It appears that Carsten Bormann <[email protected]> said: >On Jan 28, 2026, at 05:57, John Levine <[email protected]> wrote: >> >> I also am not understanding how this is going to work in Legacy Line Printer >> Format (sometimes misleadingly called "plain text") unless it's just going to >> dump the MathML into the document. > >I think RFC 9438 shows nicely how this can be done. > >[1]: https://www.rfc-editor.org/rfc/rfc9438.txt
While that is a tour de force of ASCII art (UTF-8 art?) it impresses me as exactly what we should *not* do. It assumes that the display device has a full repertoire of line drawing characters, in some of the displayed equations you can't tell what lines are supposed to be subscripts or superscripts if you don't already know what they mean, and it's hard to imagine a screen reader rendering it as other than gibberish. It doesn't help that it overloads underscores, some of which mean _italic_ and some of which are parts of identifiers. While I have some guesses about what would be better, there have been blind mathematicians as long as there have been mathematicians. Surely the RPC can find people with experience who can tell us what other math publishers do, and how we can publish math so it's readable by as many people as possible. R's, John -- rswg mailing list -- [email protected] To unsubscribe send an email to [email protected]
