[Standards] hash agility in file transfer

Peter Saint-Andre stpeter at stpeter.im
Wed Jun 15 19:51:36 UTC 2011

On 6/15/11 1:35 PM, Dave Cridland wrote:
> On Wed Jun  1 23:55:47 2011, Waqas Hussain wrote:
>> But if we go with tags, do we even need to specify that? We can always
>> add tags later in new namespaces, even defined in new XEPs, e.g.
>> <hash xmlns="new sha1 xmlns">...<hash>
>> or
>> <sha1 xmlns="new XEP-0096 namespace, or even the existing namespace,
>> since the updated protocol is fully backwards compatible">...</sha1>
> I dislike this, because it's harder for an app to say "Hey, this app
> sent me a hash function I don't understand", which can be interesting
> (perhaps not in FT, but in other cases).
> Also, it requires more work than a textual name, since we have those
> defined and maintained for us at
> http://www.iana.org/assignments/hash-function-text-names
> So therefore I'd like to suggest we consider an element: <hash
> xmlns='urn:xmpp:crypto:hash' function='sha-1' format='64|hex'>...</hash>
> I appreciate this isn't as exciting and shiny as a new namespace and
> element name for every hash, but in effect, this has already defined 8
> hash functions, so we can move on and concern ourselves only with which
> one we want to recommend for now.

How is support for each function discovered?


Peter Saint-Andre

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 6105 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://mail.jabber.org/pipermail/standards/attachments/20110615/7444e780/attachment.bin>

More information about the Standards mailing list