[rfc-i] 'gaps' in the RFC index
ah at TR-Sys.de
Fri Oct 2 10:20:46 PDT 2009
Dear RFC Editor(s), Sandy,
thanks for your quick and detailed response!
> Thank you for your email. Please see comments inline.
I see that your remarks below broadly match my expectations,
as reflected in the 'Discussion' part of my original message.
I'll be happy to see progress in this detail, but I well
recognize that this is not one of the highest priority tasks.
> On Fri, Oct 02, 2009 at 11:19:07AM +0200, Alfred HÎnes wrote:
>> The RFC index used to document which RFC numbers have never been
>> issued (and will not be issued any more). To this end, the
>> "rfc-index.txt" file, as indicated in its preface, contains
>> entries of the form:
>> #### Not Issued.
>> and these entries appear in the XML version of the RFC index as:
>> Unfortunately, the useful practice to insert such entries has
>> been ceased gradually during the last years. The last such entry
>> added was for RFC 4637, but there are many gaps left open earlier
>> as well.
>> It would be appreciated very much for clarity and visibility
>> if decisions to not publish RFCs with specific numbers were made
>> visible in the RFC metadata, in this established tradition.
> We try to go through the numbers periodically to update the "never
> issued" numbers, but there are many factors that do not allow us to
> update the numbers as quickly one might wish.
>> The upcoming transition of the RFC Editor function might be a
>> good opportunity to perform such cleanup.
>> There are three classes of RFC numbers missing from the RFC index
>> (list current up to RFC number 5650, omitting documents in AUTH48):
>> a) RFC xx00 (STD 1) series:
>> 3800, 3900,
>> 4000, 4100, 4200, 4300, 4400, 4500, 4600, 4700, 4800, 4900,
>> 5100, 5200, 5300, 5400, 5500, 5600
> Some of these will be updated to "never issued" when we next update
> the rfc-index.
>> b) RFC xx99 (Summary) series:
>> 3399, 3699, 3799, 3899, 3999,
>> 4099, 4199, 4299, 4399, 4499, 4599, 4699, 4799, 4899, 4999,
>> 5099, 5199, 5299, 5399, 5499, 5599
> These are not obsolete. We intend to update these when possible.
> However, it cannot be done until some of the issues regarding the
> numbers mentioned below are clarified, and until the numbers above are
> marked as never issued.
>> c) sporadic numbers:
>> 3333, 3350, 3907, 3908,
>> 4232, 4658, 4751, 4921, 4922, 4989,
>> 5245, 5312, 5313, 5314, 5315, 5319, 5522, 5644
> All of the above numbers have not been designated as "never
> assigned." For various reasons, an RFC number may have been assigned,
> but hasn't made it to publication yet. Additionally, RFC numbers 5522
> and 5644 will be published; they are not "never issued." RFC numbers
> are not assigned strictly in sequential order.
This explanation closely matches my observations and the
expectations I had, as originally expressed -- see below.
I had to make a decision on where to make the 'cut' in giving
numbers; 5500 was one candidate. 5650 was an easy choice, simply
picking the most round number from the last batch of published RFCs.
> We agree that the index should be updated, and we intend to do so
> shortly. However, with the pending transition, we are currently
> focusing our efforts on preparing for the transition and moving
> through the current queue.
> Thank you for the reminder; this is something that was/is on our list
> to get done before transition as well.
> RFC Editor/sg
>> Re a)
>> RFC 5000 has been published, its predecessor was RFC 3700.
>> Remarkably, RFC numbers 3100, 3200, 3400, and 3500 indeed are
>> already listed as "Not Issued".
>> It is not expected that the other gaps in the RFC xx00 series will
>> be filled; a future STD 1 (if any) will likely be assigned a higher,
>> then 'current' RFC number.
>> So most likely the decision has been made to _not_ fill the gaps
>> mentioned above in group a) any more.
>> Hence, I suggest to formally 'close' these gaps by establishing
>> "Not Issued" entries.
>> Additionally, if the old RFC xx00 practice is confirmed as ceased
>> definitely, the xx00 numbers might become candidates for assignment
>> to 'normal' documents in the future.
>> Re b)
>> Apparently, RFC 3399 once has been delayed waiting for a decision
>> on RFC 3333 being published or not. I assume that RFC 3333 will
>> _not_ be published any more. So for the sake of continuity,
>> preferably the draft for RFC 3399 (reportedly, it exists)
>> should be finalized and published.
>> This would provide for a continuous style up to RFC 3599.
>> The decision to not publish more recent xx99 RFCs any more has
>> never been announced firmly.
>> I suggest that either this decision be made now and corresponding
>> "Not Issued" entries be filed for RFC xx99 with 'xx' > 35, or else
>> it be confirmed on the list that the decision is still left open.
>> Additionally, if the old RFC xx99 practice is confirmed as ceased
>> definitely, the xx99 numbers might become candidates for assignment
>> to 'normal' documents in the future.
>> Re c)
>> Many of the RFC numbers listed above correspond to 'late' DNP
>> decisions; sometimes successors have been produced before the
>> RFCs could be published (e.g. 3907/3908).
>> I conjecture that for the majority of RFC numbers listed above
>> (perhaps with exceptions for some 5??? numbers in the 3rd line),
>> the decision already has been made, or can be made now, whether
>> or not that number is still regarded for assignment to a document.
>> In the former case, that should be documented by filing the
>> corresponding "Not Issued" entry.
| TR-Sys Alfred Hoenes | Alfred Hoenes Dipl.-Math., Dipl.-Phys. |
| Gerlinger Strasse 12 | Phone: (+49)7156/9635-0, Fax: -18 |
| D-71254 Ditzingen | E-Mail: ah at TR-Sys.de |
More information about the rfc-interest