[Top] [All Lists]

Re: draft-housley-binarytime-00.txt

2004-09-11 05:49:56

Russ Housley <housley(_at_)vigilsec(_dot_)com> writes:

So, if it were an experimental document, you would have not issues?

Experimental would be fine.  As long as there's an indication to the masses
that there's a different degree of importance between support for
extensions/options like MessageDigest and ones like BinaryTime.

(I'd still suggest though that if space saving is the primary concern, an
 approach like the SimpleSigningAttributes one would be better, since apart
 from the 15-20 bytes *per extension* saved you could use further tweaks like
 making ContentType default to "data" (the most common case), using enums
 instead of OIDs (that is, make it a CHOICE between standard-content-type as
 an enum and OIDs for nonstandard types), etc etc etc.  Peter Sylvester's
 suggestion of packing everything in with the data is even more compact, if
 slightly less flexible).