Re: [tsvwg] CALL to revoke last call: Re: Request for working group feedback on draft-kuehlewind-system-ports (6th March, 2020)

[Joseph Touch <[e-mail]>]

PS - minor tweak to be VERY clear:

	"Reassignment of Standards-Track Ports to the IESG"

The IESG is hereby indicating to the community its intent to transfer assignment to the IESG the ports currently assigned to standards-track protocols (that are not already so assigned) and that are currently assigned to parties that are no longer reachable, as determined by IANA and the IAB over the next calendar year, and this Internet-Draft is intended to expire at the end of that process.

> On Feb 18, 2020, at 7:24 PM, Joseph Touch <[e-mail]> wrote:
> 
> Cutting to the key points, hopefully to move forward:
> 
> I don’t think there’s any value in this doc trying to define a process that’s already defined in RFC 6335. It should have exactly one line of text and a changed title: 
> 
> 	"Reassignment of Standards-Track Ports to the IESG"
> 
> The IESG is hereby indicating to the community its intent to transfer assignment of ports assigned to standards-track protocols and to parties that are no longer reachable, as determined by IANA and the IAB, and this Internet-Draft is intended to expire at the end of that process.
> 
> Everything else steps on RFC 6335 in one way or another, is unnecessary, and should be removed.
> 
>> ...
>> [MK2] I thought I replied to all you mails. I noted this point. However, I believe if we do any action here, we should also go for a clean-up in order to detect outdated contact information and potentially not-used ports, as I already said several times. I'm happy to get more input from others on this point! 
> 
> The sentence I provided above is the most that is needed and allows for reassignment of “orphan” ports.
> 
> “Not used” ports is a completely different issue with a MUCH higher bar  for deassignment- it requires confirmation of the port not being used even in legacy systems in order to deassign. That’s a dangerous area to try to “clean up” in the ports and should be avoided unless absolutely critical on a per-case basis.
> 
> One final point:
>> 
>> [MK2] I didn't say that an RFC is require, I was just saying that my expectation is that most new assignmenst in the system range will have the IEFT as assignee and contact (and that is something independent of the assignment procedure).
> 
> That’s simply incorrect. Right now, nearly NO new assignments will be made in the system range as per RFC 7605 anyway.
> 
> Joe
>