Re: [Axiom-developer] html formatter

[Tim Daly]

Martin Baker wrote:
> On Saturday 30 Jan 2010 13:41:09 Tim Daly wrote:
>   
> > Are you aware of the work in Axiom Volume 11: The Axiom Browser?
> > It uses mathml for mathematics display.
> >     
>  
> Tim,
>
> I did come across it although I did not fully take it in at the time, so much 
> to learn when first looking at the Axiom documentation. The documentation 
> probably needs a bit more overview at the start of volume 11 before jumping 
> into build instructions so that newcomers can see what its all about.
>   

At the moment all of the energy is going into moving the source code 
from the
tree-of-tiny-files format into books. It takes a while to move a 
million-things-of-code,
along with rewrites and refactoring. I'm into the final stretch though 
as I only have to
consolidate the interpreter and compiler. I hope to have it finished by 
the end of the year.

Once that happens I will revisit each book with an eye to writing decent 
documentation
See the book Lisp In Small Pieces for the ultimate goal I hope to 
achieve. The books
only contain the source code for the most part. There will be much more 
once I get rid
of the 1960s-PDP-11-64K-sea-of-tiny-little-files organization.
> It does seem a very nice idea, using the browser for what it is good at and 
> use Axiom for doing the maths. Is this is what provides calculations for the 
> Axiom wiki? It would be a good way to show new users the advantages of Axiom 
> if there were a page where they could type in a formula ,or sequence of axiom 
> commands, and see the result immediately on the web page.
>   
Actually when you run the browser from the Axiom command line you can 
input expressions
into the browser and the computed result will open up a <DIV> in your 
browser with the results.
This works now. All of the Javascript/AJAX/mathml/axiom-interpreter code 
works so it is
easy to add new pages. Once I got a proof of concept working I went back 
to building up
the other books.

Arthur Ralfs is the person to credit for the mathml connection.

More of the hyperdoc pages have to be converted and there are plans for 
deeper documentation
in the browser format. The browser form already contains pages which are 
not in hyperdoc.
There are newly written pages and code examples from the NIST Digital 
Library of Mathematical Functions.
There are other pieces of work stacked up  to put into the Axiom 
browser. We really want
to support the bra and ket notation from quantum theory. (I've been 
playing with that for a
while and it would be nice if Axiom supported it better.)

> Looking at the dates it looks like development on this stopped around 2007, so 
> I guess interest must have declined, which seems a pity. 
>
> Using the command line interpreter has pros and cons, I wonder if this could 
> be a way to overcome the cons, for instance: allowing the user to enter 
> mathematical symbols such as the Greek alphabet directly into the formula 
> using buttons on the web page?
>   

The browser will certainly support unicode I/O. The question would be 
whether the lisp
system will support unicode, which would be nice because then you could 
input/output
symbols like sigma as though it were any other character. Since Axiom is 
listening on
the host end it can interpret unicode any way it likes.
> Martin Baker
>
>   
Tim