When I first started doing research, everything was paper based. I went to the library and looked things up in a card catalogue. There what I think was a two volume annual publication that allowed researchers to do a keyword search of things that had been published in the previous twelve months. They would then check the card catalogue to see if the library had whatever they’d found that looked useful and if not, would complete an interlibrary loan form. I have no idea how the library staff found out what libraries might have the book or journal that had been ordered, but it usually eventually arrived.
While I was doing my Masters’ Qualifying in the 1990s, the library of my institution got its first electronic index. This was burned on a CD-ROM and posted to the library. I think it came out once a month, and apparently it was very expensive. The chief librarian watched over it and its users like a hawk and mere students were absolutely not allowed to touch the actual CD, but it did speed up the searching.
Now, I can log into the library databases from my office or home study to do searches to find out what has been published that might be relevant to my work and usually my search will also tell me if my library has it available. I can then develop a list of items that I want to borrow, print them out and take them with me. I can find out which items someone else has the temerity to have out when I want them without having to go to the bother of checking the shelves. I can put a hold on these items and I’ll get an email when they come in. I can do interlibrary loan requests on line and I will also get an email when these arrive. I can also use Google Scholar and WorldCat. So much information is literally at my fingertips.
The upside of this is that it’s so much quicker and easier to find out what has been written that is relevant to your research, which is stimulating and exciting. It also makes it much easier to ensure that your research is well grounded.
The downside is that you are therefore expected to find the relevant material, read it and comment on it. And because plagiariasm is such a big deal it’s essential that you make as sure as you possibly can that you have found everything. If the idea that popped into your head has already been pubished by someone else (because it is, after all, a logical conclusion to draw from the facts) in a language that you can reasonably be expected to understand, and you don’t acknowledge this and explain why your formulation of it is superior or at least how it confirms the previous research, you could be putting your career on the line. This makes most researchers I know not a little uneasy, so they search as meticulously as they can.
It would be interesting to see whether it has had any effect on how long it takes for ideas to reach publication (although I have no idea how you’d design this particular piece of research, seeing I doubt that there are any “before-the-internet” statistics available)! :-) A casual glance at the material in my personal library suggests that it has resulted in more citations per journal article or book than was common several decades ago. It would be easier to design research to confirm this, but not something I currently have the time or inclination to do. And maybe someone else has already done it? :-)