On 8/26/23 21:53, John Levine wrote:
It appears that Theodore Ts\'o <tytso@xxxxxxx> said:It's not that hard to force MUA's to send plain text. But forcing MUA's to send an IETF custom HTML subset may be quite a bit more difficult.I still don't understand what inventing our own HTML dialect would accomplish beyond, as you note, setting up yet another hoop for people to jump through.
I think the "subset dialect" approach only makes sense in the near-term[*] if the list converts each received HTML message to that subset, stripping out the non-conforming HTML.
The benefit is that you (presumably) don't get cumulative layers of completely arbitrary HTML, adding a new layer with each reply. Of course it's at least possible that some MUAs convert subset-HTML in a received subject message to some much more arbitrary version of the HTML, when generating a reply, so the list's efforts to clean up the content are thwarted by those MUAs efforts to mess up the HTML in messages received from the list. But I haven't tried to survey current MUAs to get a sense of how often this actually happens.
Again, the viability of this approach is something that requires experimentation and observation, rather than speculation, to evaluate.
Keith
[*] longer term it's at least conceivable that MUAs would gravitate toward using newly developed standards for HTML in email. Even if those standards weren't the "subset dialect" that we might want, perhaps the standards could specify HTML that's specifically designed to be able to be stripped down to such a dialect.
