On Tue, Oct 13, 2009 at 12:10 PM, Melissa Terras <[log in to unmask]> wrote:

I'd agree with Gabby.

You could do a character count of extant text, to arrive at a figure of size of dataset, but even that is fraught with complexities - the roman alphabet is easily and simply stored as ascii, for example, but when you start to take unicode encoding into play for more complex glyphs, then the resulting digital "text" is much larger.

I'd say that images are essential - but for archaeological records you are also looking at context sheets, drawings, maps, matrixes, etc - its not just a case of measuring the number of characters or the size of files.

I should really know how much information is in the IADB at Silchester, but I dont. I'll ask someone who does!

Melissa

Gabriel Bodard wrote:

I'm not sure I can give an answer in figures, at all, but there are at any rate several ways to answer this question:

(1) text only, as you specify, and only allowing for one, good but slim, copy of each text, basically you would get this figure by adding together the TLG, TLL, DDbDP, PHI inscriptions, Clauss-Slaby inscriptions, and that would probably total a small number of gigabytes. Round it up to factor in the small amount of stuff that is not yet digitized, and it's still a body of data that I could carry on a USB stick in my pocket.

(2) text only, but richly marked up, in a non-terse language such as TEI XML, and with copies of individual manuscript witnesses, apparatus criticus, etc. Perhaps an order of magnitude larger than the above, but I could still fit it all on my laptop.

(3) but you really _need_ images of all this stuff--not hundreds, maybe, but at least one or two good digital photographs or scans of each manuscript page, papyrus, epigraphic support, etc. No one publishes inscriptions or papyri without photographs (especially digitally), and the amount of visual support behind a single project like the Homer Multitext is impressive. I really can't put a figure on this any more, but I'm guessing that we're talking a small number of terabytes instead of gigabytes. (Has anyone tried to make estimates like this for the Scaife Digital Library?)

(4) At this point I'm going to argue that you _can't_ just include texts any more, but also want photographs of artwork, sculpture, ceramics, architecture, archaeological contexts, etc. As you say, the possibilities are limitless, but I'm sure we could still make estimates here. How much data does the Silchester excavation produce in a season?

I'm sure this doesn't help, but I hope it helps to kick off more discussion.

Best,

Gabriel

Paradoxographer a écrit :

** APOLOGIES for CROSS-POSTING **

Hello,

Perhaps the list can help me out with the following question. Has anyone ever calculated how much server / disc storage space (in gigabytes, terabytes, etc) the extant residue of the Graeco-Roman world would occupy? I suppose this would have to be limited to textual (including epigraphic) material, there being potentially no limit to the number of photographs, drawings, plans etc that could be produced ...

I'm curious to know how the volume compares with that of current annually created digital information / publications.

Kind Regards,

Rachel Hardiman.

--

Melissa M. Terras MA MSc DPhil CLTHE CITP FHEA
Senior Lecturer in Electronic Communication
Department of Information Studies
Henry Morley Building
University College London
Gower Street
WC1E 6BT

Tel: 020-7679-7206 (direct), 020-7679-7204 (dept), 020-7383-0557 (fax)
Email: [log in to unmask]
Web: http://www.ucl.ac.uk/infostudies/melissa-terras/
Blog: http://melissaterras.blogspot.com/

General Editor, Digital Humanities Quarterly: http://www.digitalhumanities.org/dhq/

Digital Images for the Information Professional. Available now through all good bookshops, or from Ashgate at http://www.ashgate.com/default.aspx?page=637&calcTitle=1&pageSubject=324&title_id=8986&edition_id=9780