I speak at a lot of conferences and meet-ups, too many. I shouldn’t complain I’ve got books to sell! I’ve also been involved in organizing many conferences – ACCU, EuroPLoP and Agile on the Beach for the last 10 years. There are some trends emerging which worry me and I want to record. So not my usual blog post but I hope this will interest a few people.
Now so many presentations are done remotely I can speak to groups all over the world, last week I spoke to an internal company group in Florida and a couple of months ago to a meet-up in Sydney. I say Florida and Sydney but the attendees were from all over the world, Florida was just the home of the company and Sydney were the group physically met before you-know-what happened.
Increasingly “big name” speakers – maybe me, certainly people bigger than me, people with serious book sales and name recognition in the tech/agile community – are turning up at local meetings and small conferences which are now online. That is good because these groups get to hear directly from people with big ideas.
But, the fact that the big names (and here I probably include myself) are speaking to such groups means others aren’t. In particular local people, new voices, people just starting on the speaking circuit or people who will probably only ever speak to a few events organised by people they know.
My worry is that as more events move online we are perpetuating the winner-take-all culture and making it harder for new people to get started.
Arguably that is offset by fact that more conferences are reviewing submissions anonymously. However I’m not sure anonymous review is a good thing. As a reviewer I normally have two pieces of information: the talk description and the bio of the submitter. I’m looking for an interesting talk and an interesting speaker. Without the bio I only have the talk description to go on. If I select a big name I know I’m selecting a name, when I’m reviewing someone I don’t know I think of them differently.
Consequently reviews are going to favour those who can write stronger, better, descriptions. It may be naturally talented writers, or those who have had the chance to hone the description over previous submissions and presentations. Again favouring the established speakers.
Adding to that is the fact that some speakers work for companies who will help them prepare talks and descriptions. This is more likely in consultancies – or those successful enough to pay for professional help. One comment I heard from conference attendees regularly is that they prefer to hear from non-consultants, those doing actual work, rather than consultants who may have a service to sell. (There are more consultants on the speaking circuit than non-consultants because they have more time and greater motivation to be seen speaking.)
So while well intentioned anonymous review may end up having the opposite effect of that intended.
Next I’m worried about the increased use of online submission systems which create a common pool of speakers – I’ve thought about this long and hard because I looked into this when developing Mimas years ago.
At first this looks great for speakers because they can easily submit too many different conferences. It is good for conferences too because it simplifies submission and increases the pool of speakers. But again this can back-fire.
I once shared a taxi with a speaker whose next conference was in South Africa. He had never heard of the conference but an online system prompted him to submit and guess what, he was accepted! Great but…
As a conference organizer I’ve had to manage over 300 submissions for less than 40 speaking slots. To do the submitters justice requires a lot of review work. A 1:8 ratio is extreme, most conferences it is closer to 1:3 or 1:4. How many conferences can do that?
The last year I ran Agile on the Beach reviews I had over 50 reviewers. That itself makes more work. How many conferences can handle that? Some conferences don’t even have that many attendees.
I used two review rounds, again more work.
And, to be honest, if I have 30 submissions to review, the one I review first will get more thought than the one I review last. Again those who can craft a good description are more likely to get through.
So, common, pooled, online submission systems will increase the number of speakers. Less work for speakers makes more work for organisers. Either conferences will need to invest more in reviewing work or they will end up taking the people they recognise when the cloak of anonymity is raised.
And how are new speakers to improve if they don’t get selected?
Very very few conferences give feedback on submissions to those who submit and are not selected. And as the number of submissions rises the work needed to give feedback also rises. (Particularly since raw comments from reviewers often contradict.)
(Perhaps the thing I am most proud of about Agile on the Beach is that I gave feedback to everyone who submitted. That came at a cost to the conference, and even more, at a cost to me. Ultimately I had to write Mimas to do what I wanted to do and that took a lot more work than I was ever prepared to admit to myself.)
Another side-effect of common submission systems is that conferences organisers have easy access to alternative speakers. There has always been a bottomless pit of people wanting to speak but now they are easy to find. This further reduces the speakers power to ask for travel costs, let alone actual payment. Few speakers get paid for speaking at conferences or meet-ups, with luck the organiser can pay travel costs. This already means those who are successful and can afford the time, and travel costs, are at an advantage. (And again favours consultants for whom speaking is marketing. Consultancies may actively encourage their people to speak, while work-a-day companies are frequently less than keen on employees taking time away from work.)
All-in-all I think well intentioned moves (online talks, anonymous review and pooled submmision platforms) are actually making it harder for new voices to get heard. As so often happens in technology the winners win more and a greater divide opens up.
I don’t know what the answers are. Maybe my concerns are misplaced. But I worry.
Hi Allan, I’ve enjoyed your blogs on specific topics and I like this one too as I think it raises a number of good points that are often on my mind too. Having a variety of voices and diversity of speakers is so important for a profession, community, audience and for the legacy speakers leave behind.
How do speakers get to hone their craft without opportunity? Everyone has to have a first time and I think well known or experienced speakers can do a lot to support and bring on others. Allyship has become a popular term of late and I think the speaking community could embrace the concept too to lift up and showcase voices that might not otherwise have the opportunity to be heard.
Yes that might mean they miss out on an opportunity, I’d say it opens more doors even if it closes one and I think it shows a braver, boldness and confidence that there is enough to go around rather than a scarcity mindset.
I’m not without hope although I do think it will take a number of levers, nudges, feedback and encouragement to enable this to happen.
Thanks for the comment, and thanks for introducing me to “Allyship”, never heard the term before, now I’m looking it up!
Hi Allan,
I share some of your concerns, but I’m not quite as pessimistic (at least, yet).
As you know I’ve also built my own CfP system (which is now being used beyond just my own conference – even if it’s not quite ready for that!). I’ve tried to address many of the problems with such systems in my own implementation.
1. For now, each event is a silo. So even if its the same codebase, it’s a separate deployment with its own DB. You can’t resubmit to a different event with a click. I’ll be moving to a model where the user db may be common – but proposals would still need to be resubmitted for each event. That’s a deliberate choice to try to add “just enough” friction.
2. The reviews system is anonymised. I agree that there are pros and cons both ways. I want to maximise the pros of each approach, and minimise the cons. Of course 🙂 So the first round is anonymised and made easy enough (relative rankings, like Sessionize) that lots of people can do lots of reviews. Then there is a second round that is not anonymised, that takes account of things like balance (not too many talks on the same subject, making sure there are enough “new voices” etc). That’s a much smaller pool (for my conference, in some cases just me).
3. The reviews process is also randomised. This is to mitigate the “review fatigue” issue – by spreading it out. It also helps to give a more consistent number of reviews (although random, it is also weighted towards the proposals with the fewest existing reviews).
For now the biggest downside is I have not yet added the feedback feature. I agree that this is very valuable, and it’s definitely on the roadmap – hopefully before my next event.
One thing I have found, so far, is that the anonymous round has selected well for diversity. I’m particularly pleased about that, because adjusting for diversity in the second round is problematic. I do consider diversity criteria in a “tie-break” situation, but other than that can be hands off and allow it to work out naturally. I consider that a big win – and worth paying the cost of sometimes missing out on good talks that were not as well proposed as they could be.