On 2/27/21 10:00 PM, John Levine wrote:
Indeed, but that was many decades ago. There are some ways in which the
IETF is cutting edge, some in which we are amusingly backward. Most
of the people I deal with can send an e-mail that says "I highlighted
the changes in yellow" and all of their correspondents see the yellow
text. Try that here. Remember that MIME was invented in the IETF and
HTML down the virtual hall from here, both about 30 years ago.
Ok, but to be fair: HTML is a disaster for email. Way back in the
mid-1990s most of us thought it would work out ok, and more likely to
succeed than text/richtext. But we didn't really take the time to
understand the nature of the problem in either case. It's hard to
write a good html editor for email, especially one that handles inline
replies properly, and every single HTML editor for email I know of
botches this. Accidentally delete the line or invisible space before
or after a change in format and it's likely to completely mess up your
formatting, say by merging one correspondent's text with another. HTML
doesn't handle annotations well either because (gasp) text messages are
not naturally hierarchical like HTML (and its *ML predecessors) expect
them to be. HTML hasn't exactly been a stable target either, and
there's lots of variation among MUAs regarding which features are
supported. It's hard to send an email message that looks more-or-less
the same to every recipient.
(And, IMO unfortunately, a lot of MUAs take liberties with presentation
of email messages, which only exacerbates the above problems.)
At the same time HTML is so widely deployed that it's very hard to
deploy something that works better.
The specific behavior you cite above is actually due to a failure of
standardization, because the vast majority of Big Corporate environments
have settled on 1 of about 2 email products overall. Highlighting text
in yellow doesn't work as well in IETF because IETF participants are
(fortunately) still more diverse than Big Corporate employees.
Keith
