Maxim,

>> There are multiple such efforts. SubWeb (one of such efforts) page links
>> to these: https://h3rald.com/subweb/#heading__Prior_Art. The main
>> takeaway is: HTML 3.2 + semantic HTML 5 elements are the best bet right
>> now. This is indeed pretty close to what most SRFIs look like, so maybe
>> there’s nothing to change besides the ethos?
>
> I haven't read the above link, but placing bounds on what HTML can be
> used in the SRFI source documents should be good thing for downstream
> users wanting to parse them. I'd welcome such a requirement; are there
> existing tools that can validate the document is valid SubWeb?

There are no validators for SubWeb yet, but Fabio (h3rald) has expressed
the intent to make one, albeit in an indeterminate timeframe.

However, there is a validator for smolweb HTML
<https://smolweb.org/validator>, which I prefer over SubWeb. My own SRFI
273 <https://srfi.schemers.org/srfi-273/srfi-273.html> mostly passes it!

> I also
> assume that existing HTML parsers like (htmlprag) are capable of parsing
> it, which they ought to if the format really is a subset of HTML.

Yes, both SubWeb and smolweb HTML are mere subsets of HTML (or XHTML in
case of smolweb HTML,) so they both should be parseable with (htmlprag)

Thanks,
--
Artyom Bologov
https://aartaka.me