I see increasing calls for conferences to allow remote presentation. This is a terrible idea that will ultimately lead to good things. Some thoughts. »
I understand the motivation. Travel is expensive. Visas are hard. CO2 is a thing. Harder still with mobility/health issues or living far from the action. And they're gatekeepers. »
But here's the thing. Conferences are not primarily meant as write-only media. It's in the very name (roots: conversation, getting together). Focusing on talks misses most of the value. »
Also, many people probably haven't had much experience with remote talks. I've done remote invited talks a few times. It sucked every time. Probably for them, but even for me. »
Still, initially it'll be great. For a few. I conjecture there's a tipping point around 15-20%: when attendees see 1 in 5-6 papers presented remotely, they'll say WTF and stop attending. And the whole conference will collapse. Gradually, then instantly. »
And I think that's the good thing. We need to disintermediate conferences. CS publication will become more like journals in other fields. (Break the ACM stranglehold!) One difference is we now have a great DNA of slides/talks/etc., and we'll still record those and attach them. »
We've been inviting short videos to go with papers in Journal of Functional Programming. No reason to stop there. We can have a 25-minute conference talk, too. In fact, multiple lengths. All of a question of what authors want to produce. »
Costs go way down. CO2 goes way down. Visas are a non-issue. And the stranglehold of deadlines and awful conference hotels and terrible catered food and annoying jetlag and pressure on families goes way down. »
We would still need a way to recover the notion of con-ferring. Maybe make them _just_ social events (with just enough "science" to justify travel). Can be in convenient locations/times. Can be more local. Look for ideas in other fields (MLA, etc.) that have journals. »
Maybe, also, the gates get propped open to be more equitable. (Researchers probably already know how this plays out in other fields.) My conjecture: only a very small improvement. There are so many other variables. »
In short, I support remote presentation, but not at all for the reason most people are advocating, but rather as a Trojan horse to break the system and build a new and better one. •
