Having read about historical radio astronomy and practiced signals analysis, I think they were a little more sophisticated than that. The recording in this format was made for easy perusal by a human, say, the next morning. I think that was the case with this signal. No doubt they wanted to revisit that spot in the sky looking at intensity, frequency shifting (Doppler), bandwidth, modulation, etc. But they never saw it again.

Historical point of interest: the Big Ear telescope that recorded this was bulldozed to make way for a golf course.

This is still a problem. I attended a talk a few weeks ago by a member of staff on the Square Kilometer Array Science Data Processor project, and they were talking about data volumes on the order of an exabyte... per day.

Obviously, an awful lot of pre-processing has to be done to get this down to a manageable data volume for analysis (100s of TBs), but if something gets lost in processing in the early stages, it's gone forever. Even managing and archiving the processed datasets will be a challenge.

Oh, and the second stage of the project was going to add 100x to those data volume numbers.