[ih] Cluster Addressing and CIDR

[Joe Touch]

John Day wrote:
> At 15:11 -0800 1/14/03, Joe Touch wrote:
> 
>> John Day wrote:
>>
>>>>
>>>> Authors will cease to present partially-complete ideas. There will 
>>>> be fewer work-in-progress drafts. There will, in summary, be less of 
>>>> this 'good or bad' research to preserve.
>>>
>>> This is not the case and has not been the case with other groups.
>>
>> There are some people (myself included) who will cease to publish 
>> drafts. That decreases (by definition) the set of what is published; 
>> to the extent that others care, it will further decrease that set.
> 
> Somehow I doubt that ;-)  but statistically I think you would find that 
> the number who would not submit contributions would be insignificant.  

OK - so we disagree on what might happen.

>>> I have not noticed that the fact this material is available in the 
>>> group's paper trail has any effect on the amount or quality of the 
>>> contributions.
>>
>> That's nearly impossible to measure. We have no series that was 
>> explicitly not archived then archived to compare. All we have are 
>> different communities right now.
> 
> Actually, not true.  If you look at IEEE, T1, ISO, ITU, ANSI, ABA, IEC, 
> and many other groups you will find that such paper trails exist and 
> perhaps not readily available they are available.

That is not a metric. You are not citing groups that changed policies; 
they were archival from the start. Agreed that they have open exchange, 
but by design.

>> ...
>>
>>>> The thing that makes IDs unique is _exactly_ the fact that they are 
>>>> NOT archived. Were that property to disappear, there would be a void.
>>>
>>> No the problem we have is there is a void.  To modify an old adage, 
>>> those who can not know history are doomed to repeat it.  Actually now 
>>> that you mention it that may explain alot.
>>
>>
>> There always was, and continues to be a path for publication that some 
>> draft authors have chosen and others have not. Draft authors can 
>> always submit documents for Informational RFC; some have, others have 
>> not. Although there are some submissions which have been rejected 
>> (every system has its minimum standards), overall we already have a 
>> solution to this problem, and it doesn't involve archiving all drafts 
>> for historical purposes.
>>
>> I agree that the world is less informed by not having the intermediate 
>> forms of "the Shining", e.g. That is as it has been - the choice of 
>> the author. All we do by archiving drafts is to take the ephemeral 
>> track away.
> 
> This is not at all the case.  Science is a much different process than 
> writing a novel.  The exploration of the domain of inquiry and the 
> process by which it takes place is as important as the final answer.  
> More often than not the process may provide more understanding for the 
> next problem than the answer ever will.

That may be the case, but it is the authors' decision, absent a-priori 
arrangements of a forum (e.g., this forum is archived, and that is known 
by participants when they sign-up).

IDs are a place where exploration comes without the threat of persistent 
archives of proposals, _by design_. As mentioned before, there are 
plenty of other archival forums, including RFCs themselves.

Joe