[JDEV] Questions on the protocol

[Thomas D. Charron]

On Thu, 26 Aug 1999 11:20:14   Anders Qvist wrote:
>> There's a major problem with that approach, and the problem is that the
>> server would have to "know" where the ICQ transport is.  As it stands now,
>> the ICQ transport could be foobar.server.com and nobody would know the
>> difference.  You could layer that on and say icq:12345 at server.com, but why
>> even have the distinction between different addresses?
>I don't understand. If the machine isn't called icq, the current scheme
>has the same problem?

  Becouse transports are independent.  the name foobar.server.com doesn't mean a thing, but an ICQ address COULD look like UIN at foobar.server.com, with fubar serving as an ICQ transport.  When I send a message to foobar.server.com, I don;t really KNOW what the transport there IS in all actuallity.  I don;t think I responded clearly, but I've got alot going on today..  :-(

>But what if we want to handle addresses for some other messaging system
>that also use mail addresses (eg email)? Heck, any system with a
>distributed model will give the current scheme problem, won't it?
>I'm not sure I understand your intentions. These comments are more to try
>to understand than to argue my own idea.

  We're using a global namespace here.  That means an infinite non centralized system.  Same as email..  When you send an email to someone at somewhere.com, you have NO IDEA what's going to happen to that mail message behind the scenes..  someone at someaddress.com is a standard way of saying 'user at host'.  Finger uses the exact same method.  Just becouse you can finger someone, doesn;t mean you can mail them..  (MOST of the time the email address is the same as the username, but it doesn't HAVE to be).  Perfect example..  Go ahead and try to finger tcharron at my-deja.com.  Betcha don't get what you'd think.. ;-P  Handling email or finger requests would happenm THRU a transport..  Aka, tcharron%my-deja.com at mail.jabber.org.  This is simply an example, and would probrably NOT be the best way to do it..

---
Thomas Charron




--== Sent via Deja.com http://www.deja.com/ ==--
Share what you know. Learn what you don't.