Hello Pete, John, others,
For Latin Extended-G, my browsers (Mozilla, Chrome, Edge, Vivaldi, all
on Windows 11, but still without this month's system updates) all show
the characters in the U+1DF0x and U+1DF1x ranges, but not the six
characters in the U+1DF2x range. This seems to confirm Martin Thomson's
point in a followup mail. The six character in the U+1DF2x range were
added in Unicode 15.0, published in 2022.
At first glance, this would support the arguments from Pete and John K.
However, if one looks at the document that proposed these characters
(https://www.unicode.org/L2/L2021/21156-legacy-malayalam.pdf), these
characters were mostly used in the first half of the last century. The
chance that they are needed in the name/address/affiliation of an author
of an RFC or ID are *exceedingly low*. Very similar considerations apply
to the other characters in the Latin Extended-G block, and to other
characters new in Unicode versions published in the 2020ies. The Unicode
consortium wouldn't have waited that long to encode these characters if
there were an everyday need for them. So that strongly supports John L.
Regards, Martin.
On 2026-05-22 01:49, Pete Resnick wrote:
On 21 May 2026, at 9:58, John Levine wrote:
It appears that <[email protected], [email protected]> said:
Internet-Draft draft-rswg-rfc7997bis-10.txt is now available. It is a
work
item of the RFC Series Working Group (RSWG) Editorial Stream Working
Group of
the IETF.
Title: Text in RFCs
Author: Paul Hoffman
Name: draft-rswg-rfc7997bis-10.txt
I should have mentioned this nine revisions ago, but it seems to me
that a lot
of this document only applies to the Line Printer Memorial text
format, not HTML
or PDF renderings. The latter two load the fonts they use, so if
xml2rfc can
create the HTML and PDF, browsers and PDF viewers can display them.
I think you're going to have to point out examples other than the two
below that constitute "a lot", because I'm not seeing it.
There are certainly issues of strange characters that are likely to
confuse
people, but the bit in section 2 about "People whose systems do not
have the
fonts needed to display part of a particular RFC" or section 3 about
"many
[characters] whose ability to be displayed is debatable" don't make
sense in
HTML or PDF versions of RFCs.
This is the bit where Klensin is spot on. There are definitely ranges of
characters for which font distribution to assorted OSes and browsers is
at least delayed and often non-existent (Latin Extended-G [1] being the
previously cited example, which was produced in 2021 but still has no
support in MacOS or in Firefox AFAICT) which would count as "not
hav[ing] the fonts needed" or "whose ability to be displayed is
debatable". So I don't think your evaluation is correct, and I don't
think a change to the text is needed.
Pete
[1] Perhaps your browser will do something better with https://
en.wikipedia.org/wiki/Latin_Extended-G that will make it look more like
https://www.unicode.org/charts/PDF/U1DF00.pdf, but on the most updated
version of MacOS with the latest versions of Safari and Firefox, I can't
see them.
--
rswg mailing list -- [email protected]
To unsubscribe send an email to [email protected]
