RE: [Axiom-mail] Windows: where's the documentation?

[Bill Page]

On October 31, 2006 8:58 AM Paul vL wrote:
> Martin Rubey wrote:
> > 
> > I *STRONGLY* advise you to try something different: install 
> > Xming and try to get Tim Daly to give you instructions on
> > compiling Axiom with HyperTex on MS-Windows. Possibly Bill Page
> > also can help. If you mangage to get HyperDoc running and you
> > provide some documentation, I'd propose you for a bounty!
> 
> I actually made a start trying to compile Axiom using MSYS/MingW
> but got stuck.

Building Axiom under MSYS/MingW is quite easy. See:

http://wiki.axiom-developer.org/BuildAxiom

> As an experienced MingW user I know how much trouble it can cost
> to solve all these issues (I'm currently trying to build GCC
> cross-Arm tools under MSYS/Mingw, it does *not* work out of the
> box and needs lots of patching). As I also have other projects
> lying around which I like to finish, I currently do not have time
> to go along this road...

Great. Maybe you can add Axiom on Windows to your list of project. :-)

> > 
> > The great thing about hyperdoc is that you can enter a category,
> > say, for example, Ring, then press on a button labelled "Domains" 
> > and you will get all Domains known to Axiom that are Rings. I.e.,
> > Integer, Polynomial Integer, Fraction Integer, PrimeField, and so
> > on.
>

For a different point of view, let me say that I almost NEVER use
hyperdoc. I find the user interface clumsy, unreliable and of limited
use. I prefer instead to search and read the Axiom sources directly
and to use )display and )show commands in the Axiom interpreter.
 
> How would that be different from any format that allows
> hyperlinking? Hypertex is IMHO quite obscure, and I cannot see
> a big benefit for using it over standard solutions which are
> widely supported.

I agree. One complication is that a lot of the intereting information
about Axiom is actually stored in a data structure called the Axiom
database (daase). Hyperdoc uses this information to navigate. It is
not straight-forward to access this using just a web browser - but
of course there are several different ways in which this could be
made available to a standard browser.

> 
> I do not ask to ditch Hypertex, but just another option to help
> all Windows users out there.

Using Xming is a possibility but I thing most Windows users would
find the X-windows based interface quite foreign.

> 
> I'm actually quite fluent using all kind of source-to-source 
> conversion tooling, so what I can imagine is that I'll write a
> small conversion utility that converted the hypertex files into
> another format.

Great! Maybe you can discuss this with Kai Kaminshi who has
already done some of this work.

> 
> > In my opinion, that's the best way to discover Axiom.
> 
> I would like to have *any* such documentation, so I totally 
> agree here.
> 
> > Martin
> > 
> > 
> > PS: I just remember, Kai Kaminski has converted some pages 
> to html. But again, I don't think that's the way to go.
> 
> Without going in too much discussion; I still do not understand
> what benefits Hypertex has over other more standard solutions,
> certainly given that currently Windows has *no* Hypertex support.
> 

I think discussion is a Good Thing. You can't have "too much".
:-) But I agree with you. I would very much rather have a standard
web browser interface with functionality similar to Hyperdoc.
Hyperdoc was "neat" in it's day (circa 1990) but today it looks
ugly and awkward. But of course that is just my opinion...

Regards,
Bill Page.