Re: reduce the number of WGs [was Re: The Friday Experiment]

[Jared Mauch <[e-mail]>]

> On Nov 11, 2018, at 2:29 PM, Brian E Carpenter <[e-mail]> wrote:
> 
> On 2018-11-12 07:43, Michael Richardson wrote:
>> 
>> Kathleen Moriarty <[e-mail]> wrote:
>>> I was disappointed by the number of overlapping sessions for security
>>> and having to choose between working group sessions that I would have
>>> liked to attend.  This did impact some working groups that suffered
>>> from light attendance that included having less regular participants.
>> 
>> I had a similar number of conflicts this past week as previous IETFs.
>> Typically this is at least four significant conflicts, and 4 or 5
>> conflicts where I'd like to know what is going on some new group
>> (i.e. wugh), but I can't because I have to be elsewhere.
>> 
>> We need to reduce the number of WGs as well.
> 
> One of the IETF natural constants that I have never properly understood
> is that the number of WGs ~= 120 over many years. But just to quote
> from my public input to NomCom (draft-carpenter-community-leaders):
> 
>   We expect the leaders not to work too hard.  The IESG in particular
>   works just as hard as it makes itself work.  More precisely, today's
>   IESG defines the work load for its successors, by approving WG
>   charters.  If fewer WGs are approved or renewed today, there will be
>   fewer drafts to process in two years' time.  We expect the IESG to
>   say "no" quite often.  In the case of BOFs and workshops, we also
>   expect the IAB to recommend "no" quite often.  Of course, the "no"
>   should be clearly explained, and rooted in community consensus and
>   technical evaluations
> 
> If we as a community don't get better at saying "no" this problem
> will never go away.

I can think of a few WGs that are well intentioned but should likely
just go away or be merged.  Some examples, tictoc should be merged with NTP
or just become the time-wg or something similar.  Captive portal is a
problem I care about as a traveller because the variety of interactions
with the hotels/airplanes/airports is all different and should be standard
but different ones have different reasons for working the way they do and
I do not expect them to standardize this to be honest.  There’s no value
for them to interact with the IETF.

From attending homenet this week there may be some things of value to come
out of it, but as most people don’t have significant options nor do they
_want_ to have this complexity in their home network (they just want to consume
the internet, not orchestrate it all) it seems like perhaps a problem of
consuming space because you can, not because it’s a good idea.

There’s also some significant issues with the overlap and lack of interest
between areas such as from network operator things like GROW and transport
area changes.  Due to application and transport layer problems many operators
have deployed mitigations to keep their networks functioning.  The transport area
folks are very dismissive of the reasons the operators have done this in some
of the side conversations I’ve had and I fear the gap within the IETF
and fortress walls will be built taller vs talking amongst ourselves better.

We don’t have infinite time to review each others work and I suspect that most
folks are interested in this but honestly don’t have the time/resources to read
all the work that comes up for IETF LC.  I certainly don’t, nor do I expect most
other people will either.

- Jared