[Standards] Proposed XMPP Extension: Explicit Message Encryption

Peter Saint-Andre stpeter at stpeter.im
Fri Sep 2 02:40:24 UTC 2016


On 8/22/16 1:58 AM, Florian Schmaus wrote:
> On 17.08.2016 20:48, Emmanuel Gil Peyrot wrote:
>> On Wed, Aug 17, 2016 at 01:31:28PM -0500, Sam Whited wrote:
>>> Having the name be an attribute doesn't seem very friendly to me. What
>>> if I want it in a different language or locale? How do RTL languages
>>> work? Even if we assume everything is always English, what if my text
>>> doesn't work well with all names "an [OTR] encrypted message" works
>>> fine, but "an [PGP] encrypted message" does not where the bracketed
>>> word is substituted.
>>>
>>> I'd leave off the name attribute entirely; it's not necessary.
>>
>> The rationale for the presence of this name attribute is to have a
>> placeholder in case the client doesn’t know anything about the
>> encryption method in question.
>
> The name metadata should either be qualified by a lang or simply be
> removed. I feel like removing is the right thing: Clients/Libraries
> either know the encryption mechanism by its namespace, and are possible
> able to tell the user to install a certain plugin to enable the
> encryption mechanism. And otherwise they could simply say "The received
> message was encrypted with an unknown mechanism (jabber:x:encrypted)".

Agreed.

> That said: I really like the EME XEP.

Yes, it seems fine. +1 to publishing it (if the Council folks are paying 
attention).

§3.2 mentions a list of supported protocols. What happens if we add 
another supported protocol to EME? (Because, you know, we have so many 
end-to-end encryption protocols!) Do we create a registry of supported 
protocols and add another to the registry? Do we update the spec, 
perhaps even with a change to the namespace version?

Peter




More information about the Standards mailing list