[Standards] XEP-0191 and XEP-0016

Mridul mridul at sun.com
Fri Mar 16 18:19:41 UTC 2007


on-off invisibility is just a special case of selective invisibility ...
Which is one of the reasons we never implemented the other set of 
invisibility proposals : privacy list does that job already for us.

- Mridul

Peter Saint-Andre wrote:
> Tomasz Sterna wrote:
>> Dnia 15-03-2007, czw o godzinie 15:18 -0600, Peter Saint-Andre
>> napisał(a):
>>> OK, folks, make up your minds. We had all sorts of complaints that 
>>> "privacy lists are too hard", therefore I wrote up XEP-0191. So my 
>>> question is: are they complex or not? If not, I would be happy to 
>>> retract XEP-0191 and we can move forward with XEP-0016 alone.
>>
>> My feeling is, that XEP-0191 is a result of lack of the consensus on how
>> would we define simpler "subset-protocol" on top of XEP-0016. Neither as
>> extensions nor as recommendations.
>>
>> There is one thing I like very much in XEP-0191.
>> "If a service deploys both privacy lists and simple communications
>> blocking, the service MUST use the same back-end data store for both
>> protocols."
> 
> And the one thing people *don't* like is the lack of notifications about 
> changes to privacy lists, which are in XEP-0191 but not in XEP-0016. So 
> those features need to be synced up, I think.
> 
> Another failing seems to be the inability to modify a list if it is in 
> use, but notifications might solve that problem.
> 
> Some people want "layers" too, so that you could have things like a 
> corporate privacy list and on top of that a user-defined privacy list, 
> so that may be something we want to discuss as well.
> 
>> Isn't this the very heart of what this XEP-0191 is about?
>> We do like the XEP-0016 but we want to have a simpler semantics for the
>> ones that do not need the full power of Privacy Lists.
>>
>> It's like the PEP and PubSub case. And we managed to get it going.
>> Shouldn't we follow the same road here?
> 
> Quite possibly, yes. :)
> 
>> Let's leave Privacy Lists for the ones that needs them and integrate
>> simple layer on top of it, that just sets simple blocking, without all
>> the hassle of managing full lists client side.
>>
>> And once we have extracted Privacy to XEP-0016 (is there an intention to
>> remove it from RFC3921bis?) 
> 
> It already has been removed from rfc3921bis:
> 
> http://www.xmpp.org/internet-drafts/draft-saintandre-rfc3921bis-01.html
> 
>> we could address the issues we have with it
>> and fix it to get a clean way of doing simple on-off invisibility?
> 
> I still don't think that privacy lists are the right place to do simple 
> on-off invisibility. The volume of modifications to the privacy list 
> concerns me.
> 
> Peter
> 




More information about the Standards mailing list