Currently my literature surveys do not involve trips to the library and I am wondering if I am missing out on something important. What sort of material/knowledge can the library provide that I would miss on google scholar. Thanks
I recently wrote a research survey, while on sabbatical at an institution without a physical library, but with extensive electronic subscriptions.
I never missed having physical library access, even for papers dating back to the 1840s. (In fact, the historical literature was more likely to be freely available than papers from the 1980s.) Occasionally I had to use my home university's VPN to get access to a different subset of the electronic literature, and there were one or two books that I had to download via bittor—sorry, that I had to borrow from colleagues down the hall. Yeah, that's it.
My situation is different from JeffE's. I had been looking for a couple of old CS books (in automata and switching circuit theory) published in 1960's. I finally found them in an institue library. According to the librarian there, the two books I wanted to borrow had not been checked out for at least two decades.
This is just me. My research area is not very active. I wanted to find info about what was done in those books which were not cited in modern literatures. I knew the existence of the books because I saw them when I was a graduate student in 1970's.
If your research area is modern and active, I guess you don't need to go to the library. Internet would be good enough. By the way, my personal experience is that Google is sometimes better than Google Scholar.
A side note. My feeling of reading books in the library is different from staring at the computer screen.
Don't assume that a library is a building. Libraries provide the funding and infrastructure that lets you access online journals and databases, just as they used to provide you with hardcopies and card catalogs. If you think that Google scholar provides the same degree of access you are sorely mistaken. And no, I am not a librarian, but I am very grateful that I have access to a good library.
Now, the question of whether libraries should be forced to pay for access to journals is another question.
Most things can be found online, that's for sure. But the main problem is that we tend to stay in "our neighbourhood" because of that: we can only find what we are looking for while searching online resources.
If in another field researchers use a different wording for the same objects, we will never know because it is very hard to browse a large amount of papers very quickly online.
The interest of going in a library is that it is possible to crawl very quickly amongst dozens of papers from other fields, just looking for "oh, this curve looks just like mine" or "wow, but I know this equation". This way, with luck and persistence, it is possible to find new connections and/or new way of thinking about a problem.
Library may allow you to access generally online resources behind the paywall.
In other words, while using printed material instead of online resources does not bring much benefit, some online resources (like scientific journals) are often paid, and you may need a library computer to get access to them. This is one of the reasons why you still may need to go to the library.
Of course, this is university dependent.
The advantages really depend on:
- your skills with search engines,
- the talent of your librarian.
The fact that you are only mentioning Google Scholar may suggest that you may extend your skills to other media. One single person is often affected by biases in his search. A librarian, who has experience, knows a lot of sources (some not public), has experience with many tools and more importantly, can reformulate a question with less domain-related biaises.
Physical sources are more trustful; you can check the authenticity of the information they hold since the author is always named, and so is the house of publication. All what is needed for the entire reference is shown at the beginning of the physical work; whereas, the electronic source is usually not authentic; the author can be unnamed, the house and the year of publication can be dropped, etc. additionally, reading a printed material is safer for the reader's eyes and nerves than reading on screen; the latter is evidently harmful.