Hi Tim,
Thanks for airing your concerns and worries.
On Mon, Nov 23, 2009 at 8:12 PM, Tim Daly <d...@axiom-developer.org> wrote:
> Is there a plan for what happens if funding is not approved?
Yes.
> Do the servers continue?
I'm not asking for funding for servers. The actual hardware is
owned by UW (but paid for by the NSF), and all hosting of those
servers is graciously done by the mathematics department at UW. This
is an extremely valuable perk of my job, and of being at a university
that understands software development, perhaps due to the influence of
Microsoft, Adobe, and the many other local software companies here. I
sysadmin them.
> Are the students reassigned?
I'm not asking for funding for students.
> Do the Sage days continue?
Yes (see below for more). There would just be less of them.
> Does the code move to sourceforge or github?
No. Though if it weren't for their limited quotas we would already
have a mirror there.
> Does this become a free-time only, non-academic activity?
No. The Sage project has a broad base of support:
http://sagemath.org/development-ack.html
Also, there is some revenue from people buying the tutorial on
Amazon.com, donations from individuals, etc. People make
tax-deductible donations here:
https://secure.gifts.washington.edu/as_mathematics/gift.asp?page=make&Code=MATSAG
That money is extremely important, but we're not dependent on it.
Regarding students, there are a *bunch* (6 or 7?) of UW undergrads
working on Sage right now. Some are doing it for fun. Most are
getting paid some amount. Currently, the ones that are getting paid
are getting paid from an NSF grant from the Applied/PDE group in the
department---the UW math department is amazingly collegial, and the
different groups try to support each other and work together as a team
(rather than against each other). Just like we should *all* do as
math software projects.
> Once Sage becomes a non-academic
> "free and open source" project there is no hope of getting
> NSF to pay attention again.
>From my perspective, Sage had no direct financial support for about
half the time it has been around. Despite the generous support the
project has received during the last 2-3 years, I think it is
definitely not dependent on support. Just to emphasize this, consider
that most release management for the last 5 months has been done by
Minh Nguyen (an undergrad in Australia) and Mike Hansen, and that was
all 100% volunteer work.
Sage is dependent on something: it has to continue to satisfy genuine
user demand and be a fun project to work on in order to remain strong
and flourish.
> Computational mathematics suffers from one critical flaw
> (which I pointed out to the NSF). Someone in an academic
> position only gets tenure track credit for published PAPERS.
> Source code does not count and it is a net loss to write
> working code if you are on a tenure track. This makes it
> hard for someone to aspire to a tenure position and also
> participate effectively in open-source computational math.
There is some truth to this, but in my experience it also
oversimplifies things. I have repeatedly seen exciting doors open to
young mathematicians as a direct result of quality coding and
computational work they have done. Also, in my experience some
computational work has had trouble being accepted by serious
mathematicians, partly because -- like classical Italian "proofs" in
Algebraic Geometry -- it violates accepted mathematical standards by
being built on top closed source commercial software. It's not
surprising that a very young field, relative to research math, which
is thousands (!) of years old, would have some growing pains.
Perhaps all our work on Sage, Axiom, Maxima, PARI, etc., is an
important foundation that will give young mathematicians a chance to
address this problem, just like the work of Bourbaki and others was in
making pure mathematics more rigorous and respectable. I decided
that instead of complaining, I would do work toward earning the
respect of the mathematical community. I am glad that you have also
put so much effort into doing the same (and you started years before
me).
Mathematica, Matlab, Maple, and Magma have the same place in genuine
quality research mathematics as non-rigorous proofs. They are useful
-- sometimes wonderful, inspiring, and incredibly useful (!) -- but
they aren't the real deal.
> So if (when) Sage is no longer a "hot topic", funded, and a
> top-of-the-interest-pile for new papers then the momentum
> slows a bit. NSF funding is critical for keeping the conference
> papers flowing.
I don't know what you mean by "papers". However, there could
certainly be Sage Days even without direct NSF funding -- just less of
them. To prove this, let's take the next 10 already planned Sage
Days (all in the next 9 months!), and in the interest of full
disclosure, I'll describe their funding sources:
# Sage Days 18 -- Cambridge, MA (December 1-5, 2009); (funded by the
Clay Math Institute and MIT)
# Sage Days 19 -- Seattle, WA (at Shuey House Mansion, January 16-20,
2010); (funded by the NSF)
# Sage Days 20 -- Marseille, France (February 22-26, 2010); funding: France
# Sage Days 20.5 -- Toronto, Canada (May 3-7, 2010); funding: Canadian govt.
# Sage Days 21 -- Seattle, WA (June?); funding: tentative, probably
PIMS = mostly Canada; otherwise, NSF
# Sage Days 22 -- Berkeley, CA (June 21 to July 2, 2010); funding by
MSRI (= partly by MSRI sponsoring institutions)
# Sage Days 23 -- Leiden, Netherlands (July 5-9, 2010); theme: Number
theory (funded by Holland)
# Sage Days 24 -- Linz, Austria (July 17-22, 2010); theme: (funded by
RISC = Austria)
# Sage Days 25 -- Mumbai, India (August 9-12, 2010); funded by India
(some national grant)
# Sage Days 26 -- Kaiserslautern, Germany (August 27-31, 2010); funded
by the German government
As you can see above, funding of these Sage Days is not dominated by
an NSF grant directly to me. It's shared between many countries,
including France, Germany, Austria, India, Canada, etc., and a private
foundation (Clay). Why? It's due to the grassroots effort of a
_lot_ of mathematicians all over the world who have decided to put in
a huge amount of effort writing their own small grant proposals for
conference funding. This model for funding is sustainable so long as
we all continue to work together as a team. The NSF proposal that
forms the main topic of this thread is simply a request to NSF to help
me do my part.
> Without it there will be no papers and Sage
> will become "venerable" software.
I hope someday Sage becomes venerable. Only RJF could twist
venerable = "calling forth respect through age, character, and
attainments; broadly : conveying an impression of aged goodness and
benevolence" into something negative....
> It is very difficult to run a large open source project with
> no funding at all. Axiom costs me between $4000 and $5000
> per year for hosting costs, equipment, personal travel, free
> books (I buy old copies of the Axiom book and give them
> away), etc. So while Axiom is "technically free-as-in-beer"
> the reality is that it comes out of my personal budget.
>
> I assume that Sage has similar direct and indirect costs.
There are no costs for hosting, as I mentioned above.
> If these are no longer receiving NSF and/or academic
> funding (e.g. free server hosting) what becomes of the
> Sage project?
It thrives, less.
-- William
--
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org
