Celestia development volunteers?

[hank]

Hank, I'm sorry but if you wanted it THAT focused, there are simple solutions:

- You write it clearly when opening the thread (see this thread, and despite I was in the community for only 2 months, no one break the rule)
- You ask the mod for special attention on your thread in case of (note that this second point is really optional, is point one is correctly done, no problemo.

That might work for the specific case of a simple sign-up list. But in general, no. The forum threads have an inherently sequential, conversational structure that does not allow the kind of integration and organization of information that is possible with a wiki.

- Hank

[hank]

could you please explain, why you have now added plenty of real coding tasks to that particular Wiki, although NOONE of your volunteers list is able to code properly and all but Cham lack astrophysical expertise??

I'm experimenting with a simple wiki scheme for coordinating development activities. I'm using the task list from the current Celestia Road Map just as an example. Not all development tasks will require programming skills or astrophysics expertise. But I believe one of the volunteers has indicated some experience with C++ and OpenGL, and of course I'm hoping to eventually attract additional volunteers.

I thought Chris L. did install a Wiki precisely for such real developer matters in SF. So all this is getting increasingly confusing!

I'm using the open WikiSpace wiki for now because the SourceForge wiki access is too restrictive. We can move parts over to the SourceForge wiki if it makes sense to do so.

Are you suggesting that we give up the dev list at the same time? Why?

I haven't suggested that.

It has worked well for years (provided we do some actual coding work ;-) ). The present lame development will certainly NOT be improved by starting a multitude of confusing Wikis ;-)

I think there's reason to believe that some slightly more organized project coordination could be helpful in recruiting new developers. I doubt that this simple wiki approach would be confusing for them.

- Hank

[t00fri]

I think there's reason to believe that some slightly more organized project coordination could be helpful in recruiting new developers. I doubt that this simple wiki approach would be confusing for them.

Right now, there are in PARALLEL:

• SF Bugtracker (used also by Chris) to document the status of fixes for 1.5.0
• dev list
• Chris' new developer Wiki at SF (dormant...)
• Forum developer threads
• your "open" Wiki experiments

So right now there is about ONE dev documentation medium per regular developer If this is NOT confusing for new developers I don't know what could be!

-------------------------------------------------
I understand that it would be best that ALL the developers with proven (not claimed) coding knowledge would rather work jointly at this time to complete the tasks still missing for 1.5.0 final!

AS FAST AS POSSIBLE...

We have a small but knowledgable group of proper developers AND beta testers that are able to complete the required tasks pretty well. A main positive aspect is that they ALL know what has to be done!
--------------------------------------------------

I think at this time your ill-defined "Wiki experiments" are confusing and do not help in this direction. They rather contribute to defocus our actions.

If you would help fixing bugs, things would even proceed faster

Bye Fridger

PS: As you can see from today's dev list, also Chris L. is NOT making use of your new "open developer" Wiki idea...
-------------------------------------------
Today I committed fixes for three different bugs on the SourceForge tracker:

1745705: invalid(?) body frame crashes Celestia
http://sourceforge.net/tracker/index.ph ... tid=121302
Fixed by making the ssc file loader check for circular reference frames

1738359: Barycenter not included in URL path
http://sourceforge.net/tracker/index.ph ... tid=121302
Bug description wasn't quite accurate, but there was a serious bug in the cel URL code that caused star names to be omitted in some URLs. I also added code to store location names in URLs; prior to the fix, location names were omitted completely from URLs.

1733395: ReferencePoint placed in wrong location
http://sourceforge.net/tracker/index.ph ... tid=121302
This was a serious bug with reference points; the OrbitFrame for a ReferencePoint definition in an ssc file was completely ignored.

Thanks to Selden for the detailed bug reports. I've marked the bugs fixed, but left them open for someone else to double check the fixes.

--Chris

[rthorvald]

[I think there's reason to believe that some slightly more organized project coordination could be helpful in recruiting new developers. I doubt that this simple wiki approach would be confusing for them.

I think at this time your ill-defined "Wiki experiments" are confusing and do not help in this direction. They rather contribute to defocus our actions.

I think that the best introduction for potential new contributors would be a simple, static webpage on the shatters.net Celestia main site that lists the developers resources in use (the forum, mailing list, wiki, etc). This page would contain links with a short explanation of each item on it, updated as needed. No confusion there... It is the most obvious place to look for such basic info.

- rthorvald

[Fenerit]

I agree with you Runar here as well as for what you have said in the second-last post. Seem to me that Celestia to be in the web as the stars in the program or in the universe: sparse across...

[hank]

I understand that it would be best that ALL the developers with proven (not claimed) coding knowledge would rather work jointly at this time to complete the tasks still missing for 1.5.0 final!

AS FAST AS POSSIBLE...

We have a small but knowledgable group of proper developers AND beta testers that are able to complete the required tasks pretty well. A main positive aspect is that they ALL know what has to be done!
--------------------------------------------------

I think at this time your ill-defined "Wiki experiments" are confusing and do not help in this direction. They rather contribute to defocus our actions.

If you would help fixing bugs, things would even proceed faster ;-)

You are free to do what you think best, Fridger, as am I.

And I happen to think that recruiting additional developers and development assistants (such as testers), and establishing an efficient, sustainable development process, is very important to the long-term future of Celestia.

- Hank

[BobHegwood]

You are free to do what you think best, Fridger, as am I.

And I happen to think that recruiting additional developers and development assistants (such as testers), and establishing an efficient, sustainable development process, is very important to the long-term future of Celestia.

- Hank

If I may AGREE with Mr. Hank...

As you all know, I don't know BEANS about coding, texture manipulation,
scaling, various scientific procedures, etc...

However, I have STILL managed to add SOMETHING to the progress of
Celestia, have I not?

EDIT: Just added my 3rd page to the Celestia Wiki, so you guys let
me know if I'm doing anything you don't agree with.

Thanks again, Bob

[Paolo]

http://en.wikibooks.org/wiki/Celestia

http://celestians.wikispaces.com/

http://celestia.wiki.sourceforge.net/

http://celestia.teyssier.org/

http://www.lns.cornell.edu/~seb/celestia/

http://sourceforge.net/tracker/?group_i ... tid=121302

http://sourceforge.net/mailarchive/foru ... developers

http://www.celestiaproject.net/forum/viewforum.php?f=10

?????

A little bit of confusion and redundancy..... Which is the tool that the new developer volunteers should use? Please don't say ALL....

IMHO as I've said a long time ago, 1 wiki, 1 forum (eventually with sub forums) are enough. Even the mailing should be deprecated.
An updated Doxygen documentation should be very useful and appreciated too.

Develpers should concentrate on code, info and data not on documentation tools.

[BobHegwood]

http://en.wikibooks.org/wiki/Celestia

This is the one I've been working on Paolo... Since I don't know
anything about how they work, I have simply been editing
and PREVIEWING what I have done before I post the changes.

Seems to work for me. I also have trouble trying to view the HELP
pages while I'm working so, again, I just do what I think will work.
Then, I preview the changes before saving.

Good Luck, Bob

[hank]

?????

A little bit of confusion and redundancy..... Which is the tool that the new developer volunteers should use? Please don't say ALL....
...
Develpers should concentrate on code, info and data not on documentation tools.

Hi Paolo.

I've suggested using the public WikiSpace (http://celestians.wikispaces.com/) to coordinate development volunteer activity, but it hasn't really got off the ground yet.

The SourceForge wiki also hasn't been used much yet. Currently it is restricted to contributions by official developers, which seems too limiting.

The Celestia WikiBook is intended to provide information about using Celestia for end users. There is currently some development information there, but using it for development coordination doesn't really seem consistent with the WikiBook charter.

- Hank

[Paolo]

Bob, Hank,
I don't want absolutely to criticize the celestians wikispace initiative!

I've Used wikispaces and I think that is a useful tool. Allows discussions on single pages. Works like the standard wiki and has a good editor.
The real great advantage is that is "open". So also the non-official-developers can contribute and this is a great step against the extreme closure of the Celestia's developers environment.

IMHO I think that:
- in the celestia wikibook the development related content should be excerpted and moved to the new tool
- the Selden's and Christophe's websites should be translated almost entirely in this new tool.
- The Wiki on sourceforge is useless because redundant.
- The developer's mailing list is anachronistic.

Bug tracking is a real issue, but should be solved using the bugs forum.

I would like to see only better coordination of development efforts. That is the thing that in those five years is lacking more to all the Celestia project.

[Paolo]

Another question is jumping in my mind continuously in this period.

How the celestians wiki should be integrated in the current development process?

How the official developers is supposed that will interact with this tool?

In other words if we will update the roadmap, we will discuss about the implementation of the new features and so... how are we assured that those things will meet the current deveolpment process?

The official developers will use the celestians wiki?

[chris]

Another question is jumping in my mind continuously in this period.

How the celestians wiki should be integrated in the current development process?

How the official developers is supposed that will interact with this tool?

In other words if we will update the roadmap, we will discuss about the implementation of the new features and so... how are we assured that those things will meet the current deveolpment process?

The official developers will use the celestians wiki?

That hasn't been decided yet. I remain most impressed by the Trac as used on Assembla. Forum user mjoubert has set up a test space for Celestia on assembla.com as described in this forum thread:

http://www.celestiaproject.net/forum/viewtopic.php?t=11586

From what I've seen, it offers the best integration of version control (via Subversion), bug tracking, and developer wiki.

The downside is that we'd have to move Celestia from SourceForge to Assembla. That's feasible, but relative to SourceForge, Assembla is new and unproven.

--Chris

[chris]

Bob, Hank,
I don't want absolutely to criticize the celestians wikispace initiative!

I've Used wikispaces and I think that is a useful tool. Allows discussions on single pages. Works like the standard wiki and has a good editor.
The real great advantage is that is "open". So also the non-official-developers can contribute and this is a great step against the extreme closure of the Celestia's developers environment.

IMHO I think that:
- in the celestia wikibook the development related content should be excerpted and moved to the new tool

Agreed.

- the Selden's and Christophe's websites should be translated almost entirely in this new tool.

The content of Selden's website overlaps mostly with the WikiBook. It's largely up to him whether he wants to move it to the WikiBook.

- The Wiki on sourceforge is useless because redundant.

Certainly only one developer wiki is required: either the Celestians WikiSpace or the SourceForge one. I created the one on SourceForge very recently as an experiment.

- The developer's mailing list is anachronistic.

Disagree completely. The developer's mailing list will remain. It's a useful way to communicate informally with other Celestia developers.

Bug tracking is a real issue, but should be solved using the bugs forum.

Disagree. The bug tracker on SourceForge is limited, but provides a better means of tracking bugs than a forum, where threads tend to quickly diverge from the original topic.

I would like to see only better coordination of development efforts.

Yes. This will happen immediately after 1.5.0. A few certainties at this point:

- We will move from CVS to Subversion
- Developer documentation will be moved to a single Wiki (either on SourceForge, WikiSpaces, or Assembla)
- Work on a cross-platform UI will begin in earnest
- We will make it easier for Windows developers to get started. I already have updated libraries, an up to date and working VS2005 project file, as well as updated Makefiles for command line builds.

--Chris

[Paolo]

The Trac system seems good to me.
But do you really need it? Or do you need only a better bug traking system?

[hank]

Another question is jumping in my mind continuously in this period.

How the celestians wiki should be integrated in the current development process?

How the official developers is supposed that will interact with this tool?

In other words if we will update the roadmap, we will discuss about the implementation of the new features and so... how are we assured that those things will meet the current deveolpment process?

The official developers will use the celestians wiki?

To my thinking, the big advantage of the celestians wiki at present is the very low startup and overhead costs. It could be a big improvement on the forum for collecting and organizing development information and coordinating expanded development assistance activity. To some extent we're improvising here, so the flexibility of the wiki is useful. Eventually we can move to something more formal if and when it seems productive to do so. Right now the official developers are mostly focused on 1.5.0 and it would not be helpful to disrupt the processes they're currently using. In the meantime we can experiment with the wiki.

- Hank

[Paolo]

To my thinking, the big advantage of the celestians wiki at present is the very low startup and overhead costs. It could be a big improvement on the forum for collecting and organizing development information and coordinating expanded development assistance activity. To some extent we're improvising here, so the flexibility of the wiki is useful. Eventually we can move to something more formal if and when it seems productive to do so. Right now the official developers are mostly focused on 1.5.0 and it would not be helpful to disrupt the processes they're currently using. In the meantime we can experiment with the wiki.

- Hank

I'm experimenting a bit (hopefully without doing disasters), in any case in a future moving the content to another wiki will be easy.

[chrisr]

Hey, I've wanted to contribute to celestia since i learned about it. I can only program (pretty well) in Matlab. Is there any use for matlab coders?

[selden]

chrisr,

One thing Celestia is lacking is globular clusters. They can be faked by using Galaxy objects and there has been some discussion of adding a special GlobularCluster class which does something similar in an official way.

However, a globular cluster also can be simulated by an STC catalog containing a spherical distribution of several thousand Stars which have their positions (and stellar classes, etc) defined by using appropriate random number distributions. Maybe you could give that a try?

[ANDREA]

Yes, chrisr, I agree with Selden, a globular cluster generator would be very appreciated in Celestia.
Rassilon made one a lot of time ago, but it gave some problem, because the final shape was never spherical.
Here the link to the discussion and the related files, source code included:
http://shatters.net/forum/viewtopic.php?f=4&t=1320&hilit=globular+cluster+generator
Thank you for all what you'll do.
Bye

Andrea