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

40 pointsby benbreenvor 11 Jahren13 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

Pyrodoggvor 11 Jahren

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

gwernvor 11 Jahren

> 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.

fernlyvor 11 Jahren

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.

thretvor 11 Jahren

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.

cjubb39vor 11 Jahren

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

gwernvor 11 Jahren

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

prbuckleyvor 11 Jahren

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.

peteretepvor 11 Jahren

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

insipidvor 11 Jahren

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

Makenvor 11 Jahren

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

caseysoftwarevor 11 Jahren

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)

madengrvor 11 Jahren

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

arsvor 11 Jahren

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.