it to set materials for indigo.

Well, the obvious question is, will an ordinairy user, someone without any real knowledge of python, nodes in blender, and the ever evolving settings of indigo materials, be inclined to download a special build of blender to use a special exporting script to a renderer called indigo, or not. ( I say this because there is a blender build that does export the node component for users to use in scripts ). I personally think not, to be honest.

If one agrees to the above, then, as you said, the idea of zuegs to make material choices available via the exporter script is a very agreable solution for now.

After this little response to Big Fan's post that's maybe a little bit off topic in this thread, I have to vent some of my thoughts about the tendensies that evolve in this discussion in my opinion.

First I would state, as Wedge did so eloquently, that I don't want to offend anybody who has his or her own thoughts about how to channellise all attempts on scripting for the next version of indigo, but...

Of course there are a lot of people busy with develloping scripts and solutions to get all the nuts and bolts working that are available in the current test release of indigo. And of course, since all is fluid at the moment, like the tags used in indigo, the new version of blender upcoming, and the individuality of the scripters, there is a melting pot of code presented to the unexpecting visitor of this forum.
I agree completely, but as Phil stated a few messages ago, it's only 1300 lines of python script.

Creating a cvs server for that is like shooting with a cannon on a fly ihmo. I, for instance, got involved when I read someone (u3dreal) was trying to launch indigo from within blender and got stopped by the windows phenomenon to have spaces in folder names. That combined with a comment from another thread that meshlights and other light source(s) where very hard to mix got me involved.

If I had to first download and install cvs software to be able to get my improvements to you, I probably wouldn't have bothered and kept the working script for myself. Not because I'm a cvs hater, but because I've got experience with this situation, where a cvs server is set up by enthousiast people develloping something only to see after time that nobody is available to maintain it all, resulting finally in loss of effort of a lot of people who have contributed. I rather put my eggs in the basket here, so to say...

There's also a positive note in the way the scripting devellopment is going at the moment. There are no boundaries or obstacles for anyone who want to contribute. Ok, it's a little bit harder for those who only want to download something working without hassle, but what can you expect if people are develloping something for something else that's changing as we speak.

And finally - said by more that contributed to the scripting efford - for most, as for me, it's a hobby not a job. If I read that using cvs, or other means of directing scripting evolvement with as result quicker solutions to problems, for end users, my toes start to curl. It's a hobby, not a job, and pushing for more and more certainly isn't motivating...

Well, there's more I could say on this subject, but afaik this post is long enough already...

Greetings from nighttime Amsterdam,

Jan

[oodmb]

u3dreal, thanks for joining the thread, i have been waiting a while for this.

i beleive there should be a seperate developer forum, one thats not as accessable from the blender forum on indigo, possibly something the general public cant easily without specificaly wanting to find it. However, i do not beleive that this forum should be restricted access at all. EVERYBODY should be able to post to this forum (with a free easy to get account of course). however, i realy do like the classification list Big Fan has put down even though i dont realy seem to fall anywhere specific on it. what then can be done is that user acounts on this forum can be ranked in a system similar to this, which states how much that person can contribute by himself without special admin permisions the forum would have an administrator, to handle the wiki and do the final upload of official betta candidates and final releases. all users above a class 6 or 5 should be allowed to edit wiki with permisions from administrators. all users class 4 or above should have the general right to edit wiki docs. Class should be determined not only coding ability but exporter knowledge, general technical knowlege and amount of posts(gauge of seriousness). i have yet to think of a system for choosing class specificaly, possibly a user starts out "classless" with only posting rights untill that user makes a plea to the (or a) class administrator that he has certain abilities.

btw, the way i am suggesting offers no boundries to thoes who want to contribute. if one were to download the current betta, and decide to do a little code, all one would have to do is just that, not even having an acount. all he would have to do is write the proposed code, upload it and bam, done his work will be considered into being taken into the official release and if chosen, he could have a hand in implementing the code into the current release or somebody else could do it, whatever.

this might sound complicated, but in order to keep random people from changing the official version while still allowing rights to edit wiki documents, a system like this is necisary.

[Jananton]

Hi all,

well i'm a bit sad nobody mentioned me although i did alot of work
for the lastest 0.6 version of the exporter. Especially a better Gui.

Yes, you did, but everybody seems to be rather a manager than anything else these days...

Greetings from thankfull Amsterdam,

Jan

[Big Fan]

edited to remove all

[oodmb]

ok, hot idea! much less complicated then my ideas, although alowing input from class 5 and class 6 users should not be halted, just keeping them from editing important documents directly. like if somebody were to upload documentation, somebody from a higher class would update it.

as all this should optimaly be decided democraticaly, we should probably post a poll for organizational ideas, please if you post any more ideas in this forum, dont say anything against organization, We already have plenty of those.

Everybody should submit there own full length well thaught out idea about how organization should happen, and then i'll put up a poll about it, and then once the structure is decided we can choose the admins.

[OnoSendai]

Some thoughts:
I think it would be a good idea to have a new blender developer's forum, it could be a new (sub)forum just below the current blender forum, in the exporters section. This way we can separate questions about how to use the exporters from development discussion.

About CVS/SVN (source control):
Source control is nice, but as someone pointed out, this isn't really that large a project. However, once we get the new server, it should be pretty easy to run SVN off that. hmm.

I'm thinking we should have an official blender exporter script, for each version of Indigo. Then we would choose perhaps 2 people to be the chief maintainers. These people would be responsible for merging changes into the official version.

So right now, for example, we would choose an official 0.6 exporter. This would be the one available in the download system.
We would also choose an official 0.7 (development) exporter script. When it is done, and Indigo 0.7 final is done, this is the one that would become the 'official' exporter script, and put in the download section.

This doesn't stop people from working on their own script if they want, but I think it will help in channeling work into one really good exporter.

[oodmb]

you put my ideas into words (far better words at least) exactly

[crsrma]

Wow, that's wicked good news... I just counted 10 exporter versions in my menu.

I'm loving all the effort in the 0.6 exporter, especially balancing of meshlights in the GUI. Thanks much, devs!

[leope]

Where's leope after all.

I've explained my way of grab(or create)/develop/release a script in a post on this forum and i don't like team work (too time dependent, too much compromises and my interest is variable).
I think i don't full fit into the 3rd category cause my skill is very low is some areas, so it's true i can write an exporter from 0 but it can't reach a good level quality and completeness (eg. blenderqueen).
I paused my coding at the exporter script for some reasons:
1. No time.
2. Indigo is too slow for my free time and i want sleep at night. (maybe when i buy a quad core...)
2.1 I'm too slow.
3. No native Linux version.
4. 3d CG is an hobbie.
5. I prefer a blender2Indigo coded into blender (like kerkythea, povray, yafray) but my c++ skill is too low.

So "i'm waiting for":
1. new hardware
2. new exporter written by a more expert coder than me

Thanks
bye leope

[Wedge]

Hey Leope,

It is good to see you.

I understand your situation.

I want to thank you and Zuegs for your work on the exporter.

It was both your works that I first saw when I found out about Indigo and started looking at the forums. It inspired me to take part and see what I could add to the exporter.

Thanks again!