Re: [Axiom-developer] RE: proposal to create the Axiom Foundation (was: bounties etc.)

[Martin Rubey]

Dear Bob,

Bob McElrath writes:
 > I have just sent out my $100 via paypal.  

Thank you!

 > Sorry I have been out for surgery for a while, still catching up on email...

I hope that everything went well! All the best!

 > I ask that it be used for a new mathematical domain, extension, or
 > algorithm.  (As opposed to an interface, compiling, or integration issue, of
 > which there seem to be many...)  

Since you are one of the decision-makers, you can ensure this yourself :-)

 > I will look over the wishlist and see if > anything catches my eye.

Do so!

 > Can someone summarize the funds collected so far and the bounties out?
 > We have proposals on::
 > 
 >     http://page.axiom-developer.org/zope/mathaction/AxiomFoundation
 > 
 > but no concrete promise of bounties for anything specific.

No, since there has not been any agreement, neither on the bounties themselves,
nor on the pricing.

 > Also let me add my voice that I think the bounties so far are too small, or
 > the goals not big enough.  It has been noted that $100 is a nice award in
 > poorer countries.  In US monetary terms I think our goal should be $500
 > minimum, which is roughly one to two weeks of a grad student's time.  I
 > think efforts smaller than this will not benefit from bounties.  But
 > then...I'd like to be proven wrong on that...  And of course we are funds
 > limited for now.

There are several issues:

* we don't have too much money to spend, thus I regard the bounties rather as
  a way to say thank you. If you happen to live in a country where 100$ are a
  fortune, even better. (In fact, I think we should specifically target such
  countries. Mathematica and Maple cost a fortune, I don't know about Mupad.)

* I dare say that it will be very difficult to make conditions precise, under
  which some submitted work is "good enough". Therefore I argue that "good
  enough" means, that the "heavy lifting" is done, i.e., the code works. This
  is another reason to set the prices at a not too high value. I think we
  should allow ourselves to negotiate here: if some submitter says: "Here's the
  code, it works. If you want some really good documentation, I'll ask for
  another 100 bucks.", I think this will work better than insisting on
  documentation right away. Furthermore, I think that maths will be even more
  difficult to judge than some integration work.

 > Also, has it been decided who are the foundation members?  Am I one?

I think so, yes.

Martin