[Openmcl-devel] OpenMCL funding and development priorities

[Gary Byers]

On Mon, 29 Aug 2005, David Steuber wrote:

> For the first time ever I just had a look at http://www.clozure.com/.

Ah.  Our clever marketing strategy is working, I see.

>
> I've always wondered, but never bothered to ask, how is OpenMCL
> funded?  What parts of OpenMCL are currently getting the paid
> development support?

We should remember to acknowledge that Dan Knapp's work on the
documentation over the last year or so, Alex's work on the IDE, and
all of Bryan's work over the last year or so (tons of bug fixes and
other improvements) -have- been done on a volunteer basis and have
benefited and will benefit users.

The PPC64 port that'll be in the next release was funded, as has
been some of the other work that'll go into that next release and
(hopefully) the next release or two.

>From 2002 until the spring of 2004 I worked for a company that used
OpenMCL and wanted to support its development.  From the spring of
2004 until this past spring I was working on something unrelated, and
I think that there was a lot less development activity (from me, at
least) in that year than there's been in the last few months.  I hope
that that activity translates into progress, more users, and more
users for whom funding continued development makes financial sense.

>
> Also, with Apple's move to Intel, is a Linux/x86 OpenMCL in the future? (or 
> perhaps some other Linux/intel architecture since Apple, to my knowledge, 
> hasn't been too specific about what the final Intel CPU will be other than 
> the dev boxes using the PIV).
>

I haven't heard anything more specific about what CPU Apple plans to
use.

Assuming that Linux runs on whatever that CPU turns out to be, the
effort required to port from Darwin on that CPU to Linux on that CPU
is -likely- to be a lot less than the effort required to port to that
CPU in the first place.

The last time that I checked, official Apple documentation describing
the x86 Darwin ABI consisted of a link to an old Linux x86 ABI
document and an assertion that the ABIs were very similar; I haven't
looked in a few months, and don't know if there's been anything added
that'd describe whatever differences exist.