Exploration Through Example

Some questions on the submission process for Agile 2008, passed along by an Example stage committee member.

What the heck is a “stage” anyway?

I am thinking of a stage as a “conference within a conference.” It’s like a stage at a large music festival, where there’s a particular physical location where you might go to hear blues. (Plug here for Karen Carroll.) Buy a pass to the music festival and you can wander among the stages, seeing what you like. Or you might decide to hang out at a particular stage with people who share your particular obsession.

The “stage producers” have a lot of leeway in how they organize their own particular stage. That’s why the answers below should be considered specific to our stage.

How will the submissions be reviewed? What constitutes a good submission?

The first stage reviews will be public, on the submission website. The purpose of the website is to help the submitter hone the submission into its best possible form.

After submissions are closed, I’m planning to have something similar to what’s done at many software conferences:

1. Everyone votes on the submissions she’s read, using a four point scale:
   A: I will champion this submission (argue vigorously for its inclusion).
   B: I support this submission, but it wouldn’t break my heart were it not included.
   C. I think this submission should be rejected, but I could live with it being accepted.
   D. You have got to be kidding! Accept this?

   Most submissions will have a near consensus. The interesting cases are those where you get one A and four C’s (or A’s and D’s).

2. It’s possible that the number of clear accepts fits exactly within the space boundaries we have. More likely, we’ll need to make some choices. These get made in a phone conference. The structure of the conference depends on the submissions and how they match up with our available space. Here’s one possible scenario:

   We have five open slots. There are also five interesting disagreements. We discuss them. In two cases, the negative votes agree to accept. In one, we decide that we’ll provisionally accept if the proposer adds some particular improvement.

   Now we need three other cases (two definite accepts, one if the provisional accept falls through). People propose sessions from among the pool of ones that no one’s either strongly for or strongly against. Which sessions do they propose? Probably it depends on how they stack up against the already-accepted sessions: does a not-yet-accepted session complement or nicely contrast with one we’ve already got?

What keeps this from being a popularity contest, where only the most well-known names get accepted?

All of you reviewers do. The visibility of the submission process should help. If some Big Name writes a sketchy submission that displays no effort (which happens more than I’d like) and we accept it, the shame of it will follow us the rest of our lives.

There is an argument for accepting Big Names: it draws attendees to the conference, some attendees (especially first-timers) are more likely to listen to them, and it gets people who most likely are Big Names for a reason to the conference, where they can wander the hallways dispensing wisdom. Since this stage is for an oddball topic where there are no big names, that argument does not apply to us.

How will the sessions be compensated? Do they all get the same
compensation?

This is supposed to be on the website (there’s text saying so), but I couldn’t find it. Bug has been reported. Here’s what was once the correct information:

2. Here's the compensation for speakers, it's still based on session type.
  * Tutorial/Discovery session: 90 min: USD 250, up to 2 free
registrations, 5 night hotel, 1 room ; 180 min: USD 500, up to 2 free
registrations, 5 night hotel, 1 room
  * Experience Report/Research Paper: 1 free registration
  * Panel: moderator: free registration, 5 night hotel; panelists: free
    registration
  * Main Stage presenters (formerly known as talking heads): USD 700, free
    registration, 5 night hotel, 1 room
  * Free registrations and free hotel nights are non-transferable and
non-accumulatable.

This does not cover “other.” Compensation for other will be handled case-by-case. Here’s the way I think about it:

1. Things that are normal for conferences (presentations) get one free registration.

2. Things that require substantial extra work or require special expertise in crowd management or are popular enough that they will help sell the conference get free registrations and some amount of hotel rooms and money.

3. Panels get compensation way in excess of the effort required. But they’re popular among attendees, so they’re worth paying for.

How complete should a proposal be when I first submit it? What’s
considered to be “complete”?

It should be complete enough to entice a modestly public-spirited committee member or interested reader to take the time to write helpful suggestions.

“Complete for final consideration” means that every idea you (the submitter) have for improving the session or improving your argument for the session seems like more trouble than it’s worth.

That sounds like I’m being flippant, but I’m not. I don’t think completeness is something you can know a priori. You construct what completeness means by moving toward completion.

A more helpful answer might be this: does the submission describe enough that a reader can imagine what it would be like to attend that session? what they can (best case) leave with?

What’s the definition of the various “session types” - Demonstration, Experience report, Panel, Talk, Tutorial, Workshop? Does my submission have to be one of these?

Here are my understandings:

• A demonstration is where you show an audience something. The thing is traditionally a tool in use. (”Here’s how Watir works.”) It can also be a demonstration of a technique. (”Watch as I use session-based testing on this here app.”)

• An experience report is the result of some personal experience over an extended period of time (most typically a complete project). The goal is that the audience understands what happened and what the presenter learned, and that they can map that experience onto their own.

• Panel is a moderator plus N people, all sitting on the stage at once. What happens then is up to the moderator. Typically the moderator explains the topic, then each of the N people says something about it. Then the moderator may ask provoking questions, and the panelists talk about them. Typically questions are taken from the audience, and the panelists try to say something interesting and profound.

• A talk is a presentation like a demonstration or experience report, but it doesn’t fall into one of those categories. Like them, audience participation is usually limited to questions after the talk.

• A tutorial is a longer session where the presenter is someone specially qualified to teach a particular topic (preferably both an expert in the topic and with considerable experience teaching). Tutorials usually have some “experiential” component where people practice what has just been explained. Participants come to a tutorial to learn, in some detail, what someone else already knows.

• A workshop (aka Discovery Session) is organized by someone who thinks not enough is known about a particular topic and wants to do something about it. So, in contrast to a tutorial, there’s not the metaphor of knowledge flowing out of a teacher’s mind into the minds of a pile of students. There is a group of more-or-less peers creating new knowledge. Sometimes this is through some form of structured discussion. But there’s often an experiential component: a lot of learning by doing.

I am personally most interested in workshops and experience reports.

[The] consensus [of people she was speaking to] was that this year’s process heavily favors previous presenters, more so than in other years. They felt the lack of definition of sessions, lack of criteria by which sessions will be reviewed, and wide reviewership makes this year’s game nothing but a popularity contest; new folks need not apply.

Now that I think about it, what I see is a bias toward regular conference goers more than regular presenters. I think the (implicit) persona is someone who’s been to conferences and has had one of these reactions:

1. Why did no one talk about X? I guess I’ll have to do it.

2. I know as much about Y as anyone who presented. I guess there’s an audience for what I know.

3. I’ve got this fuzzy idea about Z. As far as I can tell, everyone’s fuzzy about Z. But at least there will be thoughtfully fuzzy people in Toronto next year - I need some mechanism to get them together so that we can figure this thing out better.

Whether a particular stage turns into a popularity contest depends on two things:

1. Do people just vote on proposals or do they make helpful comments?

2. Does the selection committee pay more than passing attention to the votes, or do they spend time reading and thinking about the comment threads to reach an informed opinion?

I don’t think our stage will.