John \"sb\" Werneken wrote:
> ACTION suggestion (if as a non-participant in the coding, I may do so): how
> about an "RFC UNREV-II-1: Request for Comments, to UNREV-II Group": "That
> our work carry the Free BSD License, of the type not requiring the
> acknowledgment". From what I see, ALL posters support either open source per
> se or open source with the ability for derivative work to be made
> proprietary. From what I understand, that License is accepted as "open
> source" and allows derivative work to be made proprietary.
> If it got support, we would (1) dispose of an issue; (2) establish a frame
> for decision making; (3) experience the joy of moving forward on something;
> (4) remove an apparent source of contention and/or of energy diversion.
Excellent proposals! I could live with the BSD-revised license.
I for one will not contribute any further code to this colloquium until
licensing issues are resolved to some form of satisfactory "open source"
and the development effort is moved out from the Stanford / BI
"permission to use" clause with its indemnification provision. So in
addition to having a BSD revised license, the "permission to use"
agreement would have to be formally revised.
I may continue to participate in the mailing list, especially on topics
like this one related to resolving the licensing issue. However, I
started interacting with BI and the Colloquium under the impression the
content of the effort (including the list) would become open source and
redistributable under some terms, to form the core of an open source
repository to solve global problems. I am disappointed no one from BI is
addressing this licensing issue on the list in an official way.
The beauty of a project like Squeak or Python is more than anything in
Jim Spohrer (the creator of the Squeak license) for example understands
this. He would be a good person for BI to ask for help form in
navigating what I see as a crisis of commitment.
This license issue is the single most important one facing BI to have
credibility as an open source project. It demands immediate attention. I
don't think it's the only issue, but it's the top one. The longer it
goes on, the more the "we" here diminishes into Stanford vs. BI vs.
> Suggestion Two: use source forge, with our own primary site as a mirror;
> this is an established and apparently functional method of supporting an
> open source project, to get things going.
Great idea. This would force the BI/Colloquium to use an approved open
source license, because otherwise SourceForge would not accept the
project. It would be an excellent test of BI commitment to "open
> Suggestion Three: use Eric's mission statement stuff: our intent would be
> announced, not as forking Source Forge, but as transcending it, developing
> an OHS for software development, which if and when it attained a 1.0
> release, we would replace Source Forge with it, and then continue
> development of our tool, using our tool to do so.
Great idea. However, be aware that whatever people say, the primary
benefit of SourceForge is that it is a free FTP server and free web site
for open soruce projects (with very hight bandwidth and disk capacity
and few hosting limits -- even compared to commercial hosting sites).
That is why people use it. The software isn't very important in that
sense (although the authors have every right to be proud of it).
> Suggestion Four: everybody says open source works by putting something out
> there that can be used, and then attracting improvements to it. There would
> of course be a huge gap between what Source Forge does and what Eric's
> requirements say an OHS would do, and I submit, there would be initially a
> huge gap, in Source Forge's favor, between what our initial efforts could
> do, and what Source Forge does now, but if we could at least demonstrate, by
> an actual useable application, however minimal, that our vision transcended
> theirs, we might have a start.
Great idea. But again, it would be easy to transcend their codebase, but
difficult to transcend their hosting capability. CVS is a very limited
version control system, even though it is popular.
Still, I very much agree with the notion of releasing code in early
stages, even if it is not as good as what is out there.
> AS you probably know, from Slashdot, Source Forge has attracted some
> bitching lately, to the affect that it's maintainers seem to be too busy to
> interface well with those who might wish to assist in developing Source
> Forge itself, so a move in this direction might be timely. If you visit the
> Slashdot post, please do skip through the numerous cute but off topic posts
> constituting the conversation between the two mutually amused posters, there
> is some real stuff in the thread, too.
Great idea. Again though, since SourceForge is basically a free web
hosting service, it is difficult for them to accept contributions
because they need to keep the site live. Also, organizationally, it is
a core group of people who meet regularly in a server room in CA and
know each other well, giving it a group dynamic a little isolated from
an internet-based developer community (something like what's happened
with the Colloquium too with weekly physical meetings -- i.e. Apache
developers didn't physically meet for years, until IBM paid to get them
in one room).
SourceForge is a unique resource for open source because of the high
cost of maintaining the hardware and bandwidth for free hosting.
However, the software it runs really isn't much of interest because it
only runs on one system, and runs identically for everyone. The only
way it might be of interest is if people could run multiple different
systems on SourceForge, or if it made sense to run it locally.
Developers of custom software and educational simulations
Creators of the Garden with Insight(TM) garden simulator
Porsche Boxter. You and a friend. Nine dream days from
Napa Valley to Beverly Hills. Provided by CarsDirect.com.
Click to enter.
Community email addresses:
Post message: unrev-II@onelist.com
List owner: unrev-IIemail@example.com
Shortcut URL to this page:
This archive was generated by hypermail 2b29 : Sat May 13 2000 - 07:41:42 PDT