Re: to pitch or not to pitch, IETF attendance costs

[Keith Moore <moore@xxxxxxxxxxxxxxxxxxxx>]

On 5/15/19 6:53 PM, Joel M. Halpern wrote:

First, new work comes in at different phases, not just in the BoF.

Yeah, I think we were talking about two ways of introducing new work - virtual interim meetings and virtual BoFs.  I'd expect them to be similar to one another, but not entirely identical.

Second, when new work comes in, it is not obvious to folks that they are interested in it.  Thus, they are less likely to make time to attend virtual interims (or, even worse, actual interims)

On the other hand, it's possible we're excluding a large number of potential contributors by insisting on doing so much of our work, and all of the BoFs, at in-person meetings.   

So I suspect the trick is to improve "interim" (i.e. entirely online) meetings, and our advertising of them, so that they attract a larger number of potential contributors (and also a more diverse group of contributors) than our face-to-face meetings and BoFs do.

(This is just a hunch of mine.   I'm not certain it's better, but I think it's worth exploring.)

Third, due to the nature of time zones, it is VERY hard to actually find a time when all the folks can make it.  Particularly when you do not know who the interested parties are, as they have not yet had a chance to see the work and discover they are (or are not) interested.

This is indeed a problem and a risk.   I'd hate to see the times of virtual interim meetings chosen in such a way as to deliberately or effectively favor some participants over others.   Then again, I still think it's probably easier for folks to be up late at night for a virtual meeting, then to travel halfway across the world from wherever they are and be up late at night (relative to their home timezone) for a physical meeting.   (Though there's something useful about getting away from one's normal job, for those who can.   It's just so expensive and time-consuming that there are many who cannot do that.)

Yes, this means that sometimes we sometimes use WG face-to-face time to let folsk with new ideas know that their new idea has lots of problems or is not of interest.  That is not, in my book, a waste of time.

Let me take a step back.

In my experience, nearly all face to face meetings that I've experienced in my career, were not only a very inefficient use of time, but also a huge disruption (not a good one) to creative people who needed focus to work.

I've concluded that face-to-face meetings are very useful for some things, but the only way to make them an efficient use of time is for the meeting to be tightly focused on things that are necessarily, or much more efficiently, conducted in face-to-face meetings.  

That means, for example:

• You shouldn't use face-to-face meeting time to explain things to people if those things can be communicated over email or a web site.   If there are questions about those things (from people who have read the email or site) it might be a good use of face-to-face meeting time to address those questions - if this can be done briefly.  (If people who have read the material have questions, it's likely that multiple readers have similar questions, and the answers may be illuminating even to those who hadn't thought of the questions.)
  
• You shouldn't use face-to-face meeting time to have people report their status.   Instead everyone should send a brief status email or update a status board N hours prior to the meeting, and everyone who attends the meeting should be expected to read those things prior to the meeting.   Then the only status discussions can be questions for clarification (from people who need to know specifics to do their work), and/or negotiations (e.g. "can you finish X so I can finish Y?")
• You absolutely should not use face-to-face meeting times to present documents in detail.   People should read those documents in advance of the meeting (or at least summaries of those documents) and come prepared with any questions they might have.   (Sometimes you don't actually need to read a whole document to understand whether the technical choices being made are good ones, but you do need to know what those important choices are.)
  

Potentially good uses of face-to-face IETF WG meetings include:

• Interactive group discussion of technical issues, taking advantage of the high bandwidth and better fidelity of communication (things like facial _expression_, tone of voice, and body language) that physical presence permits, which may facilitate a much more rapid convergence than would be likely than over email.
  
• Flushing out issues that might not have been fully articulated, but which may have made several people uncomfortable.  For instance, one person can speak up in a meeting and others who share a similar concern but might not have expressed it, will feel freer to speak up.   (For that matter, if there are significant differences of opinion on the topic, those may quickly become apparent, and this is also useful to know.)   This can be done on a mailing list also, but the potential to get immediate feedback from several participants at the same time (and for those participants to get a sense of the room that others share their concerns) can be very valuable at bringing those issues to the forefront of discussion.
  
• Presentations of topics that are not easily communicated effectively using text documents and simple drawings, and are of importance to the working group.  (these should be rare but do exist)
• Interaction of almost any kind that serves the WG's purpose, because people who have actually talked to one another in person have an easier time understanding one another when they communicate over email.   (emphasis on Interaction)
  

I realize that face-to-face meetings are conventional, that there's a lot of investment in them, there's a lot of mindshare behind them.   For some, they're a good excuse to get out of the office, out of town, travel to another country, maybe even experience better weather.   They also motivate contributors to get drafts updated.  I don't wish to dispense with them entirely or even mostly.   But if people are going to shell out thousands of dollars to go to a typical IETF meeting, they should get a better return for their investment than they're getting now.   And if we could make them even slightly cheaper, say by reducing the number of days invested by 1 or 2, maybe allowing attendees to take advantage of reduced air fares on some days.... well, every little bit helps.

Keith