super-tux-devel

[Super-tux-devel] SuperTux design

[<cos...@ho...>]

Hey, I'm new here and never participated in an open source project before 
but I feel like I want to help. I don't even know if the right way of using 
a mailing list is sending an email to this address but I think it's worth a 
shot.

Anyways I wanted to say I think that a complete design of SuperTux should be 
the first aim of the project first, with defined milestones and such. After 
the design document is complete the game development should try to stick to 
it even if a 'neat feature' is tought afterwards. I can help with design it 
if that position is aviable.

After the design coding should start and prelimirary multimedia just to test 
the code should be made. I can help with C++ coding, not python tought I 
think I can learn python in little time. After the game has all the code 
complete, real levels, graphics, sound, music and levels should be included, 
as defined in the design document. I can help with graphics but I'm not good 
at sounds and music(at least not yet!).

After the game is complete as the design document says it should, then new 
ideas of any kind should be voted yes or no to include them.

Well that's at least my idea of how the game should be made. It just seems 
to me like the project is rather deorganized and some persons don't know how 
to do things coz it hasn't been defined yet.

Also I don't know if who are the project administrators and who is the one 
who has the vision of what should SuperTux be like. I'm interested in that 
to talk with that person about the game design.

-Coz

_________________________________________________________________
MSN Amor: busca tu ½ naranja http://latam.msn.com/amor/

Re: [Super-tux-devel] SuperTux design

[Ingo R. <gr...@gm...>]

Eduardo Hern=C3=A1ndez <cos...@ho...> writes:

> Anyways I wanted to say I think that a complete design of SuperTux
> should be the first aim of the project first, with defined
> milestones and such. After the design document is complete the game
> development should try to stick to it even if a 'neat feature' is
> tought afterwards. I can help with design it if that position is
> aviable.

Goals for Milestone1 are documented at:

 * http://super-tux.sourceforge.net/milestone1/

Planing much further then milestone1 however is relativly useless,
since one first needs to know how milestone1 turned out, what worked
and what didn't, etc.

> Well that's at least my idea of how the game should be made. It just
> seems to me like the project is rather deorganized and some persons
> don't know how to do things coz it hasn't been defined yet.

I don't know from where you get that expression, sure its not all
perfect, but compared to a lot of other projects around its rather
well organized. And anyway, this is a open source project mainly
developed for fun, so no need to have a full fledged design document
to collect publisher money. If you want to flesh out the current
milestone1 docu and add stuff, go ahead.

> Also I don't know if who are the project administrators and who is
> the one who has the vision of what should SuperTux be like. I'm
> interested in that to talk with that person about the game design.

Join IRC, #supertux, irc.freenode.net, most people involved with the
game hang out there.

--=20
WWW:      http://pingus.seul.org/~grumbel/=20
JabberID: gr...@ja...=20
ICQ:      59461927

Re: [Super-tux-devel] SuperTux design

[Michael G. <mi...@ge...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


|
|>Well that's at least my idea of how the game should be made. It just
|>seems to me like the project is rather deorganized and some persons
|>don't know how to do things coz it hasn't been defined yet.
|
|
| I don't know from where you get that expression, sure its not all
| perfect, but compared to a lot of other projects around its rather
| well organized. And anyway, this is a open source project mainly
| developed for fun, so no need to have a full fledged design document
| to collect publisher money. If you want to flesh out the current
| milestone1 docu and add stuff, go ahead.
|

Just my $.02 - It would be nice if the TODO list on the website had some
indication of what people are working on vs. what's unclaimed.  For
example, a couple of people have sent in small patches that were
rejected because the core team is already doing that stuff.  It's just a
little hard for people who want to be tangentially involved (for whom I
think gotm is a great project) to be able to pick up 1-2 hour tasks
without worrying about stepping on people's toes.

Maybe this goes on on IRC.

- --Mike
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFAWkSMrJDiig1GCW0RAhjOAKCJ4mnzMiBDSyFu/cHcp/cZC46zPwCffUte
xFbenNFJcT00apjJZ5NaViE=
=QgA0
-----END PGP SIGNATURE-----

Re: [Super-tux-devel] SuperTux design

[Ingo R. <gr...@gm...>]

Michael George <mi...@ge...> writes:

> Just my $.02 - It would be nice if the TODO list on the website had
> some indication of what people are working on vs. what's unclaimed.
> For example, a couple of people have sent in small patches that were
> rejected because the core team is already doing that stuff. 

The problem is that small stuff would need more work to track it, than
to fix it. If I need ten minutes to maintain the status of an item on
a webpage and only five to fix it, its not worth to track the issue in
the first place. Beside that the tracking of items would very quickly
get out of sync with what is going on in CVS unless one takes extrem
causions, so it wouldn't really help all that much. Beside that there
is also the problem with people starting something, but never getting
it into a 'patch submit' ready stage, which would result in items
being marked has 'handled' or whatever, but in the end not getting
done at all.

> Maybe this goes on on IRC.

IRC is really the best to way coordinate small items.

-- 
WWW:      http://pingus.seul.org/~grumbel/ 
JabberID: gr...@ja... 
ICQ:      59461927

Re: [Super-tux-devel] SuperTux design

[Tobias <tob...@gm...>]

Hi and welcome!

Am Do, den 18.03.2004 schrieb Eduardo Hernández um 18:32:
> Hey, I'm new here and never participated in an open source project before 
> but I feel like I want to help. I don't even know if the right way of using 
> a mailing list is sending an email to this address but I think it's worth a 
> shot.

Yeah, it is the right way.

> Anyways I wanted to say I think that a complete design of SuperTux should be 
> the first aim of the project first, with defined milestones and such. After 
> the design document is complete the game development should try to stick to 
> it even if a 'neat feature' is tought afterwards. I can help with design it 
> if that position is aviable.

I have seen a lot projects coming and going, which had nice design
documents with every little detail described.
A few of them actually fullfilled their wishes but the most of these
projects can be seen as dead ones.

SuperTux's development began with one person and that's Bill Kendrick.
I can't even say if he had a design idea, looking at the 0.0.4 sources
I think he just had much time and coded to see a game running. (Yes, it
was very basic, but many people including me liked it)

Many projects including Linux began with a code base like this
and it seems to be one of the ways to attract more developers.

I was attracted by the 'playable' 0.0.4 and wanted to fix some bugs and
extend the engine. This was the beginning of a new SuperTux era, since
Bill Kendrick moved the project to SourceForge after a few bullying
mailes from my side.

Until 0.0.6 a rather complete code rewrite took place and the
communication in the team was and IMHO still is very good.
And as it seems 0.0.6 attracted YOU! Had we written little nice
documents nobody would know about the project, besides two to three
people, and it was probably already dying, because the motivation to
release new versions wasn't that exciting.

> After the design coding should start and prelimirary multimedia just to test 
> the code should be made. I can help with C++ coding, not python tought I 
> think I can learn python in little time. After the game has all the code 
> complete, real levels, graphics, sound, music and levels should be included, 
> as defined in the design document. I can help with graphics but I'm not good 
> at sounds and music(at least not yet!).

SuperTux needs artists in the first place. I see a serious danger in too
many people working on a quite little free software project like this
one. It's no danger for the project, but it can easily happen that two
programmes pick the same code or want to fix the same bug.
Therefore the mailing-list is seen as central point for ALL patches and
bug-fixes.

> After the game is complete as the design document says it should, then new 
> ideas of any kind should be voted yes or no to include them.

The funny times when we can implement features easily, will be after
0.1. Voting on every little feature would slow down the project.

> Well that's at least my idea of how the game should be made. It just seems 
> to me like the project is rather deorganized and some persons don't know how 
> to do things coz it hasn't been defined yet.

This project is organized to 70% on this mailing-list, so you may want
to read the mailing-list archieve on sourceforge. Other 20% take place
in IRC and 10% on the project's homepage, which can be seen as a place
for collecting ideas.

> Also I don't know if who are the project administrators and who is the one 
> who has the vision of what should SuperTux be like. I'm interested in that 
> to talk with that person about the game design.

This 'visions' have been discussed with the 'community' in the so called
IRC meetings. I think you can find the logs in the mailing-list
archieves.
IMHO a free software project shouldn't only live from the aims of a 1.0
version, but it should constantly make little steps towards new versions
and clean up it's API as long as it can be consired as stable.

Greetz...

Tobias Gläßer