[EMAIL PROTECTED] wrote:
>
> The clauses that Emile is referring to are an entirely different animal.
> I've inquired about recieving legal consoltation concerning these issues
> and the FSF has offered limited probono (sic). I also have a U.S.
> corporate laywer who I eventually will pay to construct or advise on the
> issues of non-competition and confidentiality. I consider these to be
> important and complicated issues. When companies don't understand Open
> Source by legal letter and spirit they envision concepts that go against
> Open Source interests.
I'm not in charge of writing contracts at Aurora but IMHO, the main
competitive advantage Aurora can have is hiring some of the core
developers/writers of Midgard. As you already know, Aurora has already
hired -or is going to hire- Alexander, David, Emiliano and Ron. Their
main task is to deliver Midgard 1.4 final in time (1st October) and will
be to develop Midgard 2.0. They also can help other Aurora developers to
develop with Midgard and give some advice for Midgard projects Aurora
has with customers or internally.
Other people at Aurora who also work mostly for Midgard are C�dric,
Patrick, Simon and I. A new developer also has joined the team on August
1st: Bruno Abitbol (I'll introduce him in my next email).
> I think it would be great to see more development being paid for by
> distinct and seperate businesses for several reasons. I don't think Aurora
> should bear all the responsibility for financing the development of
> Midgard. They can't anyways because of outside OS developers who have done
> %99 of the work on 1.4. Of course productivity will increase as we
> begin the development of 2.0 because of Aurora's financial assistance. I
> know there's at least a couple other developers who are interested in
> working full-time on Midgard and additional financing would help that
> cause.
Fully agreed. Aurora isn't a multinational company and I don't think
we'll be able to hire many new Midgard people by the end of this year.
So, as Midgard is an Open Source project (I mean Midgard in general:
V1.4, V2.0, ...), it's obvious that we'd like to see other companies to
join Aurora in contributing to Midgard development. Another benefit
would be to have a kind of international network of companies able to
develop services with Midgard.
> IMO, there's a danger of having OS projects monopolized and controlled
> when one company pays for all the development.
Agreed, but I don't think -at least at my level as R&D manager-
there is such a risk because the Midgard Association is independent and
isn't controlled by any commercial company. As long as the developers do
have control over the licenses of Midgard, it cannot be controlled by
any company.
> My own experience at Aurora has lead to a 180 degree turn in the focus of
> documentation. It's distinctly possible that this is for the better. The
> point is, I have little say over the type of documentation that I'll
Our main concern concerning the documentation at Aurora was that the
Midgard documentation was totally unclear and confusing to most Midgard
newbies. So a great effort is been done by Ron, C�dric and other to
explain as clearly as possible what does Midgard, its main concepts and
how it works.
> write. What happens when Ami believes that a RDMS is the best backend for
> 2.0 but Aurora wants LDAP and the primary developers are working for
> Aurora. I believe these concerns are well managed for the moment because
> the right people are working on the project. However, I think it's niave
> to believe that for profit efforts won't be tempted to push self
> serving business strategies on OS projects.
The X.500 / LDAP backend hasn't be selected yet as the backend for
2.0. BTW, it's very possible that there will be several available
backends for Midgard in the future (SQL, object DB, ...). The purpose of
the initial choice is just being able to deliver 2.0 ASAP. Also, an LDAP
server itseld isn't really a backend, rather an access layer to it. One
LDAP server -like OpenLDAP- can be used with different backends.
The only reason we talked more about LDAP than other backends (like
Object Databases) is that I have some experience with LDAP/X.500 and so
can contribute more efficiently. Developers testing other backends are
of course welcomed.
Bruno is also doing a great job actually designing and testing a
part of Midgard database over OpenLDAP. He and I will post our first
feedback ASAP so everybody can comment. First LDAP tests he did:
designing LDAP DIT -part of Midgard-, implementing ACL, a PHP user
interface to do some basic testing (binding, browsing, searching,
inserting, ...). We'll also focus on some of the immediate built-in
benefits of LDAP (scalability, flexibility, ACL, replication, ...).
Further tests will then be mainly load tests to see how LDAP handles
large number of entries and concurrent users.
> Just so nobody gets the wrong impression, I am not saying that Aurora is
> or will attempt to serve their interests at the expense of Midgard. My
> example scenario concerning backends for 2.0 is nothing more than an
> example and should not be interpreted as anything but an argument for the
> purpose of demonstration.
Yes, Aurora cannot and won't make any decision about the design or
the implementation of Midgard itself. We can just propose solutions as
any other potential contributor. May be the only difference is that we
have more time at Aurora to do some design / testing than most
individuals.
Regards,
Jean-Philippe.
> PEACE
>
> Ron
>
--
| Jean-Philippe BRUNON /-/=\ _ _ _ __ ___ _ __ __ _
| mailto:[EMAIL PROTECTED] | / \=|| | | | '__/ _ \| '__/ _` |
| http://www.aurora-linux.com/ |/---\|| |_| | | | (_) | | | (_| |
| Phone: +33 (0)1 58 17 03 20 \___/ \__,_|_| \___/|_| \__,_|
--
This is The Midgard Project's mailing list. For more information,
please visit the project's web site at http://www.midgard-project.org
To unsubscribe the list, send an empty email message to address
[EMAIL PROTECTED]
