- From: Michel Fortin <michel.fortin@michelf.com>
- Date: Thu, 8 Jun 2006 15:21:29 -0400

Le 7 juin 2006 ? 20:28, Ian Hickson a ?crit : > For something as big as Mathematics, we want to simply re-use an > existing language, not invent a new one. Inventing a new language > for encoding content with as wide a problem-space as mathematics > would require months, as well as the time of domain experts, etc. Defining everything unambiguously is what would take long. If someone wants to define unambiguously everything in a formula, I'd suggest he use MathML, or something else if MathML cannot do the job properly. But most people don't need machine-readable formulas: they need to comunicate mathematics to other *people*. If you take a look at how HTML handles prose, you find that it doesn't try to determine what is a sentence, what is a noun, or what is a verb. But it provides what we need the most: paragraphs, headers, lists, emphasis, etc. I think it should be the same for math content. In my view, HTML should only provide the most needed building blocks for mathematical representation and let authors define semantics beyond the specifications as required. The idea behind the "type" attribute on <var>, <matrix>, and <fence> in my draft proposal, which could also be extended to other elements, is to allow author-defined semantics to be bound to them, semantics which can then be used as hooks for style sheets. Default styles could exist for predefined semantics (<var type="vector"> could put a little arrow over the variable for instance) while authors could create new types and styles for their own purposes when needed. For now, I think creating an inventory of formulas showcasing everything we want to support would be a good first step toward creating, refining, and adopting a formal mathematical markup solution, as well as creating implementations. So I think I'll look into that. Michel Fortin michel.fortin at michelf.com http://www.michelf.com/

Received on Thursday, 8 June 2006 12:21:29 UTC