Re: Old transport-layer protocols to Historic?

[Lixia Zhang <[e-mail]>]

On Jan 7, 2011, at 9:13 PM, Mykyta Yevstifeyev wrote:

> 07.01.2011 21:53, Bob Hinden wrote:
>> Mykyta,
>> 
>> 
>> On Jan 5, 2011, at 9:44 PM, Mykyta Yevstifeyev wrote:
>> 
>>> Hello all,
>>> 
>>> There have been a discussion on tsvwg mailing list about old transport layer protocols - exactly IRTP (RFC938), RDP (RFC908,1151) and NETBLT (RFC998). Initially there have been proposed to define IANA considerations for them. But after a discussion it was found out that it would be better to move them to Historic. I am writing to request more wider discussion on this topic.
>> I see little value even thinking about this.  It's looks like a "make work" project to me.  Just because something is "old", doesn't mean it is "historic" in the sense the label is used in the IETF.
>> 
>> Regarding RDP (RFC908, RFC1151), of which I am one of the authors, both are currently labeled as Experimental.  I do not see any reason to change that.
>> 
>> Bob
>> 
>> 
> Dear all,
> 
> RFC2026 mentions:
> 
>>  A specification that has been superseded by a more recent
>>    specification or is for any other reason considered to be obsolete is
>>    assigned to the "Historic" level.
> and gives 2 reasons for making the RFC Historic: 1) RFC is obsoleted (superseded) or 2) obsolete.
> 
> Obsoleted = made obsolete. This is obvious. When one RFC replaces another, it obsoletes it, and second becomes obsolete.
> 
> What is obsolete (adj.)? Obsolete = deprecated, outdated, out of use, non-current, etc.
> 
> Moreover, RFC2026 does not set any other guidelines for setting the Historic status.

that is because only standard track protocols need such guidelines

> That is why if the protocol is out of use, even specified by Experimental RFC, it is a reason to move its spec. to Historic, in accordance with RFC2026.

First, you said RFC2026 did *not* say anything on moving non-standard protocols to Historic status.

Then you said Experimental RFCs need to move to Historic, in accordance with RFC2026.

Doesn't this sound self-conflicting to you?

Lixia