From: Charles Lindsey (chl@clerew.man.ac.uk)
Date: Fri Jun 06 2003 - 05:19:32 CDT
In <Pine.BSI.3.91.1030605125139.17018C-100000@spsystems.net> Henry Spencer <henry@spsystems.net> writes:
>On Wed, 4 Jun 2003, Russ Allbery wrote:
>>
>> Well, that's weird. Since when do standards specify the semantics of
>> syntactically invalid data?
>Son-of-1036 did syntactically permit empty headers, although it tried to
>discourage their use. So there was no inconsistency there.
>I don't believe our current draft actually succeeds in outlawing all empty
>headers, if only because of the other-header escape hatch.
>In any case, it is reasonable to state how common error situations should
>be handled, in addition to describing the semantics of correct data.
Indeed. It is telling you how to be liberal in what you accept.
And actually, as our draft stands, it is not syntactially invalid. We do
not define any headers where it is possible, we say that they are a Bad
Thing, and we say that posting and injecting agents SHOULD remove them if
they see them (hence the need to say that makes no semantic difference).
Exactly what Son-of-1036 said, in fact.
-- Charles H. Lindsey ---------At Home, doing my own thing------------------------ Tel: +44 161 436 6131 Fax: +44 161 436 6133 Web: http://www.cs.man.ac.uk/~chl Email: chl@clerew.man.ac.uk Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K. PGP: 2C15F1A9 Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5