[rfc-i] Classifying pre-IETF RFCs

Joe Touch touch at isi.edu
Wed Dec 1 07:04:18 PST 2010



On 12/1/2010 6:59 AM, Mykyta Yevstifeyev wrote:
> 01.12.2010 16:48, Joe Touch wrote:
>> Hi, Mykyta,
>>
>> On 12/1/2010 1:21 AM, Mykyta Yevstifeyev wrote:
>> ...
>>> To my mind, if RFC Database reflects the status of RFC at the
>>> moment of publishing, RFC 1000 is OK for classifying pre-1310
>>> RFCs (as RFC 1310 make the classifying system 'modern').
>>
>> You should consider that RFC 1000 was NOT published at the time RFCs 1
>> through 999 were published. I.e., if you agree that "status at the
>> time published" is the right rule, then you MUST ignore RFC 1000's
>> summary of it's informal organization of RFCs 1-999.
 >
> RFC 1000 is only listing RFCs 1-999 and mentioning their categories.
> However not all RFCs 1-999 contain the category itself.

That's because RFC 1000 - or RFC 100 (as below), doesn't set that sort 
of categorization up.

These RFCs say "this is how this RFC helps categorize what's been 
published", not "and authors should indicate what category their doc 
should be in when they publish them".

I.e., these are roadmap docs. Roadmap docs do not create streams.

>> And RFC1000 did not proscribe any classification of later RFCs, so no,
>> it's not appropriate to apply it to pre-1310 either.
>>
>> These ~1300 RFCs were not classified when published.
 >
> No, they were. Look at RFC 100, which is a summary of RFCs 1-99, and
> then RFC 1000, which is summary of RFCs 1-999. The had almost similar
> classifying system. A part of these RFCs (1-999) contain category in
> accordance with RFC1000. Why not all? - at the time they were published,
> there were no exact requirements to RFCs.

Exactly.

>> Let's not rewrite history.

In specific, if such labeling was optional, then it was.

I.e., if you want to publish an RFC that categorizes all RFCs before 
1310, or all RFCs, or all prime-number-RFCs, fine.

But that doesn't create an entry in the index with that label. THAT is 
rewriting history.

Joe



More information about the rfc-interest mailing list