Re: Media Library, pt. 2

[Derek Simkowiak <dereks kd-dev com>]

> What I think you mean is that we use C++ as a back-end and then wrap it up
> in CORBA. However, I'm very afraid that we will get a lot of BeOS code that
> cannot be used under Gnome, and vica-verca.

	If it's going to be part of Gnome libs, we'll need a C interface.
OpenGL and OpenAL are in C.

> problem - if they have COM or DCOM, damn - if they don't have any, they're
> like KDE (on their way to extinction :)

	(offtopic note: I thought KDE was using Corba for this?!)


> > As an aside, if this project needs any net-type resources (web space,
> > email forwarding, web design, etc), inter/open would be happy to provide
> > them/host the project - whatever's required.

	I too have a server sitting on a T1, but SourceForge would
probably be better; they have a full development set of software, complete
with mailing lists, user administration, domain hosting, etc.  Not to
mention professional, full-time *free* admins, backup systems, and
serious-ass bandwidth.  Everything you need for a collaborative project :)

	So I guess the next step is to get serious about forming a
consortium.  


TO-DO: 
------ 

1) Decide on a name.  Personally, if we can have an OpenGL-like API for
this (as Loki has done with their superb OpenAL library), then I'd like to
go with "OpenML" in the hopes that people will think of it as the "OpenGL
of media programming".  It could also encourage support from SGI.

	I'd be willing to pay a domain registration fee.  "openml.org" is
available, "openmedia.org" is not, "openx.org" is not,
"openmediaconsortium.org" is available.

2) Legal issues.  How, legally speaking, is a "Consortium" defined?  Do we
need to register with the IRS (here in the U.S.?)  What paperwork is
necessary?  Who will pay for the lawyers? :)

3) Find out who is interested in this.  This means

	a) Deciding who here will act as the consortium's official contact
person, to call up SGI, Apple, Loki, etc. and inquire about interest

	b) Obtain a list of contacts at the different companies.  We will
want to inquire about corporate sponsorships as well as technical
implementation issues

4) Technical design issues: I, personally, have no experience with a media
architecture.  I don't know where Corba would go, or how much of our API
would depend on the different Codecs, or what level (hardware wise, and
such) the API should operate at.  Could this API be used to control DVD
players?  Video tapes?  DATs?  3dfx-accelerated cards?  Etc.  I guess this
to-do item boils down to finding experienced people.  Be can really help
out with this.

5) Sourceforge site and web design, along with dedicated mailing lists.
Given SourceForge's intense list of features, I think they should host our
site.  Also, we need somebody with Gimp experience to design the site.


	I can help out with number 5, but I'll first need a name for the
consortium (see #1 :).  I can do the graphics and HTML work and register
the project with SourceForge.  After the site is set up, my time for this
project would be limited to 4 hours per week (for the next few months, at
least).


--Derek