I am of one mind with Aravindh on this.

I spent all day setting up Jupyter to work with FriCAS. The documentation 
is poor on this topic. This page: 
<https://fricas.github.io/install.html#hunchentoot-optional> — does not 
explain nearly enough. I had to refer to a number of other web sites to 
figure out what is SBCL, Quicklisp, Hunchentoot, and how to make them work. 
At least there is some documentation on their home pages. What is 
documented nowhere is how to put these pieces together to get FriCAS built 
with Hunchentoot baked in. I allow that this is an easy matter for a Lisp 
developer. But there are very few Common Lisp programmers in the world. I 
have experience with a number of languages and build systems, but I do not 
think I could have figured out how to build FriCAS with Hunchentoot had I 
not found Aravindh's patch.

The HyperDoc system looks very dated to me as well. I do not see myself 
using it. To make matters worse, when I ask it to solve an equation, it 
tries to launch `xterm`, which is not available on my system, and hangs. 
Annoyingly, HyperDoc pops up every time I restart my FriCAS Jupyter 
notebook. This is an annoyance that I wish I knew how to disable. I looked 
at Aravindh's patch but it seems to contain a lot of other changes so I am 
reluctant to apply it. I wish there was some way to configure FriCAS not to 
open HyperDoc at startup.

I think FriCAS would benefit from making it easier to integrate with 
Jupyter and other third party tools. This does not seem like a hard problem 
in principle, since all the pieces are already here. But having to 
customize the build instead of simply grabbing the binaries provided by my 
Linux distribution requires both effort and expertise, and all the rough 
edges make this option unattractive. A little polish would go a long way in 
making FriCAS more accessible and attractive.

I am grateful to Aravindh for making the patch available on his GitHub 
account and writing a letter to this mailing list linking to it.

On Friday, October 25, 2024 at 12:59:31 PM UTC+7 Aravindh Krishnamoorthy 
wrote:

Dear Waldek,

Sorry for my late response.

> Or for some reason you want FriCAS graphics but want to specifically 
disable HyperDoc?

Unfortunately, this seems to be my personal preference. 
The FriCAS CLI and HyperDoc GUI look very dated to me. So, I use FriCAS 
primarily as a kernel for Jupyter notebook.
For the future, my hope is that I can interface "jFriCAS" with "jlfricas" 
(FriCAS with Julia by gvanuxem) and use Julia's plotting commands to show 
FriCAS plots in Jupyter notebooks.

Best regards,
Aravindh Krishnamoorthy

On Tuesday 23 January 2024 at 02:58:57 UTC+1 Waldek Hebisch wrote:

On Tue, Jan 16, 2024 at 06:08:35AM -0800, Aravindh Krishnamoorthy wrote: 
> 2. Updates to configure.ac to disable HyperDoc by default. 

I am not sure why you want this. If you do not have X, or do not 
want to use X, then there is '--with-x=no' option to 'configure'. 
I use this option on servers where access is text-only. 

Or for some reason you want FriCAS graphics but want to specifically 
disable HyperDoc? 

-- 
Waldek Hebisch 

-- 
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 fricas-devel+unsubscr...@googlegroups.com.
To view this discussion visit 
https://groups.google.com/d/msgid/fricas-devel/feda8cf0-c39f-439a-a69b-78da97a4ac88n%40googlegroups.com.
            • ... Waldek Hebisch
              • ... 'Nasser M. Abbasi' via FriCAS - computer algebra system
              • ... 'Nasser M. Abbasi' via FriCAS - computer algebra system
              • ... 'Nasser M. Abbasi' via FriCAS - computer algebra system
              • ... Ralf Hemmecke
        • Re: [... Qian Yun
          • R... Waldek Hebisch
  • [fricas-devel] Re:... Aravindh Krishnamoorthy
    • Re: [fricas-d... Waldek Hebisch
      • Re: [fric... Aravindh Krishnamoorthy
        • Re: [... Ignat Insarov
          • R... Dima Pasechnik
            • ... Ignat Insarov
          • R... 'Ralf Hemmecke' via FriCAS - computer algebra system
            • ... Ignat Insarov
              • ... Waldek Hebisch
              • ... Ignat Insarov
              • ... Waldek Hebisch
              • ... Ignat Insarov
              • ... 'Ralf Hemmecke' via FriCAS - computer algebra system
          • R... Waldek Hebisch

Reply via email to