W3C home > Mailing lists > Public > www-math@w3.org > October 2006

RE: MathML-in-HTML5

From: White Lynx <whitelynx@operamail.com>
Date: Wed, 04 Oct 2006 20:24:51 +0400
To: www-math@w3.org
Message-Id: <20061004162451.4252143D76@ws5-1.us4.outblaze.com>

> HTML5 will not improve people's authoring skills. However, it will 
> (hopefully, at least!) improve the interoperability of UAs when handling 
> broken pages -- with HTML5 we no longer have "tag soup", because every 
> stream of input characters maps to a single well-defined DOM. 

Yep, Microsoft will take care of it ;)

> > Nobody is against experiments here, just please use your own 
> > unique name (MathML is in use already), your own namespace (if 
> > applicable) and your own content type (that is in case if 
> > experiment goes beyound boundaries of given content type).
> 
> There's no name being used, no namespace being used,

What names are advertised in the title of the thread?

> and the entire point
> of the experiment (as I understand it) is to see if it is possible to do
> it with the text/html MIME type.

Someone said it is impossible?

> > Or alternatively make any changes in markup language through 
> > channels provided by organization that developed this markup, 
> > defined relevant namespace and registered the content type.
> 
> The only markup language being changed is HTML. The only organisation that
> is actively maintaining HTML is WHATWG. The experiment is being done with
> that organisation very much being kept in the loop.
> 
> So everything seems to be being done as you request...

Since we are not talking about HTML as such, we can skip some history lessons, content type registration issues and similar stuff, so basically you can play with HTML as long as it does not harm us. The current reason for concern is MathML which in its present form is already quite problematic language, if in addition someone will replace square weels with triangular ones then we will be simply forced to leave the cart and walk down the road. Following freaks of several different groups of people that push language in several different (and often equally wrong) directions is not an option. Making mathematical markup language is not a rocket science, it is something that existed in 1994 (and even before), was more reasonable, human processable, simple and on structural level even had functionality that MathML, ECMA OMML and similar achivements of XXI century do not provide. Later something went wrong, either it is a matter of beureaucracy, irresponsible attitude, practice of pushing underresearched concepts, involving wrong resources or something else, but the fact is that things are already more complex then they should be (and would be if proper approach would be taken) and the current proposal is to invent new problems without adding ANY new functionality AT ALL. 

We have much more important things to do then playing with the tags (reminds me marquee, blink games), we are have major problems on style language side, that requires both markup and style language to be synchronized otherwise all kind of MathML extensibility without having working formatting mechanisms is myth, everything just reduces to useless retagging games, someone proposes one tagset, others another creating illusion that some progress is made, while fundamental isssues are unattended.

You may argue that one tagging is better then another, but the real issue with proposal is that you drop extesibility, wellformedness, and in addition take MathML further from XML/CSS framework.






-- 
_______________________________________________
Surf the Web in a faster, safer and easier way:
Download Opera 9 at http://www.opera.com

Powered by Outblaze
Received on Wednesday, 4 October 2006 16:24:47 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 20 February 2010 06:12:59 GMT