> The question we are asking you to comment on is:
> 
> Should the working group standardize one data plane behavior for compressing 
> SRv6 information?


Yes.

One. And only one.

Why? 

Let’s suppose that we choose N (or fail to choose).  Different networks will 
make different choices. Then vendors will be forced to support all N of them. 
Chip manufacturers will have to support all N of them. Effectively forever. 
Sure, some may fail, but by that point it becomes impossible to remove the 
support without re-verifying the chip. So no one will.

Choosing more than one is hideously inefficient for the entire industry.  Our 
job, as an SDO, is to make a single standard.

The IETF is neither a coffee klatsch nor a publishing house.  Our duty is to 
make the tough decisions for the world.

Tony

_______________________________________________
spring mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/spring

Reply via email to