How did conferences “work” before the e-mail era?

Question

Some friends and I were discussing a certain conference that has a "rebuttal" phase, in which the authors can see the reviews and reply to them. We realized that this rebuttal phase probably couldn't happen before e-mail was common, which lead us to think on the question

How did conferences "work" before everybody had e-mails?

I don't know about other fields, but in computer-science, some of the conferences are dated to early 60's. Back then there was no email and internet (were fax machines common enough?).

I have plenty of questions about how the process worked back then:

• How did one submit a paper? (I assume one had to print the paper and mail it? Was the "deadline" determined by the post-date, or maybe there was no deadline?)
• How did one get the reviews back, if at all?
• Where did one send his paper? To the program chair? How did the reviewers get a copy of the paper (I assume the chair would first need to look at the abstracts and then assign to other PC members/reviewers? or maybe the entire committee would get copies of all the papers, and there was no sub-reviewing?)
• How much time did the entire process take (seems like the overhead of mail adds quite a lot to the already-long process)

(maybe I'm taking it all wrong, and conferences back then were not peer-reviewed, and anyone that came could give a talk??)

---

Some things that I (think that I) know, and might help to complete the atmosphere of 'making science without the supporting technology':

• How would the program committee make the decision? - they'd meet in some place together, and decide.. Which means that one had to travel quite a lot to be on a committee.
• Proceedings were sent to print months after the conference, so they actually described what had happened during the conference.
• Presentations were done using a projector and transparent slides (it's funny to think about it..)

Answer 1

Conferences were organized pretty much as they are now, only slower and with lots more paper.

• Conferences would advertise by physically mailing paper fliers to potential authors (primarily past attendees) and perhaps some department chairs. Postage was a significant portion of the conference budget. Professional society publications (like Communications of the ACM and SIGACT News) would include several pages of calls for papers with upcoming deadlines. These calls for papers were typically prepared and sent to publishers/printers about a year in advance of the conference.

• Authors would send multiple physical paper copies of each submission by post (or FedEx, or whatever) to the physical address of the program committee chair. Typically, the author would send one copy for each member of the program committee. Papers usually had to be received, not merely postmarked, by the submission deadline. (Hence the probably-not-apocryphal stories of grad students flying to the PC chair's city with a box of last-minute submissions from their home department.)

• Submission deadlines were typically about six months before the actual conference date, just as they are now.

• The PC chair (or more likely, their secretary and/or students) would collate the received submissions into boxes/binders, which would be physically mailed to each program committee member. Thus, each PC member would receive a copy of each submitted paper. Shipping costs were a non-trivial portion of the conference budget. (Having never been on a committee organized this way, I don't know how review assignments were done. I assume the now ubiquitous practice of sub-reviewing was very rare.)

• A couple of months later, the entire program committee would fly to a central physical location with their submission boxes/binders, to decide which submissions to accept. Conference budgets sometimes included travel costs for the program committee, but not always.

• The PC chair (or more likely, their secretary and/or students) would send a physical letter to each submitting author informing them of the committee's decision about their paper. Often this was the only feedback from the program committee. Authors who wanted more information about why a paper was rejected often had to contact one of the committee members directly.

• About three months before the conference, authors of accepted papers would physically mail camera-ready copies of their papers to the publisher, after following formatting instructions received by physical mail. The publisher would duplicate (using an actual camera!) and bind these papers into books, along with front and back matter prepared by the PC chair and also physically mailed to the publisher.

• The publisher would ship physical books containing all the camera-ready papers, which would be distributed to the conference attendees. (Again, shipping costs were a non-trivial portion of the conference budget.) Faculty attending the conferences would often buy extra books to distribute to their students who could not attend. Other copies of the same book would be physically mailed to hundreds of university libraries and other subscribers.

Answer 2

I have heard many stories of people rushing to the FedEx station at the airport to ship off papers at the last possible minute. I've also heard stories of grad students driving from Boston to New York with all the submissions from MIT to submit them.

Answer 3

There are some vestiges of the pre-electronic era still out there. For instance, the American Physical Society continues to this day to have an annual "Sorting Meeting" for their conference, where anyone interested in helping to organize the meeting comes together. At that meeting, all of the abstracts, which have been printed out, are available, and need to be "sorted" and collected into groups that would become different sessions.

You will also sometimes see references to manuscripts needing to be "camera-ready"; this is also a reference to how production of conference proceedings worked. The manuscripts were photographed and then turned into material that the publisher used to make the final printed copies.

Of course, even after the advent of the fax, many things would still have to be done by mail. However, I think that ultimately things had to be done "on site," as it was simply too impractical to mail everything around to all of the reviewers. However, I also think it would have been much more difficult to ask for full papers, just because of the logistics of having to deal with so much material in one sitting. You would need to rely on abstracts (regular or "extended"), and then ask for full papers from the accepted papers.

(And, because of the poor quality of faxes, anything that would be submitted to a publisher would have to be done by mail. Note that this was true even after the advent of the email era!)

Answer 4

Typically, each program committee member would receive ALL the submissions (100-200 back in the 1970s and 80s), when STOC and FOCS did not have parallel sessions and were only accepting 30-36 papers. Authors with accepted papers would receive a mailing tube containing large size sheets into which their papers were to be typed (or pasted), with blue guidelines indicating the margins for the 2-column format. These were then sent to the publisher, who would photograph the pages - the proceedings pages were reduced-size versions of the "camera-ready" versions. Figures were often hand-drawn, as you can see if you look at some of those old proceedings, scans of which are available in the relevant digital libraries.