Conference poster sessions can feel like anything from one of the main reasons to attend a conference, giving participants the time to explain their work more personally than is possible via presentations, to an afterthought, hidden in the back room of a secondary venue scheduled during an already short lunch break. The former is great. I think the latter can be improved.
At larger conferences, posters seem to be mainly used for their space and time saving parallelism. It's not possible to have 227 talks at the same time, but a big enough room can hold 227 posters concurrently, as was proven at the 2017 conference on Neural Information Processing Systems which had 679 posters (one for each accepted paper) over three four hour sessions. At smaller conferences or workshops, posters seem (to me) to be primarily a tool to get more people, and more ideas involved, with the downside that the sessions often feel secondary to the main event.
An alternative to poster sessions for small conferences/workshops
Here's the vision. One of the conference sessions is held in a large room, with tables and chairs scattered around. Behind each table sits a person underneath a banner outlining a single proposition that they want to argue for. Conference attendees spend an hour moving between tables (encouraged to go to those they disagree with or find most novel) challenging the presenters to defend their claims.
Presenters may bring any amount of supporting material to help make their case. They could just bring their arguments or they could bring a demonstration. They could bring some reference material or some props. They could even bring a poster if they so desired.
I imagine anyone that wants to defend a proposition could submit a simple form prior to the conference:
Proposition you will defend (max 20 words)
How will you support the proposition? (max 100 words)
Special requests (e.g. poster board, display screen):
At the end of the session a feedback form could be used and the most persuasive would be given a short slot to present their claim in the more traditional way. Or, even better, those that voted for them would attempt to present their claim for them.
Perhaps this is common in fields that I haven't worked in. Perhaps it wouldn't work for some reason I haven't thought of, but as far as I can tell it would be a more interactive, less printing-intensive way to get the same benefits as poster sessions with less of the downsides.
Before this gets rolled out worldwide, let's agree that, at the very least, poster sessions should have their own slot timetabled for them. In the smaller conferences, where the posters are generally presented by less-established researchers this leads to a more inclusive feeling.