[Standards] XMPP team activity

Ralph Meijer ralphm at ik.nu
Wed Sep 30 15:23:46 UTC 2015


On 2015-09-29 20:51, Holger Weiß wrote:
> * Ralph Meijer <ralphm at ik.nu> [2015-09-29 19:36]:
>> I strongly believe that having competing *Experimental* specifications
>> are a *good thing*. It helps getting towards a good design that has the
>> potential to go to Draft status and beyond. Note that going to the next
>> phase, Final, requires multiple different compatible implementations.
> 
> You describe the XEP standardization process as envisioned by the XSF.
> In practice, some clients implement XEP-0096 (Draft) for file transfer,
> others use some revision of XEP-0234 (Experimental, since 2008), and
> some do both.  So somehow things aren't working out as envisioned.
> 
> Your stance on this seems to be that it's clearly an implementation
> issue, hence there's no need to question the standardization process.
> It is this stance that I disagree with.

Hi Holger,

Indeed, I describe the XEP standardization process as layed out in
XEP-0001. But the 'vision' is just those who have contributed to the
process over the years, and can be changed with enough support in the
Council. Such changes can be proposed to anyone.

As for how effective it is, depends on what goals you consider. The
primary task of the XSF as a standards body is publishing specifications
and providing a process for their advancement. As I mentioned before,
new enhancement proposals start out as Experimental. If there is enough
interest *by the community*, it will yield (hopefully interoperable)
implementations, discussion and updates to the specification, ultimately
resulting in a request to the Council to consider its advancement to
Draft. If a specification stays unchanged (in text or state) for six
months, it will be deferred.

To go with your example of file transfer, XEP-0096 was mandated in the
first protocol compliance suite: Intermediate IM Protocol Suite
(XEP-0117), that advanced to draft in 2005. It was superseded by the
protocol suites of 2008 (XEP-0213 and XEP-0216) that removed said
requirement.

XEP-0234 on the other hand, as yet has (apparently) not yet achieved
enough maturity to be considered for Draft status. It was even Deferred
in 2013 for inactivity, until people picked it up again.

So all in all, it doesn't strike me as weird that clients implement
XEP-0096 and not so much XEP-0234, yet. It makes sense to implement both
to be able to have more interoperability, at least until XEP-0234 is
both advanced to at least Draft and has sufficient client support.

Again, the XSF is not driving implementations. It facilitates the
process of standardization and the discussions around it. It is up to
people (whether Council, Board, other XSF member, or not), to drive
implementations. If you are unhappy with the status quo around file
transfer (I am, too, to be sure), you can take action to convince,
coerce, bribe, beg, or pay people (or yourself) to get implementations
and implementation experience and then ask Council to advance to Draft.

I personally don't see a way to change the standardization process we
have to improve on this.

-- 
Cheers,

ralphm


More information about the Standards mailing list