The consultants don't mention LaTeX as a potential solution. I believe that's because MathML is considered more accessible than LaTeX (eg native support for MathML in screen readers). We could ask them about it to make sure my understanding of this is correct, though.
But also, i think that LaTeX can be easily converted to MathML, right? So if authors write in LaTeX and convert it to MathML prior to publication, does that seem reasonable? On Thu, Jan 29, 2026, 6:13 PM Brian E Carpenter <[email protected]> wrote: > Paul, > > The LaTeX and MathML representations are equivalent and readily > convertible. > > pandoc --mathml try.tex -o try.html just works, and apparently there is an > underlying tool called texmath if you don't want to use pandoc. > > pandoc --mathml try.html -o try2.tex also works to convert MathML back to > LaTeX. > > I round-tripped \(K = \sqrt[3]{\frac{W_{max} - cwnd_{epoch}}{C}}\) and it > worked perfectly. > > So modulo tooling, it doesn't really matter which we specify. But LaTeX is > certainly much more friendly for sighted users. I don't find the following > much use to me: > > <p><math display="inline" xmlns="http://www.w3.org/1998/Math/MathML > "><semantics><mrow><mi>K</mi><mo>=</mo><mroot><mfrac><mrow><msub><mi>W</mi><mrow><mi>m</mi><mi>a</mi><mi>x</mi></mrow></msub><mo>−</mo><mi>c</mi><mi>w</mi><mi>n</mi><msub><mi>d</mi><mrow><mi>e</mi><mi>p</mi><mi>o</mi><mi>c</mi><mi>h</mi></mrow></msub></mrow><mi>C</mi></mfrac><mn>3</mn></mroot></mrow></semantics></math></p> > > Brian > > On 30-Jan-26 14:31, Paul Hoffman wrote: > > Given the fairly concrete analysis below, it feels like the draft should > have an explicit policy for using MathML for artwork that primarily > contains math. > > > > --Paul Hoffman > > > >> On Jan 29, 2026, at 17:27, Alexis Rossi <[email protected]> > wrote: > >> > >> We hired expert accessibility consultants to review some documents and > answer some questions to help guide our efforts for rfc-editor.org [ > rfc-editor.org], and along the way we also asked them specifically about > how to improve representation of math in RFCs. You can refer to these > documents for the outcome of that: > >> > >> 1. https://www.ietf.org/media/documents/RFC_Documents_Spot_Checks.pdf [ > ietf.org] > >> 2. > https://www.ietf.org/media/documents/RFC_Data_Visualizations_Accessibility_Guidance.pdf > [ietf.org] > >> > >> The first document's executive summary says that our current > "Mathematical formulas are completely inaccessible to blind screen reader > users." > >> > >> Some other takeaways from that first doc which are relevant: > >> - paraphrasing: people with visual disabilities are most likely to use > the HTML format for access (the other formats have major drawbacks, > including txt which lacks navigational structure) > >> - "math is not text and should not be treated as such. For example, > some alt text that sounds understandable in English speech will not be > translated into proper mathematical braille or other speech languages." > >> > >> In both of these documents, the accessibility experts recommended using > MathML. > >> > >> I'll quote the second doc starting on Page 15 because it's more > extensive: > >> > >> "Use MathML for mathematical content, which allows browsers to > translate the markup language into properly formatted equations. MathML > ensures that such content is consistently rendered across different > browsers and platforms, enhancing accessibility for users who rely on > various devices and software. For example: > >> - Screen magnification users can enlarge the equation while maintaining > its integrity and the clear visual distinction of each character. > >> - MathML content is communicated in an intelligent and meaningful > manner for screen reader users who can further explore the different parts > of the equation as desired. For a demo on reviewing a sample MathML > equation using the NVDAscreen reader, refer to the following YouTube video: > NVDA and Firefox ReadingMathML on Wikipedia. > >> - MathML is designed to work seamlessly with access technologies, such > as braille displays and speech recognition software, providing a more > inclusive experience for all users. > >> - Neurodivergent users or people with reading disabilities can > customize the formatting of text without altering the meaning of the > equation." > >> > >> On Thu, Jan 29, 2026 at 12:46 PM John R Levine <[email protected]> wrote: > >> On Fri, 30 Jan 2026, Brian E Carpenter wrote: > >> > >>> On 30-Jan-26 08:00, Carsten Bormann wrote: > >>>>> K = \sqrt[3]{\frac{W_{max} - cwnd_{epoch}}{C}} > >> > >>>> The “accessibility” use case can be easier to address if the source is > >>>> available with the RFC. > >>> > >>> True, but does anybody here know what a screen reader would do with > that example? > >> > >> My guess is that there are enough blind mathematicians that there are > >> screen readers that can read MathML or LaTeX, but once again, we need to > >> talk to people who know about this rather than guessing. > >> > >> R's, > >> John > >> > >> -- > >> rswg mailing list -- [email protected] > >> To unsubscribe send an email to [email protected] > >> -- > >> rswg mailing list -- [email protected] > >> To unsubscribe send an email to [email protected] > > >
-- rswg mailing list -- [email protected] To unsubscribe send an email to [email protected]
