SourceForge this beast (Re: call for volunteers)

[Daniel Morissette]

Hi, (the Devil's Advocate speaking here...)


imap at chesapeake.net wrote:
> 
> All very good points.   Along these lines, I would like to mention
> that http://sourceforge.net is totally impressive and they seem to have
> solved alot of the open source development problems.  They service over
> 10000 projects...  Their blurb:
> 

I'm not sure if SourceForge is really going to be the _magic_ solution
to everything.  And I'm even worried that moving to SourceForge could
bring new problems...  I've been thinking of taking some other Open
Source project that I work on to Source Forge but one reason why I
haven't done it yet is because several times I tried to access their
servers and they were either awfully slow or completely unaccessible
(like this morning).

We already have a CVS server at UMN for the project that works very
well... it's reliable and access is fast anytime from anywhere... I
would hate to have to wait 5 or 10 minutes or try 5 times everytime I
have to commit some changes to CVS.

I'm not arguing against the SourceForge idea... I think it's great...
but unfortunately its too great success may be what makes it a bad
choice for now.

However, there are a couple of ideas that have been raised in this
thread that would help a lot and we should consider them seriously:
  - Bug tracking system
  - Change log
  - Demo/documentation project

Maybe we could give SourceForge a try (mainly for the bug tracking
system), but please don't move the current (reliable) CVS server
and mailing list over there until we are sure that SourceForge can be
reliable.

Marc Jacquin wrote:
> 
> And let's give Steve the rest he deserves, we should create a Board
> Committee deciding and validating (and managing) the updates
> (software,version,documentation).
> 

I guess you're referring to the fact that MapServer development has been
mostly led by UMN until now?  Are you suggesting that decisions on the
future of MapServer should be moved to a commitee?  I would be happy to
participate in it, but would a committe really help things?  I mean
Steve seems very open to suggestions and MapServer (like any project)
needs someone to make the final decision about new features being added
or not to the core and how it should be done... otherwise the software
could end up being a mess.

The fact that Steve and UMN lead the MapServer development does
not prevent anyone from making contributions today.  For instance, we
have contributed the PHP module and more recently I have added support
for reading multiple vector formats through the OGR library (but like
SDE, it still lacks query support).  I just had to make sure with Steve
that my changes would be consistent with the overall MapServer
architecture (taking current and future plans into account).

I guess I'm not really concerned by who makes the decision as long as we
make sure that they're consistent.  But I'm wondering if more people
would really make contributions just because there is a committee in
place.  After all, there is nothing that prevents anyone from
contributing extensions (and bug fixes) today.


BTW, Thanks to Pericles for the demo documentation... it's great!  
-- 
------------------------------------------------------------
 Daniel Morissette             morissette at dmsolutions.on.ca
               http://www.dmsolutions.on.ca/
------------------------------------------------------------
  Don't put for tomorrow what you can do today, because if 
      you enjoy it today you can do it again tomorrow.