[Opensim-users] Groups Implementation Discussion (Ai Austin)

[Ralf Haifisch]

*gg*

Sure - but unless you have a scalable and robust load balanced topology as
e.g. notes/domino, you can avaoid some trouble by having an advance.

The trouble to avoid is enumeration.

Lindens did a "one fits all" group thing.

While security groups are ACL based, they still need some kinda enumartion.


So, at the time people had more and more groups and more and more members
have been in that group and sending group IM´s, the system degraded.   If
this would have been limited to communication it would have been discomfort
- but getting the enumeration in groups getting stale, so influence in
collaboration, it was a pain in the neck.

The advantage would be the goal Charles did outline:  move on to the next
frontier.   You could keep compatibility on one (security) while setting up
a more advanced (maybe XMPP, wich could be a voice basis as well) solution
for the other (communication).

Besides that - yes, collaboration needs communication.  Call me old style -
I still prefer to be able to access files, portals and print over email
@work.  I can still pick up the headset.   So I would prefer to isolate the
communication from security for availability means, as well.


But many thanks for point that both come together, brought me to lindens
problems...

Cheers,
Ralf
------------------------------

Message: 2
Date: Mon, 06 Apr 2009 09:44:39 +0200
From: Dr Scofield <DrScofield at xyzzyxyzzy.net>
Subject: Re: [Opensim-users] Groups Implementation Discussion (Ai
	Austin)
To: opensim-users at lists.berlios.de
Message-ID: <49D9B2E7.7090407 at xyzzyxyzzy.net>
Content-Type: text/plain; charset=ISO-8859-15

Ralf Haifisch wrote:
> Dear Charles,
> 
> 
> That is where i did like your XMPP approach.  
> 
> I guess the rex-chaps can tell a bit more about their experience, they
have
> been playing around with telepathy framework.
> 
> My thought was:
> - diving functionality into security groups (collaboration) and
distribution
> groups (communication)

playing devil's advocate: why should we have two group schemes, one for
collaboration and one for communication? all my collaboration always
involves
communication (i find it a tad difficult to do collaboration without
communicating).

	DrS/dirk
-- 
dr dirk husemann ---- virtual worlds research ---- ibm zurich research lab
SL: dr scofield ---- drscofield at xyzzyxyzzy.net ---- http://xyzzyxyzzy.net/
RL: hud at zurich.ibm.com - +41 44 724 8573 - http://www.zurich.ibm.com/~hud/