Currently my main priority is release, but after release I will at
the wiki.
OK.
BTW, before official release, can we have a pre-release so that I can
again check the things I care about?
In particular we should have a freeze and then look at the stuff that
should go to fricas.github.io (i.e. all the things that live under
src/doc/sphinx). I had some ideas to restructure install.rst, but I am
not sure I find enough time to clean things up.
I am currently checking that the stuff for jfricas is nicely integrated
into the fricas repo.
BTW, do you already have a description for how to build hsbcl. In fact,
to keep it simple... hunchentoot should just live in the binary
distribution. There is actually no need to change anything in the fricas
documentation to describe an installation from the git repository that
works for jfricas. Maybe, I simply document this in jfricas itself.
What I, however, would like to see in the documentation of FriCAS is
reproducible description of how the binary release was produced.
In the long run I propose to use the jupyter notebook format
(actually with the help of JupyText ordinary .input files that
follow a certain simple structure are sufficient) and store all
this in the repo
https://github.com/fricas/fricas-notebooks/
and show it via (for example)
https://fricas.github.io/fricas-notebooks/FriCAS-FirstSteps.html
The git repo would also nicely record the history of the content.
Just an idea. Of course, I'd need help to transfer important
content from the wiki to the git repo.
I am not sure what you really propose. One important point of
FriCAS Wiki is ablity to handle actual FriCAS input and show
results.
I understand.
What I currently propose is more or less a static view. Giving people
tutorials in the for of working jupyter notebooks.
Having jfricas working with a webserver locally, it shouldn´t be a big
problem to put some machinery in front of it and serve a limited access
to a jFriCAS interface over the internet. Yes, of course, that needs
hardware on our side, but is probably more convenient than editing the
stuff in Zwiki.
Do Github allow installing our executable on their server for this
purpose? My understanding was that this was not possible. And
handling input coming from net is reason for VM.
Yes, of course. Security is an issue. I am not at the stage of offering
something like CoCalc. Currently, I only want a static view, but some
day, we may switch to a web access for people to try out FriCAS, not for
serious computations.
But I am not eager to replace overengineered system that reasonably
well satisfies our need by something which drops needed functionaly.
You need not repeat. I know your opinion by now.
And I am not eager to replace one "object store" by different "object
store" which may be even less managable.
Oh, well. The zope object database is too much for my taste. Storing in
git is much simpler. And, in fact, github even offers to edit online and
commit. Have you seen the "Edit on Github" link? All that is needed for
a potential contributor is an account at github. There is not much (if
at all) knowledge about git needed.
Ralf
--
You received this message because you are subscribed to the Google Groups "FriCAS -
computer algebra system" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/fricas-devel/a96bc7db-b425-9134-e550-2ad1f8c8cbae%40hemmecke.org.
