On 01/30/11 03:27 PM, Jonathan wrote:
As one of the "non-mathematicians" (as a physical chemist who teaches
quantum mechanics and kinetics = nonlinear differential equations and
systems of differential equations, I'm not sure that is quite the
right term, but I think you get the idea), I thought I should chime
in.
1) I think SAGE has an adequate plan.
I disagree.
It is a community project. The
successful open source projects I know about serve a need for the
community(s) they serve, because the actual users contribute code that
does what they want.
Firefox, OpenOffice and FreeBSD are all successful open-source projects, but
have a documented plans with at least projected time scales. If dates are
regularly missed, then one needs to determine if they were unrealistic, and so
allow more time for future plans. IMHO, Sage lacks a real direction/plan.
Developers don't just add whatever they fancy to Firefox. Just because someone
wants to be able to do something obscure in Firefox, their code does not get
committed to the core. Instead they write an add-on
https://addons.mozilla.org/en-US/developers/docs/getting-started
which is then available for download.
This keeps the basic Firefox code reasonably compact and easy to manage. If the
author of an add-on does not develop it further and it fails to work with more
recent versions of Firefox, then so be it.
Is solving Rubiks cubes really needed in Sage, or would that have been better if
it was an add-on, which people installed if they needed it?
It appears to me that if someone wants some quite obscure functionality in Sage,
they write code and it gets added to the library. That might be really useful to
that person, as they use it as part of their research. But IMHO, unless it is
generally useful to a reasonable number of people it should not be in Sage.
Put another way, there should be a discussion about what Sage needs, how urgent
it is, and a plan drawn up.
I thought porting Sage to Windows via Cygwin was seen as important, as it will
dramatically increase the number of users. But this seems to have stalled.
2.5 years ago (June 2008 to be precise), Mike Hansen said "The first step which
should be done within a few weeks is to get a Cygwin version of Sage for Windows."
http://groups.google.com/group/sage-devel/browse_thread/thread/80cd1fa0c2f48eaa/94e06945347d6466?lnk=gst&q=cygwin+mike#94e06945347d6466
It was several months ago William said it would not take a lot of work if him
and Mike worked on it. IIRC, it would a couple of weeks of work. That was before
William start Psage, which seemed to be quite badly timed given the Cygwin port
status.
This Cygwin related page:
http://trac.sagemath.org/sage_trac/ticket/6743
has not been updated for a year, and another one:
http://trac.sagemath.org/sage_trac/wiki/CygwinPort
has not been updated for 4 months.
Michael Abshoff was *employed* to work on the Solaris port of Sage, yet went off
on a tangent and did things that interested him. I tried to help on the Solaris
port at that time, but got frustrated that changes Micheal knew of were not put
into Sage. I was simply wasting my time, so I gave up for a few years.
The key here is
that you need a few dedicated individuals directing the project and
ensuring that it moves forward.
But that is not happening.
For them it needs to be a nearly full-
time job.
Not necessarily.
I'm involved in the Jmol project and we depend on one
person who does the bulk of the work. Finding these people is
difficult because they need to be supported in some way. This means
the software has to be important for a company or this person has to
be an academic who can convince their institution that the work is
worthwhile and scholarly.
I know in the case of the Solaris port, some body was convinced to pay Micheal
to do the port, but IMHO he did not do an alful lot. In fact, several of his
changes just made things more difficult.
2) I agree that SAGE could use more exposure. I have found it better
for my teaching than Maple, which my institution has a license for. I
can't speak of recent Mathematica editions, but I know that 15 years
ago I found serious problems with it and gave up on it. Anyway, there
are lots of people who could use SAGE, but there are two key issues:
a) for Windows an install that runs in Windows as an application would
be nice;
Agreed - see comments above.
I would like to introduce someone to Sage, but I'm wasting my time if she needs
to learn Linux, Solaris, or buy a Mac first. So I'm suggesting she use
Mathematica on Windows. As much as I don't like that idea, I don't feel able to
recommend Sage. If there was a Windows port, then I might change my mind.
3) I think the issue of crackpots and bad code dragging things down is
not much of a problem.
I agree.
My example may be a little slower than many people's because I also
have very little time to contribute to this, but I still think you are
unlikely to get really bad code included using the present model.
The lack of attention to detail worries me about some Sage developers. That's a
view shared by Peter Jeremy, when he said:
"I am very concerned at this "release it now, we'll make it work later"
mentality."
http://trac.sagemath.org/sage_trac/ticket/6456#comment:67
That's a different issue to crackpots. I agree they are not having an adverse
effect.
Jonathan
Dave
--
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
--
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
