On 25 August 2011 01:41, John C Klensin wrote: >> still new (= interesting for readers) and maybe unique.
> Yes, except that DKIM doesn't necessarily sign _all_ headers, > so mentioning it specifically in this context requires a lot > more detail... IMHO "e.g., [DKIM]" would not require more details, but I'm not hot about it, and certainly agree that more details are wrong: <http://www.ietf.org/mail-archive/web/yam/current/msg00769.html> > If a primary goal is to mention (advertise?) DKIM Not by me. I just happen to know that DKIM is not like PGP or other signatures; DKIM could have "a high astonishment factor" for 4409bis readers who think to know what MICs and signatures are, but don't know DKIM. >> "Gmail, fix the SPF FAIL for me, will you." I fear my >> administrative control has limits, as outlined in RFC 5598 >> figure 4 s/transit/SUBMIT/. > Please note "formally" and observe that, if you don't like > what Gmail is doing, you are not obligated to use them as an > address or submission server. Maybe I'll revive the GMX account where SPF FAIL works, or use @xyzzy.claranet.de again. While we evaluate the last comma in present or absent references, somebody said that an informative RFC 5332 reference is no downref. Checking RFC 4897 after Sam and you talked about it I found that nothing is wrong with the RFC 5332 downref note: | At the option of the author, similar notes may be attached | to non-normative references. -Frank _______________________________________________ yam mailing list [email protected] https://www.ietf.org/mailman/listinfo/yam
