Hello everybody, [Dear IANA, please read this mail carefully. It contains some requests to you regarding the Character Set registry, in particular regarding the fact that the alias "ASCII" for the "Character Set" with preferred MIME name "US-ASCII" was lost somewhere. Thanks!] On 2023-12-18 09:21, Martin J. Dürst wrote:
Hello Stephen,
On 2023-12-16 04:06, Steffen Nurpmeso wrote:
To add that for backward compatibility the plain ASCII alias cannot go away,
I seem to remember too that ASCII was listed as an alias, and have confirmed this with https://web.archive.org/web/20051229042158/http://www.iana.org/assignments/c...
Stephen, maybe you can do a bisection to find out where this alias disappeared.
I have done this bisection (actually it is currently a trisection, because I haven't found a way to get redirects to .xml files working in the WayBack Machine (maybe such files were never archived). Earliest version without "ASCII" (after redirect) Last Updated 2013-01-23 https://web.archive.org/web/20130702181028/http://www.iana.org/assignments/c... Latest version with "ASCII: (last updated 2012-09-26) https://web.archive.org/web/20120930075956/http://www.iana.org/assignments/c... In between these two, redirect to XML, not visible in WayBack Machine: As suggested by Steffen, that the "ASCII" alias was lost in the transition around 2012/2013 from plain text to XML seem highly probable. I have looked at the various archives for the ietf-charsets mailing list, but in that time period, there were extremely few messages, and none in any way talked about removing the "ASCII" alias. IANA, can you please check your archives (version control, messages, whatever else you have) and try to find out actual or potential reasons for why the "ASCII" alias got dropped/lost? In case this was a clerical error, can you please check the rest of the Character Set registry for any potential similar errors? Regards, Martin.
<charset reviewer hat on> I don't remember ever having dealt with a request to remove this ALIAS, and I strongly doubt that Ned ever did that. </charset reviewer hat on>
i do have emails with charset=ascii in my archive,
It would be interesting to know how many (i.e. what percentage of overall mails, and what percentage in comparison to those labeled US-ASCII), and how old they are.
Regards, Martin.
and GNU "iconv -l" gives
$ iconv -l|grep -i ascii ASCII// CSASCII// US-ASCII//
so if there are any character-sets downstream consumers, let alone automated ones, they need to have the alias also in the future, how deprecated it may be. (For the MUA i maintain i fixed it manually.)
--steffen