Home | Login | Recent Changes | Search | All Pages | Help

SessionAttendance

One of the wonderful things about the AYE conference is that we have flexibility in the schedule and in who chooses which sessions. We don't make you sign up in advance and give you a ticket (which would be standard for these kinds of tutorial-like sessions). On the other hand, that makes it difficult for us to plan which sessions to locate in which rooms, or if we should re-offer a session later in the week.

We decided that we can repeat or relocate one session to Wednesday afternoon, if enough attendees have interest and the leaders are willing. (This is a change for this year. What do you think?)

We would also like to know how many people are planning to attend each session. However, I don't want you to tell me on this page. Instead, I'd like to know: If you could design a way to determine the number of people who would attend each session, what would that design look like? --JohannaRothman 2002.09.24


Wiki-based solution:
I created a page for each time slot (ie: MondayAmSessionSignup)
  • For each concurrent session:
    • Horizontal rule
    • Session Title
    • Preparers
    • Please enter your WikiWord name below if planning to attend this session
    • Horizontal ending rule

Let the participants manage themselves.

--BobLee 2002.09.25


Bob, I'm not sure I would have come up with a system that looks like the one you put in place. Is this the one that the conference is going with, or is it still open for discussion?

--DaveLiebreich 2002.09.25

Open for discussion --JohannaRothman

Perhaps Johanna actually meant "I don't want you to tell me right now". But she said "on this page" instead. So Bob told her on a number of other pages... :) -- LaurentBossavit (sameday)

Sorry, Dave, but it was easier to do than to describe, and I had 1.5 hours on my hands due to a no-show kid today. Let's improve the art by designing even better solutions. This was pretty brute-force cut'n'paste, but it spreads the Wiki contention over 9 separate pages. I also thought of an Excel spreadsheet, but then it's not participant maintainable --BobLee (sameday)

Bob -- One problem is that a small cross-section of AYE participants visit the Wiki (from Server logs, I'm guestimating 20-25% at most). The best we can do here is gather data that Johanna can project estimates from. --DaveSmith


My oh my. Bob, this is one way. Your design is certainly one technique, and works for people who wiki, and who are unlikely to change their minds at the conference. Is there another alternative you or anyone else can see?

Here are the alternatives I thought of:

  • Do nothing. As in previous years, let the facilitators run with things. If they don't have enough people, don't sweat it.

  • Prepare signup sheets for the conference, and tell people about the signup sheets when they register and at the Sunday night dinner.

I knew I was stuck when I could only think of 2 alternatives. You've given us a third alternative, which means there are several more out there. Any other ideas?

Laurent was closer when he tried to interpret my ambiguous statement. What I wanted was designs, not for people to tell me which sessions they wanted. (Is that clearer?) --JohannaRothman 2002.09.24


I was tempted to answer Johanna's original question (what would the design look like) with "A Crystal Ball". Coming up with an actual design, now... that's harder.

How accurate does the proposed solution have to be ? How far in advance do you need it to yield a number ?

A fairly accurate but late solution (as opposed to Bob's early but perhaps less accurate) would be to count the number of people milling in the anterooms of each session, about five minutes before the sessions start. (There's a third alternative.)

How do you react to the numbers ? What do you look for - sessions that are overfull ? Sessions that are (close to) empty ? Uneven numbers ?

I quite like the notion of not having to sign up in advance. I don't have to commit to a schedule according to my current learning needs, which could turn out to be suboptimal given what my learning needs became by november.

What if we had presenters sign up for participants ? There's a solution that doesn't have the "participants who don't Wiki" issue - it's a lot easier to get the smaller number of people to agree on one method for signing up. ;)

LaurentBossavit 2002.09.26 (26 minutes into the day)


Prepare signup sheets for the conference, and tell people about the signup sheets when they register and at the Sunday night dinner and at the beginning of each session (on the first day or two).

Put the sign up sheets on the wall or on stands next to the rooms that the sessions will be in, with a pencil attached to it by a string... or put all of the sign up sheets in a central location, perhaps arranged in a nice time / room-grid.

KeithRay 2002.09.25


Laurent,

I like the presenter signs up for participants!

I am not a sign up sheet person myself. But from a conference host POV it can help us figure out what space (small, medium, large) to allocate for what sessions, but I don't think anyone wants to freeze your schedule into a bronze plaque.

Reality is that regardless of the best plans, something different happens. :) Johanna accounted for that in her suggestion to "do nothing" and go with the flow.

- BeckyWinant 9-25-02

I think you should do nothing & go with the flow. When asked to in the past, I have sent in my name by email, written it on sheets of paper on the desk or the wall and maybe half the time I change my mind, often in the last 5 minutes.

Johanna, I think this is herding cats with a vengance. No matter what you do, we will all end up where we belong, we will all have a truly wonderful time and we will all learn enormously. Isn't that why you and I are back again? (OK, that and the hugs!)

SherryHeinze 2002.9.25 (3 more hours in my day Laurent)


I wrote up a long (and dare I say, eloquent and insightful) response for this page, then fumble-fingered it into the bit bucket. Days worth of thought, gone in an instant. Sigh.

The following is but a hint of the glory of the original - I hope it will suffice. Or something like that.

Johanna, what do you want to know, and what would you be willing to do/change as a result of that knowledge? Would you move many sessions around if you discovered that a large number of people wanted to attend conflicting sessions?

What are your goals? Do you want to make as many attendees happy as possible? As few unhappy as possible? Hosts instead of attendees?

With regards to the presenters signing up attendees - as a non-presenter, I want to let all presenters know that I am an excellent attendee. References provided upon request. The bidding for my presence can take place on DaveIsHere.

--DaveLiebreich 2002.09.25


Dave,

Bravo! I loved the performance and insight.

BeckyWinant 9-25-02


Hosts: ...a small cross-section of AYE participants visit the Wiki (from Server logs, I'm guestimating 20-25% at most).
  • Interesting Dave, historically, how many AYE attendees came but never signed into the guest book? (2000, 2001?) Those metrics are pretty invisible to a first timer like me.
  • JR - how variant are the room sizes? With a max of 4 concurrent sessions, does it become a squeeze and a vast horizon with wrong guesses? I would have assumed that the "under 100-150" would allow relatively easy accomodation even with skewed attendance. I also expect some periods people may choose to visit 1:1 with hosts. After 2 years, isn't there enough track record to estimate with?
  • Is this just a "J" preference that session size things should be settled?
  • Re: repeating 1 session on Wednesday PM - that's one day I'm already schizo on - I wanted to attend both Weds. PM sessions already!
  • I don't regard signing up for a session as a commitment, but as a rough intent. I may well discover different priorities at ground zero, may wander through several concurrent sessions, or cut out to meet side groups.
  • I think advertising my sessions of interest may help others choose, and seeing others' choices may show interesting patterns, too.

--BobLee 2002.09.26


You guys are all way too smart for me. Here is the situation I would like to avoid: We had to cancel sessions last year, due to not enough people arriving in the first 5 minutes. Later, over lunch or drinks, people said, "Oh, I really wanted to attend that one, but this other session won out at that time. Can't we repeat that session that was canceled?"

Dave, you asked what I would like to have happen:

  • I would like to maximize the contentedness of both the hosts and guests. I'm not looking for happiness, but the feeling or realization "That was a good conference. I learned something and didn't waste my time."
  • I would like to avoid "That was a waste of a morning/afternoon" perception on the part of either a host or a guest.
  • Given the constraints that each host can only be in one place at one time, and that we have only 4 concurrent session rooms, we have some flexibility in the schedule but not complete flexibility. I'm willing to move things around, if I thought that would increase contentedness.

I don't mind sessions with low numbers, unless those session require a minimum number of people greater than 1 person. I think I'm only doing one session that requires 3 people. All other sessions require only one person. (Of course, the people who attend will get more out of the session if more people attend, but that's true for all the sessions.)

I like the idea of presenters signing up for participants. And, I'm concerned that I'll miss out on some people I haven't met yet. I was dismayed last year at the closing when I realized I hadn't met at least 5 people all week!

Sherry, I've long held held the belief that we participate in the thing we need to at the time we need it. It's possible I'm trying to solve an unsolvable problem, or a problem that doesn't need to be solved, but what the heck. If I don't try, I can't possible succeed.

BTW, if I was trying to herd cats, you'd know. I'm not even guiding here, just exploring :-)) --JohannaRothman 2002.09.26 (am EST)


I know you aren't really trying to herd even cats, Johanna, and you know AYE attendees way too well to try to herd us. It really was just an unsolveable problem reference:-)) --SherryHeinze 2002.9.26
Another suggestion. If there is a mailing list for those who have registered, send out email and ask that people either:
  • Email their top x workshop preferences
  • Email a list of all their preferred (at this time) workshops
  • Come to the wiki and select their top x preferences...
  • Come to the wiki and select all of their preferred (at this time) workshops

and

  • Identify those sessions for which their is a time conflict, so perhaps they can be rescheduled (that would be a real motivator for me ... even without promised sucess!)

THEN, follow up with an "It looks like x & y sessions will not go; z and q may be overcrowded; a and b are better if they change places; etc." and get feedback again.

A lot of work? maybe........ but just a thought.

DianeGibson (who found quite a few troublesome conflicts between sessions she wanted to attend last year... of course, I don't know if there was enough TIME for all, but....) 9-27-32

  • PS I did go and sign up for sessions, indicating where I was undecided (in a new section called "CONFLICTS". Even limiting myself to the MOST SERIOUS conflicts, there was one every day..... Could get messy!

Hi Johanna.

My literal response is simply this: Pick any solution to reduce the odds of cancelling sessions due to attendance and leave the rest alone.

  • It seems to me that people not getting to every session they might want is a sign of an abundance of riches. Not meeting people ditto. Why is this bad?
  • By default, I suspect people will go away thinking: "That was a good conference. I learned something and didn't waste my time." One way to mess that up would be to try to over optimize the scheduling and attendance so much that the optimization task gets in the way of conference content and / or becomes miserable itself.
  • Has somebody mentioned us raging P's? I'll decide where I want to be when I get there, if then. How are you going to optimize for that?

You can do some good, I think, by applying one of the tractable solutions to the unattended sessions problem, and leave the rest of the aspirations alone. If anybody particularly wants me at their session, drop me a note, and we'll see.

The rest is WiseGuyMusing, abstracted out to an annex where noone need look at it.

- JimBullock, 2002.9.27


Diane -- my suggestion is to add your name in each sessions you're interested in, with a question mark indicating which one is less likely... or a parenthetical request to repeat a session. My two cents.

-- KeithRay 2002.09.27


I like the way the signup is working, except that the sample is too small so far. Can we have a couple of web connections at the registration table, so people can make choices when they register? (if they haven't done so already)
This would allow us to get a pretty good idea at the conference, at least - Sunday night probably - so we could make adjustments rather easily. - JerryWeinberg 2002.09.29

That's a great idea. Defer the decision until the data stabilizes. For myself, I'll consider signing up for various sessions as soon as the list I think I want to attend is the same any two times in a row. As it is, the list changes 50% at least, every time I look at it, and more like 200% each week.

- JimBullock, 2002.10.1


Jerry, Johanna: If you really want registration software, get your request in now. The collision-avoidance feature of this Wiki will frustate people if a lot of people try to change the same page(s) at the same time. I could put together something that would work better for registration. I suspect, though, that a lower tech, higher visibility solution (e.g., post-it notes) would be preferable.

DaveSmith 1 Oct 2002


thanks for the offer, Dave, but I'm guessing that won't be necessary. If we have only one station, that should be sufficient, at the rate people register. One station would pretty well prevent collisions. Two wouldn't be a problem, either, I think. - JerryWeinberg 2002.10.01
Another slant on this "Gee, I wish that those sessions weren't concurrent!" is: it will raise attendence next year - to catch what they missed this year. That's OK, too! Does it really require fixing? -- BobLee 2002.10.02

I just noticed the conflicts entries in various of the session signups. I'm conflicted about every potential choice. I want to attend all the sessions. Should I list these conflicts on the sign-up pages? In aid of a consistent data set, of course. Are we going to measure conflicts as an indicator of overall session quality, or maybe calibration of the offerings to the attending comunity? If so, I'll give you all the "goodness" indicators you can stand. (While I'm on that topic, I've never understood people who go to conferences then spend their time not attending the sessions, or complaining about the sessions, or both. If you're not interested, why show up? I so don't get that.)

- JimBullock (I'm just trying to be helpful here. Really. No, honest.) 2002.10.2


Jim, I have been to AYE every year, and I always want to go to every session. Usually there are only 2 in a time slot that I want to go to desparately, but not always. I can multi track, but I can't be in 2 places at once.

SherryHeinze 2002.10.3


"Two places at once" - one way to solve this is to only have one place, as in an OpenSpace . This makes sense to me - the world is all one place anyway. Breaking it up into arbitrary pieces and calling these pieces "sessions" is one way of managing the complexity of the whole.

What are the problems being solved by partitioning the conference into "sessions" ? Is that setup actually solving the problems it's meant to solve ? What other problems does it create (I suppose this is the question we've been discussing on this page) ? Could these problems be solved differently ?

LaurentBossavit 2002.10.04


Jim,

Here's one person's perspective on why conference sessions are avoided. At a typical three day conference where sessions are 45 minute powerpoint presentations, I can only sit through so many. I pick the people who I enjoy listening to, people I want to hear because I've read something they wrote, and the topics I'm interested in. If a slot has none, I do something else. Too many of these is a conference I don't want to attend again. By day three, I might not want to sit through 45 minutes of anyone talking at me.

AYE is unique. The first year we (hosts) were taking hall duty to chat with anyone who wnted to sit out a session. We were lonely hall monitors. Next year we stopped doing that and started attending sessions ourselves. Many of us have the same conflict - which one to go to?!

- BeckyWinant 2002-10-04


I've only just come across this page, so two additional cents. I don't think I could favor any solution that requires sign-ups in advance, whether "advance" is via the wiki or at registration or 5 minutes before the session. I much prefer the open-ended feeling of everyone going to whatever session they want to, based on whatever their choice is at session start time. Opportunities, not constraints.

When we attempted session sign-ups at registration at AYE 2000, people still went to whatever sessions they wanted to, independent of what they'd signed up for. Every session has to be ready to run, whether or not there's advance info that people are planning to attend. As presenters, we need to be able to accept the risk that people may not be interested in what we're offering.

For myself, I hope people will want to attend my sessions, but if no one shows up, I get to go to other sessions and learn from them, so I win either way. NaomiKarten 2002.10.11


Just to be clear, Naomi, my proposal wasn't for signing up, but for polling to get some idea in advance of how much interest there is in each session. We have different sized rooms, so this will help us choose rooms, and maybe give us time to do some rearranging if we find there's conflicts that could have been avoided. I also like to have some idea for my exercises if I'm going to have 5 or 50 people. I can work without it and just respond to what's in the room, but I think it helps me do a somewhat better job, in some cases, if I have an inkling in advance. And, like you, I don't mind if a session cancels so I can attend someone else's session. - JerryWeinberg 2002.10.11
Jerry, actually my comment above wasn't in response to your proposal, but rather just some thoughts that came to mind as I skimmed all the comments on this page. But thanks for the clarification anyway. I agree with you that polling to get a general idea of turnout is useful, inkling-wise, especially in planning simulations and such. NaomiKarten 2002.10.14


Updated: Monday, October 14, 2002