Re: [DotGNU]SourceForge drifting

[tony stanco]

> David Sugar <address@hidden>

>I think Savannah, and what is now happening there, is the codebase for
>the future, not Sourceforge or whatever happens to the pre-fork
>codebase.  Being able to setup a Savannah anywhere will be very useful.
> It is a mistake in my view to have "one" monster site.  There are
>plenty of projects organized around a small community or a campus, and
>having a Savannah one can role out to manage them locally would be very
>nice.

A monster site is usually going to be a problem, because it can try to
monopolize or is a single point of failure issue. I talked to Larry Augustin
about a month ago about placing SourceForge at the universities that I deal
with for FreeSoftware development, because they have huge pipes, can act as
local centers and they have professors and students, who we should invite
into the community. I thought it would have been better to do that with
Savannah, but talking to Richard and Loic during the summer, I thought
Savannah was only for GNU projects for the foreseeable future.

>At most, what would be nice is a meta-aggregation site, such as thru a
>search/directory engine for "published" projects on Savannah-like sites.

Exactly. Distributed Intelligence with a good searchable directory linkage
is a good overall model.

> Not all projects need to be published widely; sometimes someone might
>create a small site and set of projects for personal work, for example.
> Another example might be a company or organization which has some
>custom projects and some public projects.  Being able to move projects
>between such sites easily could be nice also.

Yes.

>I think being able to deploy Savannah in a distributed manner, and
>having a global directory/search engine site that can sit on top of it
>would meet everyone's needs very well.  I also think much of the work
>that has gone into Savannah will eventually make this become reality.

There is a very good chance I can get some grant money from the NSF for this
with George Washington University and Cyberspace Policy Institute as
sponsors. I think $500,000 is possible.

We would need documentation and specs for the application. There is a
December 6 deadline for some IT projects. I would have to check to see if
this can be fit in there.