[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [Axiom-mail] mathaction

From: Bill Page
Subject: RE: [Axiom-mail] mathaction
Date: Thu, 9 Sep 2004 14:12:23 -0400

On Thursday, September 09, 2004 7:37 AM Martin Rubey
> I was just browsing and I'd 
> like to make a few suggestions -- apart from repeating that
> I find it just great!

> * I like 
> a *lot* better than
>   I find it easier to read and use, better structured and
>   simpler-looking.

I agree, however from my point of view the Plone portal
environment plays a rather different role than the pure wiki
environment (MathAction) even though some of the functionality
of these two environments is the same. Some the differences
are purely technical design issues, some are historical and
some are more fundamental.

Plone ( originates from a web content management
development stream. The idea is to provide a way to maintain
a certain "corporate look and feel" but to allow the content
of a web site to be created and managed in a distributed
manner. Plone provides a number of different kind of "objects"
(only some of which are currently implemented in the Axiom
portal). Objects can be things like "file folders" that
contain other objects, then can be "documents" that display
as nicely formatted HTML pages, they can be structured things
like bibtex entries, events of news items that are linked by
time and date to a calendar, or they can be external files such
as PDF documents or binary files such as programs that need to
be downloaded to be accessed by the user.

Plone does this in a integrated and controlled manner that
enables "draft" content to be created by users and to be
"published" by a site manager etc. Objects in Plone have a
"state" that is related to the workflow associated with the
lifecycle of an object. Users of the portal have associated
"roles" and the combination of object state and user role
determines what can be done to/with each object. It even
allows a developer to create new object types to be created
and managed entirely through the web. Yes, this is a rather
complex environment.

*One* of the object types in Plone is a 'wiki page' that has
the same functionality as the pages that you see in MathAction.
This is where the main overlap between these two environment
exists. Unfortunately, technical limitations mean that it is
not possible for the actual content from one environment to
be directly accessed from the other.

The wiki web interface has a very different origin than content
management - considerably less 'structured'. In a wiki the
emphasis is on collaborative web page authoring. But many
wikis do not even require the authors to identify themselves.
Besides just "for the pleasure of communicating and sharing",
wikis have been used with at least moderate success for joint
distributed authoring of documentation.

The wiki that implements MathAction and that also provides
wiki objects in Plone is called ZWiki ( Both
Plone and ZWiki are implemented using the object-oriented
database and application development environment called Zope.
Besides Plone and ZWiki there are many other web applications
that have been built using Zope. Zope in turn is implemented
in Python. And oddly enough, the Python language bears a
great similarity to the Axiom programming language. ZWiki had
already been extended by Bob McElrath to include rendering of
LaTeX coding, so the extension of ZWiki to provide the Axiom
and Reduce interface was as they say "almost a match made in

> Since it seems that there are things on Zope/Plone which
> are not on Zope/Mathaction, I suggest that you move those 
> things worth moving to MathAction and make MathAction /
> FrontPage the principal page, which really should become
> axioms main homepage, I think.
>   [This part I cannot do]

I am very glad that you are thinking about this. It is
possible (I think) to make the first page that you see in
Plone a wiki page. Then, besides all the other bells and
whistles provided by Plone, the initial interface between
these two systems would be the same. But there are many
things in the Plone that are not possible (or at least
awkward) in the pure wiki environment - for example, the
bibtex literature references and management of event
notices etc. What specifically would you like to have in
MathAction that is now currently only in Plone?

*If* we continue to maintain two different environments
like this, then I think the most natural progression would be
for some things to be developed *first* in the more open
and less structured MathAction wiki environment, and then
if they have a more lasting relevance, to move into the
Axiom Portal Plone environment for longer term access and

> In the same spirit, the links to the mailing lists and to 
> savannah should be added to MathAction/FrontPage and the
> page should be tidied up a little.
>   [I can do this, if somebody tells me: go ahead]

Please go ahead! You don't really need permission.
> * The table of contents needs cleaning up. I suggest:
>   - mathaction/AxiomAlgebra should contain (as subpage)
> the various layers as it does now
>   [I can do this, if somebody tells me: go ahead]

Yes, please go ahead.

>   - each layer should contain the docs for the types it 
> provides. (as layer0 contains attreg and dhmatrix now)
>   [I can do this, if somebody tells me: go ahead]

Yes, I have very glad you are interested in this. I think
it has a lot of potential to make Axiom's algebra more
accessible to everyone. And also to make it possible to
cope with the complexity and get on with making improvements.
It seems to me that one thing that has happened to Axiom
(and some other computer math systems) is that it has reached
a kind of "plateau of development" that is constrained by
the available tools. Maybe (just maybe) this is one way to
move ahead.

>   - it would be great if these pages would be *identical* to 
> the pamphlet files. It should be possible, shouldn't it? Maybe
> you could make a new style called "pamphlet", which produces
> LaTeX in the background using the "document" command and then
> applies "latex2html"? This way, we could move the complete
> src/algebra directory at once into the wiki!
>     I believe that this would considerably speed up the
>     transition to documented code!!!
>   [This part I cannot do, but this would be number one priority]

Yes, I agree completely and I am actively working on this.
More details later and probably on the Axiom Portal itself.

> Well, that should be enough for the moment.
> thanks again,
> Martin

You are welcome. And thank you for your contributions so

Bill Page.

reply via email to

[Prev in Thread] Current Thread [Next in Thread]