Comment on section 3.2 The Keywords header.
As I understand it, there is no precedent for giving specific keywords
specific meanings in this context. One could therefore object that this
proposal is not properly backwards compatible with existing practice, because
existing use of these keywords may not carry the specific meaning used here.
Not that I've seen the "keywords" field used in practice, or used it myself
outside of obscure humour value in Usenet posts many years ago. But hey --
devil's advocate.
I see two possible ways around this. One is to define a specific new field;
the other is to make the keywords a bit more specific. Obviously Eric was
attempting to avoid defining new fields, and that's a fair tack, but defining
a new field is an option. To make the keywords more specific, we could have
"can-spam-advertising" instead of just "advertising" (using "can-spam-" as a
namespace of sorts), to reduce the plausibility that it will conflict with
any existing usage.
On the matter of internationalisation, I would say that these are meant
primarily to be machine-readable tokens. The MUA can translate the tokens
into an appropriate natural language description or icon, etc.
As an aside, I can see the mere presence of the "keywords" field becoming a
major touchstone for spam if this proposal is adopted.
Regards,
TFBW
_______________________________________________
Asrg mailing list
Asrg(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/asrg