Re: [Iasa20] draft-haberman-iasa20dt-recs-00.txt

[Stephen Farrell <[e-mail]>]

Hiya,

On 11/07/17 19:56, Jari Arkko wrote:
> Stephen,
> 
>> I guess I understand how the DT got there, but I'm a bit
>> disappointed that so far, we only seem to have addressed
>> the "should the IETF incorporate" aspect of the iasa2
>> work.
> 
> That’s taken a fair amount of discussion, but in the
> document Sections 5.2 and 5.3 are not about that.

Note that my comment wasn't meant as a criticism of
the DT - I really can see how that would end up being
talked about first and take time and take up most
of the -00 version text. (Disappointed maybe wasn't
the right word, sorry.)

> 
> I guess my question to you (and more generally all
> of you) is whether we’ve missed options or aspects
> that we should have explored. The idea with the
> meeting next week is to talk more about potentially
> missing options or what we should put into to the analysis
> of those options, than what the “Right Answer” is
> for the IETF yet.

For me, the main things that need improving with IASA
nearly all relate to transparency and I think that got
maybe one mention in the draft. (Didn't check back,
sorry if I've gotten that wrong.) I would have hoped
that a default-open statement would have been included
and to see if that'd differ based on the different
organisational options.

So I'd be interested in the analysis covering which
of the options might be most transparent. I worry
there's a big risk that transparency++ might get
lost if we undertake the two more radical changes
to the organisation.

> 
> So, do you have thoughts on what aspects we
> are missing and which should be documented?
> Or are you unhappy with the level of detail or
> analysis of the current options?

Not sure it's that useful to note but there are
things with which folks seem generally very happy
(e.g. AMS staff today) and I don't think the draft
covers not-breaking-those. So what other things
do we not want to break? Lack of membership is
one I'd like to keep, there may be more. So maybe
consider documenting things we like too?

> 
>> - I didn't find section 6 clear or convincing but
>>  it's also hard to see how one might document
>>  such an analysis at this point. Maybe when I
>>  read it more closely or we see a presentation in
>>  Prague, it'll be clearer, but as of now, the DT's
>>  analysis isn't that clear to me. It could be that
>>  I change my opinion about IASA++ if/when I do
>>  understand the DT's analysis better, but I'd be
>>  surprised if that were the case.
> 
> At this point, we’d love to hear your thoughts
> on aspects that should be taken into account.
> Lets worry about what the right conclusion is
> later.

Fair enough. But TBH, I think I still need to see
the presentation in Prague to get what's being
said in the analysis. I'll try give it another
read in the meantime if I can and respond further
but not sure I'll have a chance.

Cheers,
S.

> 
> Jari
>