Re: [dnsext] WG Review: Recharter of DNS Extensions (dnsext)

[Jelte Jansen <[e-mail]>]

Andrew Sullivan wrote:
> On Wed, Jul 08, 2009 at 12:45:53PM +0200, Shane Kerr wrote:
> 
>> The current style of DNS standards work seems to be that we have a
>> number of people working full-time on DNS stuff, and occasionally they
>> think of something that should be standardized, so when they have a few
>> moments they write something down and present it to the dnsext working
>> group. This seems to work pretty good (except for the huge mass of
>> confusing RFCs, hopefully an effort to fix that will arise again some
>> day).
> 
> Well, the last effort was undertaken as a planned attack on the
> overall problem.  It failed because we didn't get text.  It could be,
> however, that a less centrally-planned effort would work.  If someone
> were to write a "DNS Guide to the Perplexed" as an I-D and just ask
> for comments, perhaps it would bear fruit. 
>  

Everytime this issue comes up I'm wondering where we got so deep into the RFC 
model that we are actually discussing writing a new RFC to counter the problem 
of having too many RFCs...

My general issue with that still stands; the DNS-in-a-nutshell RFC will be 
outdated as soon as dns-I'm-still-alive-ext publishes a new extension. It's what 
we do, and despite multiple attempts to kill the working group, no hero has 
succeeded yet. The update for Nutshell will be an RFC itself, effectively 
multiplying the DNS RFC growth rate.

Not that I'm against the effort; and overview of DNS in it's current state, with 
references to the myriad of RFCs and other known extensions/exceptions, would be 
fantastic. But IMHO it would be better implemented as either a website or a 
non-RFC publication. And it would only be useful if actually kept up-to-date.

>  
>> Maybe we should collect small DNS work in a separate area and then
>> create a "dnsfix" group every now and then to push them through the
>> standards process? It hardly seems ideal, but...
> 
> Creating a new group would be even more work than just rechartering
> DNSEXT to do the work, I note.  But perhaps your suggesting includes
> an implicit step, "Shut down DNSEXT."  Since the fondest hope of every
> working group chair is supposed to be the closing down of the working
> group on the grounds that there's no more work to do, I plan to be the
> last voice opposing that suggestion if indeed you are making it. 
> 

Speaking from recent personal experience, classifying dnsext-related work as 
"small" work can be harder than it seems. There's also the protocol-police part 
that's in both the current and the proposed charters to consider when thinking 
about shutting down btw.

I think it's worth it to try and get IFXR-only in the charter now (or consensus 
that the work is not worth it), which will save us at least one recharter. But 
it appears that frequent recharters are inevitable with an immortal working group.

Jelte

--
to unsubscribe send a message to [e-mail] with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/namedroppers/>