Back to: Beer in DC | Forward to: Panel discussions at SF conventions, part 2: keeping the panelists happy

On panel discussions at conventions

I'm traveling (in Washington DC, acquiring blisters as I tramp round museums) so I'm a bit short on conversational topics right now (unless you want some poor quality vacation snapshots of the Apollo 11 command module, for example). However, there is a topic that has lately captured my attention, and this is my bully pulpit, so ... you know the drill.

The topic of the day: panels at conventions. If you've been to a conference, be it a science fiction convention or an academic one, you probably know the idea: a bunch of talking heads on stage, discussing some topic of interest to the audience (and possibly taking questions).

The reason I want to talk about them: I'm in demand for panels at SF cons. In fact, I've done seventeen of them in the past month and about 25 this year so far, and probably something like 300 of them in the past decade. I've been on some amazingly productive, fascinating discussions — and some that were completely unorganized fiascos. And in order to avoid having to repeat myself, here's my brain dump on how to organize a panel discussion in order to minimize the risk of things going wrong.

This is part #1 in a series of four blog postings on the topic of panels at conventions. And today I'm going to talk about that most universal of considerations — the environment. We don't usually give a lot of thought to the venue in which a panel discussion takes place, but that's a mistake. Get the environment wrong, and the panel is doomed to failure before it starts. So here's what I know about how you should arrange the environment for a panel.

There is a standard format for most panel discussions. In general, if you've got a group of speakers and an audience — and the two are not interchangeable — messing with the standard format is a bad idea.

The standard format is: start with a room, with walls and a door. The venue needs to be reasonably sound-proof, lest outsiders drown out the panelists (or disrupt their thoughts, which can be equally bad). Screen partitions dividing up the floor of a vast, echoing echoing aircraft hangar like venue with other panelists trying to out-shout each other over the dividers do not work. Neither do rooms located adjacent to a hall full of happy clappy types holding a revival sing-along, or a band rehearsing for the evening concert, or a disco. (I speak from experience of all of the above.)

The room should not be too hot or too cold. In practice, this means it needs working air conditioning or central heating and windows that open. Sauna huts are not suitable venues for hour-long discourses on the philosophical roots of H. P. Lovecraft; refrigerator warehouses don't work too well either. General rule of thumb: it will be warmer in the room when there is a packed audience inside. Human beings put out roughly 0.1-0.2kW of heat per person, so an audience of 60 people is like having three electric fan heaters going at full blast. If it's warm when it's empty, it will be unpleasantly hot when it's full.

Lighting helps. Not too much lighting; at one recent convention the main function area had a stage, and stage lighting — and a low ceiling, so that there was a bank of stage lights glaring at the panelists from just above eye level: every time I glanced at the audience I ended up with purple after-images burned into my retinas. If the panelists are asking for sun screen and mirrorshades, this is a bad sign.

There needs to be a table at the front, with chairs for all the participants (too many chairs is okay; not enough chairs is bad), and chairs for the audience facing the participants. Putting the participants on a stage or illuminating them with floodlights is optional, but the audience need to be able to see their faces. A circle of chairs with nobody in a position of authority sounds nice and politically correct, but once you get past about 20 seats it gets unmanageably big; also, it encourages audience participation ... which may not be what you want, if you're running a panel where some talking heads are meant to be holding forth from positions of authority.

If there are seats for more than about 20 people in the audience, you need to provide the speakers with amplification. Whether you think they need amplification or not in that particular room is irrelevant — some of your speakers will have laryngitis, some will be shy and speak in a low whisper, and on occasion it may be necessary for the moderator on the panel to firmly overrule a too-enthusiastic member of the audience.

Ideally each speaker needs a microphone in front of them while they're on stage. (Wired mikes in front of each occupied seat at the table are fine.) Sharing a mike with a fellow speaker who has halitosis is Not Fun. Neither is playing capture-the-flag with an obsessive hoarder who has got the Talking Stick and won't let go. If it's a big room (more than 20 seats) and the panel is going to take questions, budget for an extra radio mike and a gofer who can scurry around the audience, otherwise the audience are going to be treated to the panelists answering questions that they can't hear.

It's a good idea to have pre-printed name plates available for the panelists to stick on the table in front of the audience, so that the audience can tell them apart. Such name plates (or just folded pieces of paper) need to be IN LARGE PRINT so that people at the back of the room can read them.

The panelists will probably need something to drink if they're talking for an hour. A jug of water and some paper cups are an absolute minimum; make sure there are enough clean cups for each panelist, and keep the jug full during the panel. Too much is better than too little; estimate half a litre of water per panelist per hour and you won't go far wrong. Beer, coffee, and other liquids also work well and a well-run conventions usually have a green room with staff ready to take orders from panelists, and deliver their chosen beverages.

Okay, that covers the environment (albeit minimally, without much depth; I may have to add to this as I go on). Tomorrow I'll talk about the care and handling of panelists.



Any chance a few nice, not-stalkery fans could buy you a beer this weekend, if you'll still be in DC?


Or if we can't find them, us.


"Once More With Ceiling!"

Green room delivering drinks is a UK phenomenon, in the US, drinks are in the Green Room proper. This is a Yet Another Effect of corkage -- US conventions can get corkage in certain spaces fairly trivially, for UK conventions, it is effectively impossible without staggering amounts of money (read, impossible for anything short of a UK Worldcon, and difficult for them.)

Since the bar is the only source for drinks, and you really don't want to put the Green Room in the bar if you want to actually have panelists appear on panels, the delivery system evolved.


One other basic requirement for the room: big enough! I've been at a Worldcon where several popular speakers were put into small rooms that only had seats for about 50 people, and 200 wanted to get in.


This is great, coming the day after I got my second con invite of the season. I'm always interested in being a better panelist and looking forward to the rest of this series.

The "environment" one I hate is when they change the room for a given panel, tell the fans, tell the gofers, but don't
tell the panelists...


Pen and paper.

Nothing fancy, as long as it works, but sooner or later somebody is going to want to make a note of something.


"Human beings put out roughly 0.5kW of heat per person"

Incorrect. It's actually 100 W, not 500 W.

2000 kilocalories / 1 day = 96.9 W
2500 kilocalories / 1 day = 121.1 W

People doing intense physical work can put out far more heat; burning 5000 kilocalories in 8 hours equals 726.4 W. You won't see that except at a convention of people trying to break the world sheep-shearing record (to take one example from an Internet search), etc.


"otherwise the audience are going to be treated to the panelists answering questions that they can't hear."

Unless things are going particularly well it's usually worth having a repeat or summary of the question from the stage.


WRT @8, how you deal with audience questions that were not spoken into the microphone, Charlie? I often listen to recorded panels over the internet, and question time tends to drive me nuts because you're always left guessing what the questions were.


On Disrupting Panelists' Thoughts

When you consider how facile panelists are at disrupting their own thoughts, they really don't need outside assistance.


Branko @ 8: As someone who's moderated a fair few panels, my preferred approach is to restate the question for the benefit of the audience. This also gives the panel a few extra seconds to think about it...


Oops; (a) that should have been 'Branko @ 9', and (b) 8 kind of answered it.


As a lighting/noise tech who has to do 100+ conferences and talks a year, I can say that the main factors in deciding the environment is cost and amount of punters expected.
Even a small scale conference, with only (say) 200 delegates can cost in excess of £4000. This is with minimal stage lighting, a single rear projection screen and two mics (one at a lectern and one wireless hand-held for Q&A with the punters). If you want to add extra mics, particularly radio mics, then factor in a cost of roughly £500 per mic, depending upon whether it's a simple wireless stick or a proper body-pack. And this is without the cost of the venue, which can be extortionate to say the least! So, if you want 20 panel members to be individually miced up, then the cost'll shoot up to £16k. Doesn't sound too bad, but it depends on the level of event. For something like the M$ worldwide forum or the NUT, I'd expect in excess of 8000 delegates/day. But, the production budget on each of those is probably around £100k, at least.
So, it's impractical to give each panel member a mic unless it's for something massively big, and even then I've never really seen it happen.
Lighting's a whole other ball game... If you don't want conferences lit, that's fine, but most people like to have something that makes panellists stand out from the punters. I agree, lack of height sucks, and can make for an uncomfortable experience if you're on the panel, but sadly most suites used for panels are low-ceilinged, so it's virtually impossible to get enough height to make things comfortable...


Tim @13: to mike up 20 panelists at an SF convention implies four parallel program streams with 4 panelists/1 moderator per panel. That's a big SF convention. Likewise, rear projection screens don't really show up except at the high end of that scale. What I'm talking about is a brace of four or five wired mikes on the podium, left in situ for each successive panel to re-use, and a single "simple stick" for a gopher to walk the audience with.

As volunteer operations, SF conventions run on a shoestring using volunteer labour; you can probably factor a cost of £50-100 per mike, and they will almost certainly be reused and shared between conventions rather than rented or purchased specifically for the occasion.


I'll be interested to read this whole series, both as a former convention programming head (both on the San Diego Comic-Con, back when its attendance ran around 30K, and on various local San Diego conventions) and as a panel participant (mainly on gaming panels and on philosophy/politics panels). Mentioning that range of scales, in fact, brings up a point that some of your other commenters have already mentioned or implied: that optimal organization depends in some measure on convention size. A well attended panel at ConDor has 25 people in the audience and gets by fine with no microphones at all; a well attended one at Comic-Con might have hundreds, or, these days, thousands and desperately needs them—and, fortunately, the convention can afford them.

I'm curious about your estimate of 500 W/person of thermal output. My reading up on human metabolism gave me a ballpark figure of 50 W per square meter, and an average adult has around 2 square meters of body surface, which comes to more like 100 W/person. The human metabolic span is apparently a factor of 10, like most mammals, which means that a very physically active person might put out 5-10x as much—call it 1 horsepower (~750 W) as a ballpark figure—for a couple of minutes. If you can point me at a source for the estimate you're using, I'd be interested. (Books I've looked at recently that discuss the issue include The Mechanics of Pre-industrial Technology, Bicycling Science, and a few pages in The Ecological Implications of Body Size, from three rather different angles.)

Of course, even going with 100 W, a room with 100 people is enjoying 10 kW of heat output, a nontrivial space heating effect. So the basic conceptual point is sound even if we don't agree on the magnitude.

Bill Stoddard


"I'm curious about your estimate of 500 W/person of thermal output. My reading up on human metabolism gave me a ballpark figure of 50 W per square meter, and an average adult has around 2 square meters of body surface, which comes to more like 100 W/person."

Either way, combined with a form of fusion, this will give the machines all the energy they will ever need!


NB: I'm about to go back and edit the text slightly to take into account some of the points raised in these comments. No, I'm not trying to censor you; I'm trying to make a better article for future reference.

Bill @15: memories of undergrad physiology lectures dredged from the vasty deep.


Charlie @ 13:

In the UK at least all cons hire in main tech like lighting, mixing desks and microphones. The problem with buying stuff like this and passing it from con to con is just that; organising the shipping and safe arrival of such kit isn't easy or cheap. Keeping it running is also a problem -- batteries corrode and leak, stuff gets damaged in store and stops working etc.

When we hire stuff the hire company delivers the kit tested and working to the con site and makes it go away again afterwards. If there's a problem they come and replace it there and then, not a week later after the con is finished.

There are a few bits and pieces that get personally loaned to cons sometimes, like video projectors and some audio gear, but they tend to be for smaller program items and smaller cons like Novacon. Main hall/main program tech is always hired in, in my experience.


Hi Charlie,

Emily from WSFA here (blonde hair, red sweater). It was very nice to meet and chat with you last night. I'm glad I got the book and I hope you enjoy your time in the DC area! Also, thanks for posting this convention stuff - it may be helpful to people who are, say, planning a Pratchett convention. :) So I'm glad I now know about your journal, too. I'll keep an eye out for more helpful posts. Cheers!


In the distant past, I've known of cons borrowing slightly esoteric kit. Maybe more on the media side, but NTSC-compatible video was one I recall being a problem. Remember, this was in the days of videotape, and if you were lucky a PAL set might sync to the frame rate, but no colour.

Amongst the changes since then (in the UK, anyway) are the rules on portable electrical equipment. There's a need for regular checks, and certificates from qualified testers. It's not a big cost over a year of regular use, but it's awkward for borrowed equipment.

I suspect a lot of people get away with ignoring it, but they could have problems with their insurance if something does go wrong.


WRT@4 Having a room that's too big is every bit as bad as having one that's too small. Some people just can't hack large audiences. Having a few people in a massive room feels pretty bad, even though the same number in a small room would make a very successful intimate atmosphere.

Venues never have the perfect mix of room sizes. Audience sizes are difficult to predict and unforseen events can force changes of timing and venue that can throw out the best laid plans. There will always be compromises.


Charlie @14:

our parallel program streams with 4 panelists/1 moderator per panel. That's a big SF convention.
I wouldn't consider four tracks of programming to be especially large. When I ran programming at Sacramento's Eclecticon (about 500 attendees), there were four simultaneous tracks -- probably a little bit too many, but not excessively so. BayCon (about 2000 attendees) tends to have something like eight or ten panels at a time, as I recall. And most of them have only one or two microphones, for cost reasons.

Speaking as a conrunner, I'd love to have all of that equipment, but the costs get prohibitive. At ConJose in 2002, we were told, more or less, "One microphone is cheap; two or more gets very expensive, because you have to have a sound board and a technician on duty if you have more than one microphone."


Kevin @22, I spent time with one panel at Minicon (typing for Peer, so he could read what the other panelists were saying), and they had three mics for four panelists without a soundboard and tech. I think they were just all hooked up to the same speakers. Same with closing ceremonies except there were five mics. (Ran con sales table and volunteers the rest of the time.)


This is largely a difference between unionized and/or city owned convention centers and non-unionized hotels. It's much harder to use your own equipment at the former, and more expensive to use theirs. "You want to turn the lights off? That will require a union electrician, for a three hour minimum."


If you rent mics, they cost way too much (but make the hotel happy). When you're allowed to supply your own kit, mics cost around $10-20 to buy. (Hotels charge that much to rent mics per day.)


A couple of comments of tech here.

First of all lights. I can guess which Con Charlie is referring to with the issue of being blinded, as Tim noted there is always a problem with lighting panels in inappropriate rooms, added to the fact that often you have to light for cameras for the big screen. On of the issues is basically if you want your face lit so that people can see it then the lights have to point.....guess.....any ideas.....yup on the face, and hence into eyes. there are some tricks of the trade that can help but considering this is the main requirement (as well as making sure that the stage looks good without too many shadows and bright patches, these can be limited. It's not unheard off for actors to be completely freaked the first time they see an audience when the lights are dim!

Mics are another matter, I agree that each panalist should have their own mic and at least some radio mics for the audience to ask questions with. Moderators should ensure that only those asked through the radio mic are addressed!

As for costs, tech costing for a con isn't the same as for a professional event. Even disregarding a marked reduction in cost for using volunteers, the equipment can be managed between programme items IF THE PROGRAMME IS WRITTEN WITH THIS IN MIND!!!!!! therefore no programme sub-committee should be complete without the tech ops manager being involved. A good TOM can save you quiet a bit of cash not only from good at con logistics but also from a good relationship with their suppliers.

Buying or Hiring is always a big issue if you are an ongoing event that runs from year to year with the same committee and staff then buy. If not then Hire. you'll thank your supplier at 3 in the morning when they arrive with the spare projector bulb for the Rocky Horror Picture Show programme item!


I'm curious about the "table in front" bit -- in my experience (doing mostly political or academic panel discussions), the table in front is helpful (to put drinks and papers on it, and to hide the sight of nervous legs of bored participants), but it a more open format (thing: TV talkshow, some armchairs or sofas) works better if a less formal style of discussion is wanted. But of course, even in such a setting, places to put down drinks, possibly mikes and so on are necessary.


...if you want 20 panel members to be individually miced up...

A wonderful image. Cheered me right up.


Charlie, are you going to mention "Handling Questions From the Floor" in one part of your 4-parter?

IMO and IME this is usually done poorly. I think there's (at least) 3 distinct problems.

Problem 1 is, probably obviously, people don't "wait for the mike" -- when there is one or more (!) but you covered that -- because they can't see where it is. IMO it's the job of the "panel wrangler" to "set up" the question session by pointing out the mike-runners, and it's up to the conference organizers to put the mike-runners in some kind of uniform. Logo'ed t-shirts are enough, preferably in a colour that will CONTRAST with the bulk of the attendees!

Problem 2 is that speakers often don't know how to say "no". Again, a good panel-wrangler can set this up. "We're going to take questions now. The mikes are there and there. Wait for them. If a panelist feels that a question is beyond the scope of what we're focusing on here today, they can simply say "Beyond Scope" and we'll move politely and respectfully on to the next question".

Problem 3 is the audience questions. No-one, and I mean no-one, wants to hear your 2 minute enthuse about whatever before you ask your question. Not the panelist. Not the audience. No-one. Again, a good wrangler can control this. "This is not a 'make a long comment with a question tacked on' session. Keep your questions short and to the point - thanks!"

IME American audiences are the worst with 3. "I just wanna says" and "I just wanna tell yas" are of interest to absolutely no-one but you, folks. If you were interesting, they'd put you on a panel...


Shan @29:
Your comments on point 3 remind me of Monty Python's Meaning of Life when the American argues with Death:

"Shut up you American. You always talk, you Americans, you talk and you talk and say 'Let me tell you something' and 'I just wanna say this', Well you're dead now, so shut up."

I love that movie.



About this Entry

This page contains a single entry by Charlie Stross published on April 4, 2008 4:00 PM.

Beer in DC was the previous entry in this blog.

Panel discussions at SF conventions, part 2: keeping the panelists happy is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.

Search this blog