[Standards] Call for Experience: XEP-0368: SRV records for XMPP over TLS

Philipp Hörist philipp at hoerist.com
Tue Feb 11 16:43:00 UTC 2020

> 1. What software has XEP-0368 implemented? Please note that the
> protocol must be implemented in at least two separate codebases (at
> least one of which must be free or open-source software) in order to
> advance from Draft to Final.

Gajim / nbxmpp

> 2. Have developers experienced any problems with the protocol as
> defined in XEP-0368? If so, please describe the problems and, if
> possible, suggested solutions.

Mixing DIRECT and START TLS is problematic i guess its not supported by any
library that offers some kind of abstraction over sockets.

For example in GLib you can connect to a SRV record, GLib resolves the SRV
record mixes it and does happy eyeballs to connect to the address.
If i have to mix i cant use this, i have to write my own resolver code, i
have to wait on 2 SRV records to finish, and have to mix myself.
Its unreasonable to ask libs to implement such a behavior, seems quite
unique to mix different SRV records.

The benefit is not really clear, Applications would want to favor Direct
TLS its less roundtrips. We are treating something the same which is not
the same. And one has a clear benefit.

This rule will most likely be ignored by clients.

> 3. Is the text of XEP-0368 clear and unambiguous? Are more examples
> needed? Is the conformance language (MAY/SHOULD/MUST) appropriate?
> Have developers found the text confusing at all? Please describe any
> suggestions you have for improving the text.

Remove the mixing of SRV records

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.jabber.org/pipermail/standards/attachments/20200211/db9afbd4/attachment-0001.html>

More information about the Standards mailing list