John Levine wrote:
In article <584815fc.40...@sorbs.net> you write:
David Hofstee wrote:
The X- type headers are deprecated... https://tools.ietf.org/html/rfc6648

Oh now there's a bad idea if ever I heard one...
If you read the document, you'd know that it said that if people actually
use an X- header it's too hard to change it, and there's not exactly a
shortage of ASCII strings one could use for header names, so if you need
to make up a name, just make one up and don't use X-.

Yeah, I read it... Still a bad idea...

Legitimate eXtension headers as X- are easily filtered as "this is something you shouldn't pay attention to because it's not part of any standard". Take away the X- and you go back to the 'ok what is legitimate and what is not' situation... with a never ending list of updates when people request (and are granted) legitimacy as standards... I for one have a crap load of X- headers, originally I just wanted to create my own, but when I really thought about it - I prepended 'X-' as I realized it was actually a good idea.)

(not that anyone couldn't put random crap in anyhow - but it was more that there was legitimate headers which was a finite list, and then there was the X- headers which were open season... then anything else was "whoooa! what have we here then?")

Regards,

Michelle

_______________________________________________
mailop mailing list
mailop@mailop.org
https://chilli.nosignal.org/cgi-bin/mailman/listinfo/mailop

Reply via email to