WebMath: MathML conf. report

[Alan Kelm]

Date: Wed, 3 Jul 2002 17:45:18 +0200
Subject: Re: WebMath: MathML conf. report
Mime-Version: 1.0 (Apple Message framework v482)
From: Paul Libbrecht <paul at activemath.org>
To: webmath at camel.math.ca
Content-Transfer-Encoding: 7bit
In-Reply-To: <200207022221.XAA30029 at penguin.nag.co.uk>
Message-Id: <DFFFFE9F-8E9B-11D6-B045-0003934D43BA at activemath.org>
X-Mailer: Apple Mail (2.482)


On Mercredi, juillet 3, 2002, at 12:21 , David Carlisle wrote:
> I know there has been a lot of interest getting mozilla for mathml on
> the mac, and design science who make the mathplayer mathml extension for
> IE have a public page saying they would like to have a mac version of
> MathPlayer but the Mac version of IE does not have the interfaces
> needed, so a Mac MathPlayer would require a new version of IE for teh
> Mac from Microsoft.

Well, indeed, it's not going to come I fear as it is this ActiveX 
technology which is, in essence, a complete security disaster where 
MacOSX developments tend to care a lot about security.

> I don't personally have easy access to a Mac (and, as I said in response
> to your question at the meeting, I've never used one) but if _anyone_
> does know of (or can help produce) a good plan for mathml on the mac the
> working grouop would certainly be pleased to promote that, and if it can
> be integrated with the XSLT stylesheet to work in a cross platform way
> I'm more than happy to add whatever's needed...

As far as I know there is:
-> a poor (i.e. withou font metrics) port that has worked some time ago 
on special Mozilla Mach-O builds (the Riscky Mozilla), it may or may not 
include MathML in current releases, I did not check. However, for a 
proper rendering, there are two major bugs: Bug 107146 - "ucvmath not 
working with Fizzilla" 
(http://bugzilla.mozilla.org/show_bug.cgi?id=107146)
Bug 74821 - "Implement GetBoundingMetrics() on the Mac" 
(http://bugzilla.mozilla.org/show_bug.cgi?id=74821).
The latter is the hard part, for a good implementation, only ATSUI is 
the good technology and its brand new and pretty hard to master (BBEdit 
didn't manage it yet...). It will bring complete Unicode metrics however 
(which Linux does very badly, e.g. for russian characters).
(several other bugs exist about MacOS implementation of MathML)

-> Amaya under XWindows (requiring some work in terms of installation, 
and giving the poor user-experience of Amaya)

> However at the last MML conference 18 months ago, MathML couldn't be
> easily be rendered in standard versions of any "standard" browser on any
> platform although it was available in special mathml-builds of mozilla, 
> and
> in Amaya. Now at this MathML conference if you are on windows
> or unix then all you need have is the latest version of internet
> explorer, netscape, mozilla, and amaya, and you can easily view mathml
> documents. I hope that we can be pleased with this progress even if
> the Mac is lagging behind.

We all hope it's going to happen, presumably on Mozilla, but the Mac 
community is sufficiently small that its intersection with the Math 
community is very small. Plus, on MacOSX, there's three implementations 
of Mozilla: Carbon Mozilla (slow network, good graphics), Mach-O Mozilla 
(fast network, good graphics, not-mature), Chimera (Cocoa-based, not a 
complete Mozilla (e.g. no XUL, no sidebar), but good graphics and good 
network, very fast)

A central community effort is really needed, I guess.

Paul


--AAA699647.1025766246/camel.math.ca--
-----------------------------------------------------------------
WebMath at mail.math.ca - WebMath Mailing List
To unsubscribe:
via Web:     http://camel.math.ca/cgi-bin/wcms/webmath.pl
via e-mail:  send message a to majordomo at mail.math.ca with
"unsubscribe webmath" in the BODY of message
List Archives: http://camel.math.ca/mail/webmath/
-----------------------------------------------------------------