[ih] Exterior Gateway Protocol

[Guy Almes]

Brian,
   I have literally not thought of this since about 1990, but ...

   Designing such an oracle would not so much be technically difficult, 
but it would surely fail for non-technical reasons.
   For example, with total inter-AS link-state knowledge, and with a 
definition of what "good" is, you could probably quickly compute a good 
set of routes.
   But I suspect that different competing ISPs would have a different 
definition of what "good" means.  This was very evident even in 1990.

   	-- Guy

On 9/2/20 10:16 PM, Brian E Carpenter wrote:
> On 03-Sep-20 12:47, Tony Li via Internet-history wrote:
> ...
>> Namely, it explores every possible loop-free recovery path until all possible paths are known to have failed. Yes, this can be painful, but what can you do short of an oracle?
> 
> Was an oracle ever seriously contemplated for the Internet?
> 
> SNA sysgen amounted to an oracle, but was of course notorious because the whole network had to be restarted to install a new set of tables.
> 
>     Brian
>