How did researchers find articles before the Internet and the computer era?

Question

I find it hard to write a literature review, even though I have Google, the Internet, and all free, very easy to search for information tools. I have to search for days to find a piece of information that is related to my research. I really would not be able to do that without the help of the internet and the technology.

This really makes me wonder how people in the past did that:

• Did they have to read the entire book to find a piece of information? I know they might have used the index, but still this wouldn’t have given them the details of what they were looking for.

• Does that mean people in the past (before the Internet) worked harder to achieve their degrees?

• Does that mean that research before relied less on references?

Answer 1

We depended on libraries and librarians. Grad students would spend hours in, say, the math section of a good academic library, going from book to book and taking copious notes (on paper, of course).

But, often enough, the next paper we needed to look at wasn't in that library at all, so you would go to the librarian and ask for a loan of the resource from another library.

But it was also an important technique to use the librarian as a 'knowledge expert' who could, and would, suggest things for you to look at and places to look. I don't know if librarians still get the training to do that.

And, of course, you could ask your colleagues for hints about which rocks to uncover to find the gems.

Answer 2

One point that the other answers have passed over is that there were various services the libraries subscribed to which surveyed the literature and provided abstracts and cross indexing of the primary journals. Science Citation Index was mentioned in a comment, but there was also Science Abstracts, which had been published since 1898. These were hefty print volumes, and consumed a significant fraction of the library's shelf space and budget.

You'd typically start with a paper your mentor pointed you to, use the Science Citation Index to see who had cited that paper, then look up those papers in Science Abstracts, march off to the shelves to find the appropriate bound copies of the journals, and then plug nickels into the copy machine. If your institution didn't have that journal, well, that was what the interlibrary loan was for.

Answer 3

(Comment extended to post:)

My impression is that part of the answer is "they didn't", or more precisely "they were only as good at it as their own knowledge and that of their communities". In particular, at least anecdotally, many things in mathematics were discovered in parallel for lack of easy communication and inter-visibility. [This is complementary to @Buffy's answer, which explains why people found anything at all.]

I have tried to put this on solid footing by comparing numbers of references in a 1970 issue of a mathematical journal vs. in a 2019 issue of the same journal. This turned out to be surprisingly nontrivial. Firstly, it's not clear if I am comparing apples to apples, since most journals have changed their publication criteria and sometimes even their (implicit) subject within these 49 years. Secondly, papers in almost every part of mathematics have gotten much longer (by a factor of 2.8 in my sample). Thirdly, empirically, Project Euclid bans your IP if you load more than about 20 PDFs in rapid succession, and Sci-Hub is slow and has captchas. There might be a way to do such research using MathSciNet, but I am nowhere near proficient enough at its use.

So I ended up comparing Proceedings of the AMS (due to their long back-catalog of freely accessible issues), and came up with this ("reference" means "bibliography item", not "place where a bibliography item is being cited"):

I picked 13 of the papers from Proc. AMS 141 (2019) #12 (more or less picking the first 13, except I skipped a few from the Abhyankar cluster since he writes and cites in rather idiosyncratic ways). The average paper has 1.5 references per page:

I picked 13 of the papers from Proc. AMS 24 (1970) #1. The average paper has 1.2 references per page:

Should we really compare references per page? There are good reasons to assume that the number of references per page should decrease as papers get longer, since the references cited in Section 1 won't normally be disjoint from the references cited in Section 2. As a consequence, the discrepancy between the above numbers looks even starker.

I can only explain this discrepancy in two ways:

1. the literature has grown much larger, and not just by the addition of new disciplines but also by more people writing about the same discipline;

2. (as the OP observed) finding relevant references in the literature has gotten easier thanks to the Internet, Google Scholar, etc.

I don't know how to properly disentangle these two causes.

Answer 4

Although I haven't really known this time myself, I think it's important to mention the much more crucial role that conferences and journals used to play in the dissemination of specialized knowledge. A researcher would usually try to attend the conferences of their field and get their local library to subscribe to the relevant journals in order to keep up to date with what people are doing in their research community.

At the time researchers would preciously keep these massive conference proceedings whenever they attend a conference. I know a few senior researchers who still have their offices shelves full of old proceedings and journal issues. In order to keep up one had to follow the series of conferences/journals relevant to their field, but it was feasible because there were not as many publications (and publication venues) as nowadays. They didn't have to read the entire volume but at least they would read the most important publications and make a note of potentially relevant ones. To some extent they relied less on references indeed, and they relied more on the reputation of a journal or conference to cover the recent progress of the field.

The question of whether they worked harder is subjective. The nature of bibliographical work was different and certainly more time-consuming, but researchers were not expected to process more than what is humanly possible: overall the research system was proportionate to the technical constraints. Like for many things, I think that people used to make more out of fewer resources: exploiting the resources they had access to in a deeper and more extensive way, whereas nowadays we often can only afford to skim through the massive amount of literature... I could keep rambling about disposable research but that would be very subjective and out of topic ;)

Answer 5

It was a LOT harder before the internet. Time spent traveling to and from the information source is eliminated by instant transmission of data.

If people were lucky enough to be able to spend long stretches of time in a library, the amount of information they could find was, for those days, a "lot." There were these things called "card catalogs." They were furniture, basically, in which long, little drawers were kept, and the drawers were organized by subject, and inside each drawer were hundreds of little cards, and you'd find information about one individual book on each of these cards and determine if you wanted to go walk sometimes long distances and up elevators or stairs to "the stacks" (where the books were) to find this particular book or not. (Google Dewey Decimal System.)

You'd write down the ID info about all these books onto what would often become long lists. Then you'd go to another part of the library, where the books were ("the stacks") and pull books, many of them hardbound, off of shelves. Lots and lots of books sometimes. Yes, they'd actually read "the whole book"!!! (Couldn't help but laugh hard at that question.) And sometimes not. And they'd underline sections in pencil (and ink, unfortunately), and place books face down with pages open to pages they needed to look at as their hands went to and from the typewriter, and they'd put bookmarks in between pages in an effort to be able to go back to references and material they wanted to put into the project they were working on, as they attempted to create an effective structure without Select, Copy, Cut, Paste.

If they couldn't type, they'd write it out by hand on pads of paper, and cut paragraphs and sentences with scissors (I'm not making this up) out of one section so they could rearrange the sequence of their text by taping it and gluing it to other cut-up pieces. (Google the word "mucilage.") Before this stuff called Whiteout was invented (by a mom, apparently, with stuff in her kitchen), typists had no way to correct an error on a page; they'd have to start the entire page over again.

The invention of the IBM Selectric typewriter was revolutionary and a gift from heaven because you could correct backwardly about 12 spaces. You could check books out with a library card, and you had to return them usually within two weeks or pay a small fine. And if you didn't return the book, you had to pay for the book. (One of the many good things that broke down in the social decay of the last few decades was that people started stealing books from libraries, something nobody ever did before.)

Some libraries had little tiny rooms (e.g. New York City) that they'd let serious researchers use privately for months. And they could get "a lot" (by the standards of the day) done. But it was a small fraction of what you can get done now. Someday, when you're in your 60s or so, a young person is going to ask a question that will make you laugh because you can't believe things have changed so much that young people don't know what something used to be like. It was a lot, lot, lot harder to get a degree, build a business, make almost anything... than it is now. I mean a lot, lot harder.

Answer 6

To add to the other answers: Many publishers and professional societies produced bound paper indices of journal articles such as Math Reviews by the AMS.

Answer 7

I'm surprised that no one has yet mentioned preprints of research articles. At least in mathematics, it used to be when you published an article in a reputable journal that you were given a number of paper copies of your article that you could give to colleagues. These in turn were often photocopied. Almost every time I went to my advisor's office in grad school I would emerge with another paper (either one of his own preprints or a copy of a preprint that someone else had given to him). By the time I had finished my dissertation, I had amassed a collection of a couple of dozen papers which were directly related to my topic.

When you went to conferences, many people would bring with them preprints of their own papers. You might emerge from such a conference with a half dozen new preprints.

Nowadays I don't accumulate many paper copies of articles very often (though I might still print up one I download if it strikes me as particularly important). But over the years (roughly) 1988-2003, I must have accumulated over two hundred preprints (and copies of preprints) that were given to me by colleagues. Even now I have a file cabinet full of them.

TL/DR: Before the internet, networks of researchers would keep much of the relevant articles in circulation.

Answer 8

Does that mean that research before relied less on references?

Plural-of-anecdotally, I've referred to several mathematics papers from the 1940s and 1950s and it's quite common for a 20-page paper in that era to have between zero and two references in the bibliography. Even today, papers in pure mathematics don't tend to cite a whole lot but I'd say that something like ten to twenty citations would be the common range for a 20-page paper.

As for your title question, it was very common for researchers to not find articles before the Internet and the computer era. For example,

The Bellman–Ford algorithm [...] was first proposed by Alfonso Shimbel (1955), but is instead named after Richard Bellman and Lester Ford, Jr., who published it in 1958 and 1956, respectively. Edward F. Moore also published the same algorithm in 1957, and for this reason it is also sometimes called the Bellman–Ford–Moore algorithm. (Wikipedia)

---

The Floyd–Warshall algorithm [...] was published in its currently recognized form by Robert Floyd in 1962. However, it is essentially the same as algorithms previously published by Bernard Roy in 1959 and also by Stephen Warshall in 1962 [...] and is closely related to Kleene's algorithm (published in 1956)... (Wikipedia)

---

[Dijkstra] re-discovered the algorithm known as Prim's minimal spanning tree algorithm (known earlier to Jarník, and also rediscovered by Prim). Dijkstra published the algorithm in 1959, two years after Prim and 29 years after Jarník. (Wikipedia)

Answer 9

Chem Abstracts worked well. Look at the volumes (physical, take down from shelf). Then go to the bound journals (in volumes). See if article is good...if so xox it and stick it in a file folder.

Trace footnotes back (look at articles, copy relevant ones, sometimes going couple layers back...you find quickly the seminal reviews). This is even same process done now.

If uni doesn't hold something, send in an ILL request. (Librarian hunts it down in other libraries.) Articles would often be faxed. Books mailed.

Books tend to be dated and less useful than review articles. But still, they are much quicker to skim (just physically to do so...and you have the TOC and index). Often looking at what is physically next on shelf is useful.

P.s. Now that Google is here, I'm actually amazed how BAD many kids are at looking anything up on Google. Very little "fu". I mean there are wizards also. But very frequently, people who can't even do the basics. See it all the time in classes, work, forums, etc. Q&A sites an issue too. (Nothing wrong with asking for help after making an effort. But those to lazy to Google are a plague that drive discussion quality down and take community for granted.)

Answer 10

"In the old days", I think there was much more emphasis on, and valuing of, scholarship, in the sense that there were (usually older...) people who had good memories, and had paid attention, and knew of many things. Published and preprint-only. I myself was struck by this possibility most powerfully as a grad student at Princeton long before computer look-up was possible: many of the math faculty were aware of an amazing terrain of current and older work.

In some regards, this was very important there, because very many of the most important books and journal volumes were checked out, all the time... and grad students had very low "recall" priority. So to be able to find out what was in those unavailable sources, informally, very quickly, by talking to scholars, was excellent.

Unsurprisingly, the high-level "pre-processing" done by very-capable scholars, as opposed to search engines, allowed many huge speed-ups in searches... and I think still does so.

And, with live scholars, as opposed to current fairly-generic search engines, one of the fundamental difficulties... not knowing the keywords... was easily overcome.

Indeed, from what I can see, quite a lot of the "research" posted on arXiv has failed to connect with much prior research, visibly due to failing to connect to the proper key-word (or author) world. Understandable, but "computers" do not magically solve that problem.

One of the biggest changes is simply the possibility of typing things up oneself, in presentable form, and easy error-correction. This was a significant bottle-neck as late as the late 1980s. The even bigger change, unimaginable in 1985, was the possibility of "publishing" by simply putting things on-line. "Organizing" this, if we pretend that's what we're doing, to any degree, seems to require new ideas or concepts, that are not yet "here". E.g., for math, arXiv is very useful, but: lots of pointless stuff, and many things don't appear there. (I periodically check the web pages of a list of people to see what they're doing... in addition to looking at arXiv daily.)

But, yes, "in the old days", it was important to "be on the inside", and/or be connected to people who were, etc., to get the "preprints" which would not have been (could not have been) publicly available for a year or two (due to publication lag in math... which sometimes was infinite).

And, as other answers have noted, in those days, in math, conferences really were the places where new things were announced... that might not appear in "publications" for a year or two, if even that!

We can also mention the vagaries of "physical mail", especially transatlantic... :) Things could take a month or so, or never show up at all. And, too, long-distance phone calls in the U.S. were pretty expensive. People tried to arrange to charge their grants... :)

One summary of the state of things was that there may have been a more coherent common body of "known" things, that specialists/experts all were aware of... and novices were trained-into that scholarship.

Answer 11

Another resource was memory, the one inside your head. We would remember a bit of fact we read n years ago, remember what we were reading n years ago, and start there.

Answer 12

As a math grad student in the 60's I thumbed through the orange Mathematical Reviews when it came to the library every month. I could read a paper there, or write the author a postcard asking for a reprint. I still have lots (including reprints of my own papers) but rarely use them because it's easier to find a pdf on the net than my hard copy.

Nowadays I often don't even have to know where to go for the literature. A query on stackexchange returns proofs and references.

Answer 13

Most of the methods used have already been pointed to. I'll add that Index Medicus was de rigueur for life sciences, as it doesn't look like anyone has pointed that out.

Oddly, I find that having to use the Scientific Citation Index taught me an understanding of how important tracking citation of seminal work can be in a way that graduate students don't really seem to understand.

The biggest difference I find between now and then is that the immediate ability to know how many times a paper has been cited without having to page through volumes and volumes of the SCI means that you immediately know what the seminal papers in a field are. It makes it much, much, much easier to bring oneself up to a workable level of knowledge in an area that isn't your own.

Of course, the other difference is the hours in front of the copy machine in dark library stacks.

On a separate point, another factor that might influence the number of citations in a paper over time is how EASY it is these days to pop a reference into a paper, with the advent of powerful reference managers that interface seamlessly with text and word processors. Think about adding that last minute reference when it meant retyping a whole document, changing all your superscripted numbers!!!

Answer 14

Amusing quote from Dirac on this subject:

'I worked on it [quantum theoretical interpretation of kinematic quantities] intensively from September 1925. During a long walk on a Sunday it occurred to me that the commutator might be the analogue of the Poisson bracket, but I did not know very well what a Poisson bracket was. I had just read a bit about it, and forgotten most of what I had read. I wanted to check up on this idea, but I could not do so because I did not have any book at home which gave Poisson brackets, and all the libraries were closed. So I had to wait impatiently until Monday morning when the libraries were open and check on what a Poisson bracket really was. Then I found that they would fit, but I had one impatient night of waiting'.