On Wed, Feb 12, 2014 at 10:31 AM, Abdussalam Baryun <
abdussalambaryun(_at_)gmail(_dot_)com> wrote:
The draft needs to show that if a wg-chair is presenting (in meeting
or on lists) an individual draft to the WG for adoption, and no
answer/reply comes out, that does not mean against nor agree, I
recommend there is no value of no effort from the WG. Some WG may not
respond for adoption because they had no time for making a decision,
so I hope the Chairs don't misunderstand and make wrong decisions.
Calls for adoption are particularly interesting cases, because if there's
no response, then as chairman I have to assume there's no interest. This
could be because of lack of time, of course, but if merely checking the
draft over for adoption isn't worthy of effort, then I can't really expect
that once adopted, people will suddenly find the time to do serious work.
In other cases, however, silence can mean assent, or at least no interest
in arguing against. If a particular open issue in a draft is resolved by a
small group (or even a single person) picking a solution that works for
them, then an absence of complaints, despite not being quite the same, nor
as desirable, as enthusiastic and vocal support, still leads to the same
conclusion.
As for providing engineering reasons... I always prefer strongly to see an
solid technical reason for a position. It helps enormously in judging the
roughness. But if the reasoning is aesthetic, market-related, or a gut
feeling, that's nevertheless useful information. Since we deal in rough
consensus, not perfect agreement, even a real technical issue is not enough
to break rough consensus sometimes, but any reason might be - and in
general terms, I want to encourage open dissent.
This last principle is important for two reasons. The most obvious is that
we want to ensure that self-censorship doesn't cause people to avoid
speaking up when their concern is well-founded - especially true for less
experienced participants who may misjudge the value of their contribution
in either way. The other is that someone voicing just a nagging feeling
that the solution is "somehow wrong" might well jog other people into
noticing some real flaw.
On aesthetic and market-related reasons, two examples:
a) At one stage, the SASL working group were close to abandoning
human-readable names for new, GSSAPI-based, mechanisms. People argued
against this purely on the basis that it would have discouraged people from
implementing these newer mechanisms. This isn't a technical or engineering
argument, it's purely aesthetic, but it's nevertheless important.
b) With the on-going discussions about pervasive encryption, it's important
to recall that for many parts of the world, it's still against export
regulations to allow >56-bit encryption in binary software (ie,
non-open-source) to be exported to certain countries. There are strong
technical arguments against "export grade" encryption.
Dave.
(With apologies for the lack of flippant sarcasm in this message)