41

I’m a mathematician (after a fashion), working in industry, and I find great value in older books about geometry, written between around 1850 and 1920. I notice that my favorite books are often offered for sale on the internet, and, after I purchase them, I typically find that they came from some academic library, somewhere. I have mixed feelings about this: I love the fact that I can own the books, but it worries me that libraries are ditching them.

Now my question: when academic libraries get rid of books, do they typically scan them, first? Or, is there some systematic effort to convert older (no longer copyrighted) materials into digital form. I can’t stand the thought that this stuff will be lost.

8
  • 10
    Google Books has many millions of books scanned. That’s where I’d check. I very much doubt many libraries are scanning books on the way out as an individual project-they’re selling them due to lack of interest, after all. May 1 at 6:35
  • 9
    Don't forget that the interlibrary loan system implies that not every library needs to keep copies of all books, especially those that are seldom requested.
    – Buffy
    May 1 at 11:11
  • 12
    I've retrieved books from library disposals on several occasions; each time the stated strategy was about disposing of duplicates (the library had had multiple copies) or older editions, with the claimed intent that there would always be one copy of the book remaining available.
    – Lou Knee
    May 1 at 11:59
  • 6
    I worked at a university library and it is common for librarians to identify books in the collection as not needed any more, due to age, condition, duplication, lack of recent circulation and/or a note from a professor saying that the material in the book is covered in newer books. These books are then "withdrawn" from the collection. If they are to be sold, they should be stamped "WITHDRAWN" somewhere at the beginning of the book. Unfortunately books are also sometimes stolen or lost. You should be able to ask the library if the book is still part of the collection.
    – Wastrel
    May 1 at 13:22
  • 3
    On the plus side, in the past 15 years I've gotten books that I never expected to find for purchase (before amazon, that is). Not only that, but many of the 100+ such books I've gotten over the years are nearly untouched and in excellent condition. Two recent acquisitions, each about $50 and essentially untouched, and each I'd been searching off-and-on for many years for a copy under several hundred dollars: Naber's Set-Theoretic Topology and Łojasiewicz's An Introduction to the Theory of Real Functions. May 1 at 16:47

5 Answers 5

8

Other answers here have mentioned that libraries typically do not scan their holdings, and have described why this is both more expensive and more complicated than you might expect.

It's worth mentioning, in this context, that although the MLA sector (museums, libraries and archives) seems rather homogeneous from outside, those in the business see those three strands as being very different from each other.

In particular, the big difference between archives and libraries is when and why they chuck things out. And all three have to be selective about material, for the brutal reason that they have finite amounts of shelf-space (or digital analogue). I once heard someone giving a talk in this area remark that ‘the dirty secret of the library sector is the “deaccession facility” (sc. the large skip) at the back door’.

An archive deletes/destroys/discards stuff at accession time. Archives are typically given stuff rather than actively seeking it (a national archive might have a very specific obligation to hold on to government materials no longer in active use, and might have elaborate rules concerning what they are obliged to accept and whom they are allowed to show it to afterwards). When material is offered to the archive, the decision is made whether to accept it or not, made on legal grounds perhaps, or bearing in mind the possible interests of future historians. Once it's admitted to the archive, though, the typical plan is that it stays there.

A librarian, however, would regard themself as serving current information users, as opposed to historians yet unborn. They are more willing to accession things, and also more willing to deaccession them if they're not earning their keep. Different libraries will obviously have radically different notions of what counts as a book ‘earning its keep’. A national library might well have an obligation to be given and to accept every book published in the country. One library might see its user population as including the scholars who would eventually want to read that 19th century geometry book; another might see that book as simply taking up valuable shelf space that its users would prefer to be filled with a new purchase.

Both are perfectly rational libraries. Libraries also know each others' holdings, so they'll typically try not to throw out the last copy of something: ILL is your friend and theirs.

Museums are a bit of both, but dealing with miscellaneous lumps of stuff other than books.

43

No. As indicated in other answers they do not scan them, and usually they scrap them. When you see ones being sold, this is actually a library being considerate. Some libraries send them for pulping or even landfill in huge quantities.

It is because, often, academic institutions and their librarians see the purpose of the library different to us, and their attitude to the books can be different to ours. This is also why they do not scan the volumes; saving the knowledge is often not seen as the purpose of their library.

I have felt strongly about this in the past and my office contains many tomes I "rescued" from waste skips behind a library!

Many academic libraries do not see themselves as repositories of past knowledge but an information resource for supporting student study. If a book is not "borrowed" by a student or researcher then often they feel it has no place in the library. I have, in the past, pointed out, that students consult many books from the shelves without taking them on loan and there is no record of this ephemeral use of the books. For example, when I set an assignment on the history of a topic students may consult many older books in their search for information but borrow few. However, the librarians feel there should just be a single book called "The History of X" that students can borrow. How that book comes to be created is not their problem!

The librarians problem also stem from not having enough funding to create the archive space needed for storage. This is why serious researchers have to depend on their National repository library, which does act as an archive, but is often not available to the average undergraduate student doing a class paper.

Although having several volumes of six figure mathematical tables, including tables of high entropy random numbers on my shelf, the library could not justify their existence on theirs.

13
  • 3
    @CodyGray I think scanning books is still fairly expensive isn't it?
    – DRF
    May 1 at 17:25
  • 21
    @DRF - Not especially expensive, but certainly time-consuming and also a copyright nightmare
    – Valorum
    May 1 at 19:16
  • 5
    I am surprised that you actually had to go dumpster-diving for those books though. All the libraries I have known had some special "free for the taking" shelf/table/box near the entrance where those books were put for a week or two before actually being tossed.
    – mlk
    May 1 at 19:41
  • 15
    "students consult many books from the shelves without taking them on loan and there is no record of this ephemeral use of the books" -- but there are commonly signs in libraries that say, "Leave books on a cart, do not return to shelf." While this is in part to keep ephemerally used books from being accidentally reshelved by the patron in the wrong place and being lost for a time, it is also so libraries can "track which items are used by people, even if those items are not checked out". So if patrons obey these signs, there is a record.
    – nanoman
    May 1 at 23:50
  • 3
    @Valorum: In the US, the books OP is talking about ("written between around 1850 and 1920") are all in the public domain anyway (because at that time, US copyright had a fixed term). Other countries' laws may differ, but in most countries, a work is in the public domain if the author has been dead for 70+ years.
    – Kevin
    May 3 at 7:27
15

When academic libraries get rid of books, do they typically scan them first? Not typically with commonly available works. If a country operates Legal Deposit most works have been archived in national repositories. Other libraries can borrow from national libraries and inter-library loans mean library visitors can get access to works no longer present in the library they visit. Since the 1990s there has been a transition from physical access to materials to digital access. The primary benefit for institutions is the savings in manpower (book handling and walking the shelves) and floor space (more study spaces and less shelving).

Is there some systematic effort to convert older (no longer copyrighted) materials into digital form? Yes. This is one aim of the Internet Archive, originally established to archive the web, it now digitally archives most media types, including books. It works with many larger libraries to digitise books. People who wish to preserve knowledge contribute to their effort to retain human knowledge.

2
  • 2
    One can use the online search feature of WorldCat to get an idea of which libraries have a physical copy of a book. The interlibrary loan of physical books is important. I read a lot of books that way. May 1 at 15:56
  • 4
    Also Project Gutenberg.?
    – Barmar
    May 1 at 18:10
13

I saw the dismissals of parts of a couple of academic libraries and they didn't scan the dismissed books. It's just two examples, but I suspect it's the commonest case. The effort to scan lots of books in a short time is huge and the personnel is usually limited (dismissals frequently happen in bulk during reorganisations or restructures).

On the other hand, there are libraries, like the Biblioteque Nationale de France, which offer the service of photocopying or scanning rare books at a price that includes the payment of the copyright fees (side note: I actually used that service from another country, and as a memento for those who come here asking how to find rare books, I can say: learn to use library services around the whole world!).

12

No, it's not typical to scan books before throwing them out - unless the book is important

I've seen a number of answers saying how unfortunate it is that libraries don't scan them, and this strongly underestimates the work involved in a useful scanning project.

I've spent roughly three years working as a developer on a large digitization project for an institution with roughly 300 years of important archives. Getting a useful scan involved these steps:

1) Working out copyright/existing agreements: This isn't an easy step. Some books can be in the public domain. Others need the copyright holder to be contacted, which may take some considerable detective work, in the case of older works.

2) Preparation of the source material: This is typically done by an archivist, possibly working with a bookbinder. Typically, in the past, you needed to remove the book spine, to turn it into flat sheets - it's a bit less necessary now as there's software that will work out how the page bows and flatten it, but for good quality scans removing the book's spine and restoring it is still the gold standard

3) Scanning: To do this at any sort of speed takes a large, highly automated, and expensive setup. Our test system could do 1 page every minute. We were up to 10 by the end of my involvement in the project, and this required 2 full time employees working with the material (admittedly, these were unique items, you could go faster on books going to be thrown out anyway, but you still need to ensure decent image quality

4) Doing stuff with the scans: This is where I come in! having a giant server full of book images accessible to the public is fine, but not very useful. We'd typically add metadata to each book scanned, we might do OCR on the documents, which is pretty imperfect still, so needs someone to skim through for mistakes. If any of the books are being released under a licence, you need to put access controls in place.

5) Keeping funding for the project so all your hard work doesn't just vanish: surprisingly, the hardest step out of the 5. Academic institutes are normally happy to commit funding for long enough to get a nice news headline, and have a thing to show off at the next benefactors meeting.

Basically, the TL:DR of this is that doing this kind of project, even on a small scale, takes a lot of person hours. Your academic library probably has limited resources to preserve, say, older textbooks - they might reasonably argue that if the material is still covered in modern textbooks, and the book isn't of particular importance, something else should get priority.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .