Friday, May 8, 2009

The Next Age of Discovery

The Wall Street Journal today published an article by one Alexandra Alter on the preservation of manuscripts, digitally: Writ large in the article is Fr. Columba Stewart, head of the Hill Manuscript Library in MN.

1 comment:

Jonathan Jarrett said...

There's some really interesting stuff there, but the writer was given some fairly partial information I think. The scanning of papyrus rolls without unrolling them is already ongoing: I saw a presentation about it last Leeds. The Codex Sinaiticus is in some sense part of St Catherine's Sinai's collection, obviously, but it is kept at and was digitised by the British Library; I have a post coming that mentions this and provides links. But those points aside, this is a really interesting piece littered with splendid examples. Thanks for linking to it.

The problems it raises about people only knowing about what's on the web are a live issue, but I don't think it's as much of a problem as they suggest. That is, I don't think it's a backward step: people have always not known about obscure stuff. What is happening is that some of that stuff is being given a big profile boost; it isn't making the other stuff recede except in relative terms. As long as we can continue to get across the idea that not everything is on the web nor e'er will be, I think that's OK.

What does give me pause is this idea that digitisation creates "a disaster-proof record". I mean, that depends on your disaster! Yes, a digital record of everything that was in Cologne would have been really nice to have just now; but equally, digital storage is so tied to file formats and hardware technology that anyone who thinks it's a permanent record is fooling themselves. Can you still read documents you wrote on your Commodore or in Works or Write? Favoured archival picture format at the moment is TIFF, but not many image viewers on a regular PC or many web browsers can view those, and they certainly can't deal with the possibility TIFF files (not redundant, it stands for `tagged image file format') have for multiple layers or tagged metadata. In fact, many image viewers just wipe that information when loading the file. Bye-bye archival data! So though it's very important, creating a digital record only opens the way to a long cycle of periodic updating and resaving for accessibility. And then there's hardware: how long do hard-disks last for? Was it backed up? To a system that a power failure won't frag? To tape? How long does back-up tape last? Are your ten-year-old CD-Rs still playing right? Will you still be able to read that DVD-R when the format wars stop? and so on.

For reasons like this, actually one step that's really important with a digital record, especially of a text or an image, is, paradoxically, to print it out on archival-quality paper and store it somewhere dry and Ph-neutral. Then you have a back-up for when the power fails and the only machine with a CD drive is in a museum and won't boot...