In reading through the various options for formats, I can clearly see that the USENIX Security ‘25 program committee co-chairs and the USENIX Security steering committee have put in a lot of thought and effort into objectively looking at all available options, and I more than look forward to feedback from other members of the community.
I would however like to bring forth something that I personally consider to be a glaring issue when it comes to the planned conference format, and specifically the following sentence:
Each accepted paper will have a talk slot, but some papers get longer talks (tentatively set at 15 minutes) and some get shorter talks (~1 minute)
I understand that these are preliminary and relatively arbitrary numbers regarding the length of talks, but considering the general trend in security conferences to slowly become larger and larger [1], I believe that it is eventually, if not immediately, going to get to a level where indeed, most presenters will only have time to introduce themselves and little else.
Looking at the FAQ, at least one author is expected to present (and probably also have to pay for a registration to the conference). Looking at this from the POV of a presenter, I would be disappointed to be given a slot of a couple of minutes to present a work that I have been working on for months if not years. I understand the A-3 proposal is meant to counterbalance that, but I think most people would agree that a poster does not have the same gravitas a talk does. Furthermore, presenting a poster to multiple people that may wander from poster to poster within a very limited time slot is likely to be a daunting task for many authors.
From the POV of an attendee (I attended Usenix '23 as a co-author), I found myself to be interested in a great number of talks which I attended, missing a few here and there due to track overlap (as others have already shared). If those talks were to effectively be preludes to a poster session, I do not believe I would have enough time to visit and understand them.
Finally, I strongly believe many presenters would probably be put off by the possibility of paying thousands of dollars (registration+travel+hotel+other expenses) and a week of their time for a given talk slot of a couple of minutes - especially for a work that they put months of effort in. I am of course unsure if this is feasible financially, but waiving the registration fee for a presenter might "soften the blow" here.
For these reasons, I do not believe that shorter talks is really the way forward, or sustainable. I don't think there is a perfect (or even great) solution, but I strongly suspect that the proposed format, if followed, perhaps not immediately but within a couple of years would have to be changed yet again.
In reading through the various options for formats, I can clearly see that the USENIX Security ‘25 program committee co-chairs and the USENIX Security steering committee have put in a lot of thought and effort into objectively looking at all available options, and I more than look forward to feedback from other members of the community.
I would however like to bring forth something that I personally consider to be a glaring issue when it comes to the planned conference format, and specifically the following sentence:
I understand that these are preliminary and relatively arbitrary numbers regarding the length of talks, but considering the general trend in security conferences to slowly become larger and larger [1], I believe that it is eventually, if not immediately, going to get to a level where indeed, most presenters will only have time to introduce themselves and little else.
Looking at the FAQ, at least one author is expected to present (and probably also have to pay for a registration to the conference). Looking at this from the POV of a presenter, I would be disappointed to be given a slot of a couple of minutes to present a work that I have been working on for months if not years. I understand the A-3 proposal is meant to counterbalance that, but I think most people would agree that a poster does not have the same gravitas a talk does. Furthermore, presenting a poster to multiple people that may wander from poster to poster within a very limited time slot is likely to be a daunting task for many authors.
From the POV of an attendee (I attended Usenix '23 as a co-author), I found myself to be interested in a great number of talks which I attended, missing a few here and there due to track overlap (as others have already shared). If those talks were to effectively be preludes to a poster session, I do not believe I would have enough time to visit and understand them.
Finally, I strongly believe many presenters would probably be put off by the possibility of paying thousands of dollars (registration+travel+hotel+other expenses) and a week of their time for a given talk slot of a couple of minutes - especially for a work that they put months of effort in. I am of course unsure if this is feasible financially, but waiving the registration fee for a presenter might "soften the blow" here.
For these reasons, I do not believe that shorter talks is really the way forward, or sustainable. I don't think there is a perfect (or even great) solution, but I strongly suspect that the proposed format, if followed, perhaps not immediately but within a couple of years would have to be changed yet again.
[1] : https://csconferences.org/