The vault where Harvard keeps millions of books(gizmodo.com)

40 pointsby benbreen11 lat temu13 comments

The vault where Harvard keeps millions of books

http://gizmodo.com/a-glimpse-inside-the-hidden-vault-where-harvard-keeps-m-1684024865

6 comments

I'm partial to the Elmer L. Anderson library and the Minnesota Library Access Center located at my alma mater, the University of Minnesota. It might not be quite as large as Harvard's store, but instead being located in an warehouse, it's located in two lime stone caverns carved into the bluffs above the river.

  The Andersen Library has been characterized as "geology-friendly;" and indeed, the geology of the site is an integral part of the underground structure. Behind the concrete portal, two enormous caverns, each two stories high and the length of two football fields, lie below a 30 foot layer of limestone plus an additional 30 feet of topsoil, clay and gravel. The solid limestone serves both as the structural roof of the caverns and the foundation for the Andersen Library building above. [1]

[1] https://www.lib.umn.edu/andersen/about-building https://www.youtube.com/watch?v=ihLwcqSJzF4

gwern11 lat temu

> When I was an undergraduate at Harvard, I knew the Harvard Depository only as it appeared in the online library catalogue. I could click a button and a book would magically appear on campus a couple days later. By my senior year, I could click a button and an email would pop up in my inbox a couple days later, with a scanned PDF of the book section I requested.

Nice.

fernly11 lat temu

Interesting. They are implementing, in effect, an LRU cache algorithm: scanning books as they are requested. The next person to request that book gets an email in 30 seconds, not two days.

thret11 lat temu

What are they discarding? It's more of a FIFO isn't it? I assume that when they have no new requests they are scanning books based on some other criteria.

cjubb3911 lat temu

Perhaps it's best described just as an infinite-sized cache. I would assume they have enough request to very rarely be idle.

gwern11 lat temu

Copy-on-write might be a better concept to allude to. You're making a new version, but efficiently, on demand.

prbuckley11 lat temu

Here is a link to the 24 minute documentary about the Harvard Depository...

http://librarybeyondthebook.org/cold_storage/

Very cool how they use the depository floor plan as a way to navigate the film.

peteretep11 lat temu

Bigger, prettier, older: http://en.m.wikipedia.org/wiki/Bodleian_Library

insipid11 lat temu

Not really, debatable, barely: http://en.wikipedia.org/wiki/Harvard_Library :)

Maken11 lat temu

I fail to understand why they haven't scanned all the books yet.

caseysoftware11 lat temu

I used to work with the Library of Congress' digitization efforts, so some perspective..

I know you said books but the larger problem is that sometimes the equipment to play back the media is no longer available.. or damages/changes the media in some way.

For books specifically, to get the entire "work" many scanning efforts are destructive in that you have to cut the binding of the book to get the full page.. because if there are notes in the margin, that is part of the work. For some researchers, the medium (material) of the paper itself is significant and could be useful. It can help establish age, printing process used, provenance and a variety of other aspects.

If you just care about the words printed on the page, it's much easier. But that's only part of the picture. (pun intended)

madengr11 lat temu

I'm surprised the retrieval lift is not automated; probably has something to do with unions.

ars11 lat temu

Perhaps.

It could also be cost. Automated machines are expensive and only worth it if there is a high volume. Perhaps there's just not enough volume to make it worth it.