[Axiom-developer] Re: bounties

[Bob McElrath]

Page, Bill address@hidden wrote:
> The idea is apparently to use donated (as well as some
> corporate sponsor) funds to pay relatively small incentives
> (but which carrying a *big* emotional multiplier of recognition
> for people who have been doing voluntary open-source programming,
> some extraordinary work that has remained quite unrecognized
> of several years).

I have even put forward some of my money for bounties.  (Unrelated to
this topic)  I don't know how many other people are putting forward
money for things they want.

> I would really like to see this idea extended to our little
> realm of computer algebra systems.

Do you think one could put forward a bounty for, say, an algebraic
extension?  In academia we generally have a very different
motivation/reward system involving journal publications.

However, Axiom does, and will continue to need pieces implemented which
are mundane, from an academic perspective.  (e.g. nobody could/would
write an academic paper about it)  Axoim is playing catch-up with Maple
and Mathematica.  Do you think bounties could work for this?  I can
imagine poor graduate students implementing some of this, and being
highly motivated by the bounty.  In my own little world, I would like
the equivalent of FeynCalc (and general Quantum Field Theory tools) for
Axiom, but I can't justify dumping a lot of time into that, because a
publication on it is unwise, and would not be respected (as physics) by
my field.  (e.g. it's not physics, but it's very important)

The other obvious way to get "mundane" things implemented is for people
in academia to put their students on it.  But, this is not necessarily
good for their students.  We would rather have our graduate students get
publications...  Students younger than Ph.D. candidates would be
excellent for implementing some of this, as they can do it in the course
of working a problem set (which is "mundane" in an academic-publication
sense) and may not conflict with the goals of their advisor.
I suppose we won't know if this will work until it's tried.  But I see a
lot of potential.

Hey...I have an idea, let's try.  Does Axiom have any existing funding
sources?  Could one write an academic grant for this, and establish a
pot for paying bounties?  The advantage of bounties is that they're
cheap, compared to hiring a developer.

Right now I volunteer $100 of my personal funds for Axiom work.
(because I'm a single post-doc and don't have a lot of expenses ;)  What
do people consider to be an important goal that is not already being
worked on, that we could reasonably expect someone to pick up?  This
kind of thing needs a rigorous definition for completion of the bounty,
and test-cases accompanying it to ensure correctness.

I'm going to put a lot of thought into applying for a small grant... I
mean, the arxiv got a grant, and is a critical tool.  This seems just as
worthy.

--
Cheers,
Bob McElrath [Univ. of California at Davis, Department of Physics]

    "It's not the people who vote that count. It's the people who count the
    votes." -- Joseph Stalin

signature.asc
Description: Digital signature