David -- (Private) I know very few mathematicians who have thought about it who "trust" MathML. For myself I will be prepared to live with it, but I doubt that it achieves what it claims beyond the level of engineering mathematics. Have you seen my draft on notation? He may have circulated it to the cathedral chapter. But nobody replied, so I don't know. If that body feels that it lacks the expertise, it should get help. The draft may be found at http://www.albany.edu/~hammond/gellmu/notation. It is plain text and it is *very* drafty. (I think that I showed it to Miner.) Are you familiar with Richard Fateman's experiment with a table of integrals? Oh! Eitan Gurari's stuff with Sebastian brings up the question whether the MathML Cathedral Chapter understands that DVI, like the new scalable vector graphics (SVG) proposal at W3C, is XML. And did Adobe understand this when they made PDF? Hmmm... DVI processors usually seek, as I recall. (But probably not Geoffrey Tobin's "dv2dt".) That means that DVI is not browser ready. Still at an authoring site you could upgrade it (analogous to Eitan Gurari's game, but with regular special-free DVI) and hope to make browser fodder, couldn't one? I would not be a good quick judge of whether it is practical; I'm just not close enough to DVI. -- Bill