[Date Index]
[Thread Index]

[Date Prev] bullet [Date Next] bullet [Thread Prev] bullet [Thread Next]
bullet
Re: idea for digital libraries
bullet bullet bullet bullet

Interesting idea. A few thoughts: - the RFP talked about digitizing a significant new corpora. What would we do here? I don't think FTP will cut it. I think it has to have sex appeal, like your EDGAR database or a global schoolhouse or something - I'm not sure how setting this thing up will help create a global common space. It sounds like maybe you want to set up a professionaly administered and programmed repository for e-publication? - it seems to me the primary appeal of what you suggest is something like a production Dartnet, i.e., a piece of infrastructure that provides value because it has dedicated function that doesn't get thrashed by all the random activity of the rest of the Internet (and doesn't itself thrash the Internet). If that's the case, the reviewers might kill it by saying in effect, "let the commercial guys build fast links, that's what they plan to do anyway". A different angle might be to claim that this thing is a great boost to collaboration technologies (mbone is nearing its limits now), but that seems to apply to NSF's previous "collaboratory" vision more than to digital libraries

Some smaller technical nits: - I'm not convinced you need a high speed network to ease the load from popular read-mostly objects like FTP files. End node caching should work ok for that - what does it mean to have rules on how to check things in and out of the library? Are you talking about locking documents for update?

I don't mean to shoot down your idea. Just to stir the pot, so we can see what comes of it. - Mike

bullet
[Date Prev] bullet [Date Next] bullet [Thread Prev] bullet [Thread Next]
bullet