Re: [Edm] [arch-d] Call for Comment: <draft-iab-use-it-or-lose-it-02> (Long-term Viability of Protocol Extension Mechanisms)

[Vittorio Bertola <[e-mail]>]

>     Il 26/08/2021 06:24 Eliot Lear <[e-mail]> ha scritto:
> 
> 
> 
>     I might point out that some features are intentionally blocked.  Random pings and traceroutes into a network is generally considered unfriendly and have been used as a means to identify points of attack.  There are also features that are unintentionally blocked, simply because they are not understood.  One of the challenges we face is the design choices that permit or deny such blocking.  It's made all the more challenging because protocol designers may not fully understand the weaknesses in their designs.
> 
The "Limiting participation" section is also affected by this. The first paragraph sounds like advocating that we should have less diversity in Internet operations or protocol implementations, so that we can have the few operators agree among themselves on whatever is necessary to make new features work. (This was indeed the approach of the DNS flag day, which was only possible because there are many DNS operators but very few DNS implementers.) Then, the second paragraph provides examples which are rather related to preventing in-network blocking and meddling by middleboxes and network operators.

I think both approaches can help extensibility but create other problems - this should be discussed or at least mentioned.

--

Vittorio Bertola | Head of Policy & Innovation, Open-Xchange
[e-mail] mailto:[e-mail] 
Office @ Via Treviso 12, 10144 Torino, Italy