[Axiom-developer] Re: RISC

[Bill Page]

Alfredo,

On Sat, Jul 26, 2008 at 10:43 PM, you wrote:
> Hi Bill,
>
> For those no so fortune to be able to travel to Europe
> ( I cannot even travel in the states :-),
> is there gonna be a summary some where of what it
> was discussed ? or at least interesting news :)
>
> Maybe a video...:p...yeah I know I am asking for too much.
>
> Thank you,
>

I am quite sorry that you and some of the other Axiom and Aldor
developers and users were not able to come to the meeting. I am sure
you could have provided valuable input to the discussion. Notably
absent also was Tim Daly, especially since the main developers of the
two other other Axiom-related projects were there (Gaby, and Waldek).

Here is a fairly accurate updated schedule of the workshop prepared by
the organizer of the workshop Ralf Hemmecke:

http://portal.risc.uni-linz.ac.at/Members/hemmecke/aldoraxiom2008

I think that overall the organization of the workshop was very good -
much thanks to Ralf! And in spite of often being arranged at the "last
minute", the talks and discussions actually fit together into a quite
coherently whole. But sorry, there was no video or audio recording.
:-( In my opinion this is something that we need to try harder to do
in the future.

One very good thing that was done this year for the first time was to
combine both the Aldor and Axiom workshops into one. This meant that
Stephen Watt and a few other people were present specifically because
of Aldor and vice-versa. This was good for two reasons: Stephen of
course was deeply involved in the development of ScratchPad before it
became Axiom and the main developer of Aldor so he was able to offer a
lot of important background on both projects.

Secondly I think it helped strengthen the argument that Aldor should
have a more open open-source license because of the amount of
development work occurring in the Axiom forks in comparison to Aldor.
At the main ISSAC meeting which occurred just before the workshop
William Stein, the main developer of Sage, emphasized exactly the same
issue.

Besides many technical discussions on language design, comparing
developments in SPAD  (especially in OpenAxiom) and Aldor, there was
at least some small progress towards a more open license. One reason
for this is the work being done by Ralf Hemmecke and Peter Broadbery
on the Aldor-Axiom interface. There a few runtime-related files in the
Aldor distribution that are required to build the interface and
everyone agrees that it would be a good thing to be able to distribute
and build the Aldor-Axiom interface without having to include the
Aldor sources and it's license. More information about that will
follow in a few weeks I hope.

For me one of the most surprising and exciting developments in the
last year was that an entirely new and very active Aldor developer has
appeared thanks mainly due to the ubiquity of the Internet and the
still rapidly growing open source culture. Pippijn van Steenhoven
discussed his work on the internals of the Aldor compiler using
"state-of-the-art" tools and methods that are common in other open
source projects, such as valgrind and other profiling tools. As a
result of this work Pippijn has already been able to isolate and fix
some important causes of segfaults that plague the Aldor compiler. You
can both see (and use!) the results of his work here:

http://xinutec.org/~pippijn/en/projects_aldor.xhtml

Please ask more questions. It seems a bit hard for me, although it is
still current, to really summarize the results of the workshop. As
time permits, I will try to add more in follow-up comments. I hope
some of the other people who were there will also offer their
perspectives and comments here.

Regards,
Bill Page.