Re: Last Call: <draft-eggert-tcpm-historicize-00.txt> (Moving the Undeployed TCP Extensions RFC1072, RFC1106, RFC1110, RFC1145, RFC1146, RFC1263, RFC1379, RFC1644 and RFC1693 to Historic Status) to Informational RFC

[Mykyta Yevstifeyev <[e-mail]>]

Benjamin,

RFC 4644, that is a reason to have this document written says:

>   The RFCs listed here define extensions that have thus far failed to
>     arouse substantial interest from implementers, or that were found to
>     be defective for general use.

I. e. these options are out-of-use and did not gain the implementators' 
popularity. Nearly the same is with IRTP, which I referred to.

Nevertheless I agree that this purpose of this draft is to say "if you 
implement TCP you don't need to implement these bits anymore"; the 
purpose of moving i.e. IRTP to Historic is contiguous - "this technology 
was eventually defined but revealed itself unacceptable/uninteresting; 
further implementations are discouraged"

Mykyta Yevstifeyev

04.02.2011 2:00, Benjamin Niven-Jenkins пишет:
> Mykyta,
>
> On 3 Feb 2011, at 15:03, Mykyta Yevstifeyev wrote:
>> However I'd like to raise some questions not directly connected to this document.  I wonder why those who said a few weeks ago that historicizing some documents in the similar situation is not appropriate do not object now.  The arguments of these folks were that RFC 2026 sets the criteria for Historic status as 'replaced by other doc' and did not consider 'being deprecated' (what exactly we have in the current case) as weighty reason for historicizing document.
> I am not sure which of the many "move to historic" proposals you have proposed recently you are referring to but IMO there is a difference between your proposals and that of Lars, namely:
>
> Your proposals fell into one of two categories:
> 1) Protocol X is old so we should make it historic for housekeeping reasons
> 2) URI Y has never been used so we should make it historic
>
> Whereas while Lars' document is doing some housekeeping it is really saying "if you implement TCP you don't need to implement these bits anymore" so it has a clear value to people writing new TCP stack implementations.
>
> In comparison your proposals were housekeeping for the sake of housekeeping and provided no value to the wider community.
>
> HTH
> Ben
>
>