Re: [108attendees] Introducing the Meetecho Virtual Hum tool

[Pete Resnick <[e-mail]>]

On 27 Jul 2020, at 22:48, Keith Moore wrote:

> On 7/27/20 11:23 PM, Bret Jordan wrote:
>> I have yet to see a WG meeting that hasn't used the hum like a vote.

You have obviously never been to a meeting that I've chaired, Bret.

>> We just need to accept the fact that the hum is a pseudo
>> anonymous voting mechanism. Yes we want to believe we are not 
>> voting, but in reality we are. Sorry. Maybe you do not see it, 
>> because you want it to be something it is not.

I think you may have had a very narrow experience of humming in the 
IETF. It does happen that it is used as anonymous voting (and that's 
part of what motivated me to start on RFC 7282), but it's certainly not 
universal, and I doubt that it is a majority of the time. In the 1990s, 
when I started attending IETFs, it was certainly not used as an 
anonymous vote. Post-2000, it definitely got much worse. I do think hums 
are still over-used, and sometimes ill-used, and that is why we've had 
WG Chair lunch sessions on exactly this topic. But I think your 
evaluation of it always being a pseudo vote is way off the mark.

> I have yet to see a WG meeting that /has/ used the hum like a vote.

Well, that's surprising as well. I have seen hums used as more-or-less 
anonymous votes, where the louder hum was taken as the outcome, even 
with a small but significant number of people humming against. But I've 
seen lots of instances of hums being used well, and it seems to have 
improved since 7282 was published.

> Part of the idea of humming is that a simple majority is completely 
> insufficient to justify a decision, and there's no specific threshold 
> for doing so.   Another part of the idea of humming is that each 
> participant can hum at a different volume level according to his or 
> her level support for the question, so that strongly held opinions can 
> carry more weight than weakly held opinions.
>
> Both of these are useful properties for a group that makes decisions 
> by rough consensus.   For instance, humming isn't as vulnerable as 
> voting to tyranny of the majority.  Of course a hum isn't by itself a 
> determination of rough consensus, but it can be a useful way of 
> determining whether rough consensus is near or distant.

Totally agreed on the above.

Bret, my apologies for being immodest, but I really do think you should 
take Ted's advice and read RFC 7282. It might not represent the WGs 
you've attended (and I'd be interested to hear offline which WGs those 
were), but I do think that many IETF chairs and participants believe 
it's a good guide.

pr
-- 
Pete Resnick https://www.episteme.net/
All connections to the world are tenuous at best