Hi Milan,
Sorry for sending this replay to the list, but your mail server sent me
back this post three times in the last 24 hours:
MTA remote ultra6.eskimo.com: diagnostic SMTP: 550 5.7.1 <your adress
was here>... Access denied
> I would like to thank you first for your distrib and descriptive
> packages for Maxima.
>
Nice to get feedback from users!
> I have been using the distrib package over a last couple of weeks - my
> students having Math Knoppix CD, then rebooting whatever computer they
> work on, put some place the package files and then load it into
> Maxima. The Math Knoppix CD has Maxima 5.9.1 installed.
>
It's a pity they don't use a newer version; there has been lots of
changes since 5.9.1.
> The distrib package has been working without problems. However, the
> descriptive.mac does not work with Maxima 5.9.1.
I know this. In fact, IIRC the first version of 'descriptive' was
intended to be compatible with version 5.9.1, but the code I wrote
couldn't run under gcl. I can't remember the details now, but
I think it was related with the unability of gcl to manage functions
with many arguments. As a result, the first "official" release of the
descriptive package was designed for 5.9.2; there were in this version
some new functions that helped me to circumvent the gcl limitations
(Barton helped me with this).
> I then edited the
> descriptive.mac file and managed to get all functions working, except
> for the dataplot. I code examined that piece of code several times,
> but have not been able to understand why it is not working. I wonder
> if you would be able to find some time to look into that, and I am
> attaching the file I modified, in case that might be of help to you.
Looking at the changes you made I suspect your knoppix-maxima wasn't
built with gcl.
With respect to dataplot, it calls function 'listofnumbersp', which
calls function 'every', which I think was included in version 5.9.2 as
part of the nset package. Try this definition and let's see what happens
(unfortunately, I don't have a 5.9.1 version installed at hand):
listofnumbersp(y):=listp(y) and not member('false,map('numberp,y))$
Thanks a lot for your interest in these packages. When one knows that
someone out there is making use of one's work, that gives some
additional energy... really
Yours,
--
Mario Rodriguez Riotorto
www.biomates.net