[ih] History of Network Operations

[Miles Fidelman]

Jack Haverty via Internet-history wrote:
> Hi John,
>
> IMHO, the Hewitt system is interesting but isn't really very similar 
> to IP Source Routing.  The difference is that, in the IP environment, 
> Source Routing was a mechanism for the *user* of the network to 
> override the routing decisions that the network would normally make.  
> For example, assuming your OS provided a facility for a user program 
> to send and receive "raw" IP datagrams, you could write an app that 
> would send datagrams through the network along a path that the network 
> itself would never choose.
>
> That concept of user-directed routing allowed users to write programs 
> that were useful for debugging network problems, especially situations 
> like performance issues where the "network management" tools 
> themselves reported that "everything is running fine" while the users 
> were complaining "The network is broken. Fix it!".   It also could 
> help verify that the routing mechanism was actually behaving as 
> expected.  Complex real-time distributed software of course never has 
> bugs...?
>
Is that the only reason?

I seem to recall that there were approaches where a local IP process 
would query a "route server" to obtain a route, then insert that route 
into the source-routing fields.  Not so much user-directed as 
distributed routing control.

Miles

-- 
In theory, there is no difference between theory and practice.
In practice, there is.  .... Yogi Berra

Theory is when you know everything but nothing works.
Practice is when everything works but no one knows why.
In our lab, theory and practice are combined:
nothing works and no one knows why.  ... unknown