This is the mail archive of the mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: why is the SC relevant (was Re: SC issues revised)

Tom Lord wrote:
> 	2) Given a formalized agenda, lower the barrier to multi-party
>            custom development contracts: where several customers who
>            compete in other domains have a clear path to pooling
>            resources for their common interests.  Being able to point
> 	   to an SC RFI document in a contract is an example of a 
> 	   lower barrier in this area.

That would be nice, but getting for a "consortium" to pay for
development is difficult.  Also, the SC member don't have time
or inclination for heavy-duty selling.

Two related ideas:

* Given some desirable infra-structure change, such as a re-written
C++ parser, or complete Java AWT implementation, it might be worth
trying to spec them out, and then ask for bids.  For example Red Hat
might state that they are willing to complete the AWT implementation
for $200k. They or the SC would accept confidential comittments from
companies that they are willing to pay for (say) 20% of the work.  When
5 such customers have signed on, then contracts would be signed, and the
work scheduled. RedHat or the SC could announce that there are 3
comittments to co-fund a project and that we just need two more.

I don't know if such a scheme could work.  Variants of it have been
tried in the past, with limited success, that I can recall.  But if we
have more formal process including publicly visible project proposals
and confidential signing up of customers, perhaps it would be worth it.

One problem is avoiding a bidding war.  Suppose Red Hat committs to
implementing AWT for $200k, and Joe Schmoe Consulting comes along and
says "my overheads are less, and I'll do it for $100k".  Would that be
good or bad?  Would that possibility discourage Red Hat from posting a
bid in the first place?  How would one evaluate the competency and
reliability of Joe Schmoe?  Should that be up to the SC or up to the
companies ponying up the money?

* Another idea is the formal consortium idea, with annual membership
fees etc.  This is how the X Consortium worked - for a while.  Other
possible models include W3C, XFree86, Gnome, KDE, and Apache.  I have no
idea how successful these are at getting corportate sponsorship for
development projects.  Trying to start up something like that seems
like a huge undertaking, though we could probably use the FSF as
an umbrella corporation.  It would probably be easier when the
economy is better.
	--Per Bothner

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]