No, you can't use MathML here ... yet. It will take a little while, but I'm thinking it will happen. There will be a front page announcement when it is allowed.
Update (April 2004): There are some problems. It looks like MathML is supposed to be in properly formed XML, which E2 sure is not. Also, the validation engine is rather weak, so many MathML features will not be available. OTOH, maybe we could just try it out.
- N-Wing
MathML 2 at W3.org: new window, same window
generic MathML (no version): same window
I started at http://www.w3.org/TR/2002/WD-MathML2-20021219/chapter3.html#presm.mi and worked my way down.
TODO: some MathML examples, to see if user can see it
List of all tags and if they should be allowed. Things I haven't checked yet say so. TODO: I should relook through here; some reasons are now invalid
| tag | meaning | allowing? |
|---|---|---|
| maction | interactive equation | no - neat, but takes more complicated checking than we currently have available |
| maligngroup | no, for now (not clear at a quick glance how to use it) | |
| malignmark | no, for now (not clear at a quick glance how to use it) | |
| math | root tag for all MathML stuff | yes |
| menclose | no, for now (too complicated for a quick check for safety) | |
| merror | error message | no - intended for programs that generate MathML, so it is useless here |
| mfenced | ([{grouping}]) | yes; it may be nice to allow the attributes open, close, and maybe separators later |
| mfrac | fraction | yes |
| mglyph | adds a new character glyph | no need, and requires a bunch of attributes (that weren't checked for safety) |
| mi | identifier - variable, like x | yes |
| mlabeledtr | labeled table or matrix row | no (see mtable comment) |
| mmultiscripts | prescripts and tensor indices (this is sort of like sub- and super- scripts) | yes |
| mn | number | yes |
| mo | operator - functions, braces, etc. | yes |
| mover | something overhead | yes |
| mpadded | adjust space around content | no |
| mphantom | hidden elements - mainly used to get proper alignment (example: vertical adding of variable expressions) | yes |
| mroot | nth root | yes |
| mrow | horizontally group a sub-expression | yes |
| ms | string literal | yes; maybe also allow attributes lquote and rquote |
| mspace | space | probably not, unless this is common in generated stuff - the main thing is it requires some attributes, which can be used to mess up page display (example: gap 100000 pixels wide) (attributes not checked for safety) |
| msqrt | square root | yes |
| mstyle | style change | no |
| msub | subscript | yes |
| msubsup | subscript-superscript pair | yes |
| msup | superscript | yes |
| mtable | table | no, for now (may "need" to allow this though, so we can have matrices and such; in that case, have to check which of the many attributes to allow) |
| mtd | table data | no (see mtable comment) |
| mtext | text - comment, etc. | yes |
| mtr | table or matrix rw | no (see mtable comment) |
| munder | something underneath | yes |
| munderover | something underneath and something else overhead | yes |
Random Thoughts:
- It would be nice to have a superdoc that converts TeX and (anything else?) into MathML. I'm sure not going to write that, though. 8P
- It would be nice to also allow content markup, but that is a whole other set of tags to check, which I don't feel like doing anytime soon. They could be added piecewise, if several people complain nicely.