[Standards-JIG] MUC (JEP-45) privacy & control

Trejkaz trejkaz at trypticon.org
Mon Apr 17 13:34:47 UTC 2006

On Monday 17 April 2006 22:47, Ian Paterson wrote:
> I understand that a lot of developers have used and are fond of IRC. But
> do we need to care quite so much about IRC users that, in order to add
> an IRC feature, whose benefits could arguably be achieved in a better
> way, we complicate our protocol? Doesn't that go against the Jabber
> philosophy of simple clients?

It's true that if the privacy filtering has to be done on the client, then the 
client needing to track a user's changes in nickname and so forth would be a 
complex hassle.

In general the solution to this sort of problem is not to reduce 
functionality, but rather to move the filtering to the server.  In this case, 
perhaps it could be moved to the MUC component, in which case MUC would just 
need a simple block/unblock protocol.  This might be similar to what people 
were talking about for invisibility, but with room nicknames instead of JIDs, 
which the MUC service then translates into the real JID (hence anonymous or 
non-anonymous are handled identically.)

> AFAIK the vast majority of people who chat use Web based chat rooms, not
> IRC clients. In my experience, the people who use these services hate
> the abuse that goes on more than anything else. Features like anonymous
> and semi-anonymous rooms just make it easier for the abusers.

Whereas this is true, it's definitely a cultural thing.  There are many 
examples out there on the Internet where an almost perfectly anonymous forum 
doesn't degrade into a flame war.

In reality, someone who likes abuse enough will just get themselves another 
JID, so you're only making life slightly harder for the people performing the 
real abuse.  An analogy might be giving everyone in the world an ID card, as 
a means of reducing crime.  The real criminals are the ones printing new fake 
IDs, and it's not like knowing who someone is prevents them performnig the 
crime in the first place.

> It is far too late to make any changes to JEP-0045 that aren't 100%
> backward compatible. However, perhaps we could lower the bar for
> potential implementors simply by adding a note to make it *clear* that
> both server and client support for anonymous and semi-anonymous rooms is
> optional.

As far as the client goes, what extra does a client have to implement in order 
to support anonymous rooms, vs. not supporting them?  I was largely under the 
impression that the main difference was being able to see the identity of 
people in the room, in which case an anonymous room would require less 
functionality on the client.


             Email: trejkaz at trypticon.org
         Jabber ID: trejkaz at trypticon.org
          Web site: http://trypticon.org/
   GPG Fingerprint: 9EEB 97D7 8F7B 7977 F39F  A62C B8C7 BC8B 037E EA73
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 191 bytes
Desc: not available
URL: <http://mail.jabber.org/pipermail/standards/attachments/20060417/fc35d6d9/attachment.sig>

More information about the Standards mailing list