Re: [port-srv-reg] Fwd: Re: [DNSOP] New version of document for review

[Lars Eggert <[e-mail]>]

Hi,

On 2010-12-2, at 19:04, Magnus Westerlund wrote:
> Joe Touch skrev 2010-12-02 04:46:
>> Catching up, was this ever addressed?
> 
> No, even I forgot about them.

oops.

>>> I've attached a diff with some mostly minor edits. Most of them are
>>> simple English language nits such as:
>>> 1. Comma reduction (a topic which I'm very sensitive to since it's one
>>> of my major faults when writing)
>>> 2. Capitalizing the first letter of bullet points
> 
> Well, if the above are real issues then RFC-editor will do something
> about them.

Agreed. Or we merge it in case we need to rev after IETF last call or IESG evaluation.

>>> More substantive issues, in more or less increasing order of importance.
>>> * In Section 3 I think the readability would improve by switching the
>>> first and second paragraphs.
> 
> Not done, but not that important.

Agree.

>>> * In Section 8.1 (and/or perhaps elsewhere?) I think it would be useful
>>> to suggest (perhaps at the SHOULD level?) that when appropriate the
>>> administrative contact e-mail address should be a role account, and the
>>> problem this is designed to mitigate (individuals sometimes leave the
>>> company/organization that is responsible for the assignment resulting in
>>> a dead e-mail address).
> 
> Not discussed currently, but something that IANA themselves can make
> clear in any form.

Or we add a sentence into the current working version now, to be included whenever we issue one.

>>> * In Section 6 (and elsewhere) there does not appear to be a normative
>>> reference for the division of port numbers into the Well Known,
>>> Registered, and Dynamic categories.
> 
> Still not present in that section.

*Is* there even such a reference? I thought it was a historic artifact.

>>> * Section 7.2 mentions several suggestions to designers for reducing the
>>> number of port numbers that they need for an application. I think it
>>> would be useful to add 2 explicit suggestions to that list, one is the
>>> idea of a "master" application with one Registered port number that can
>>> coordinate communications between the various components of more
>>> complicated applications without requiring each element of the
>>> application to have its own assigned port number. The other suggestion I
>>> think should be made explicitly in the document is the use of multicast
>>> DNS to avoid port number assignments altogether.
> 
> The mDNS part is only present in section 3. However, the current draft
> is much more focused on the basic principles in suggestions on how to do
> anything. I think this bullet belongs in Joe's future guidelines document.

Agreed.

>>> My final area of concern is the idea people have that without an
>>> assigned port number from IANA that no firewall administrators will
>>> allow their traffic. You mention this issue briefly in 7.2, and in
>>> Section 9 (Security Considerations) you include the text that I wrote in
>>> number 2 of "PLEASE NOTE THE FOLLOWING" on the
>>> http://www.iana.org/assignments/port-numbers page, both of which I think
>>> are good things to include. However I believe that it would be useful to
>>> have the whole concept described in more detail in 7.2. In my
>>> communication with port number applicants this issue came up over and
>>> over again, and was either the primary or sole consideration in filing
>>> the application in the first place; resulting in more than one
>>> otherwise-spurious application. I won't quibble if my opinion on the
>>> importance of this topic isn't shared by others, but I felt it was
>>> important to mention it.
> 
> If more discussion around this is wanted, I don't think it belongs in
> the IANA policy document, more in Joes guidelines doc.

Not sure, if it comes up again and again, it may make sense to put it into both documents.

Lars

> 
> So in summary I don't see anything substantial that we have missed. Some
> things for a guidelines doc to consider.
> 
> Cheers
> 
> Magnus Westerlund
> 
> ----------------------------------------------------------------------
> Multimedia Technologies, Ericsson Research EAB/TVM
> ----------------------------------------------------------------------
> Ericsson AB                | Phone  [phone]
> Färögatan 6                | Mobile [phone]
> SE-164 80 Stockholm, Sweden| mailto: [e-mail]
> ----------------------------------------------------------------------
> _______________________________________________
> Port-srv-reg mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/port-srv-reg