04.06.2012 14:45, Ed Summers:
> I don't think Linked Data precludes the notion of a record, or even
> that the record will be made obsolete. Much to the contrary, I think
> Linked Data brings the notion of the record into the spotlight on the
> Web. The reason why I say this is that using Linked Data principles
> means assigning URLs to the things we care to describe (books,
> authors, subjects, etc); and when those URLs are resolved (by your
> browser or what have you) you get back a representation  of that
> resource--which in my mind amounts to what we have historically called
> a "record".
I see, so it will mean that the whole concept of "exchange" will
no longer be about self-contained packages of data and of countless
locally held static replicas of authority data that need to be updated
every so often. Which of course was/is quite burdensome and wasteful
> That being said, I think there is still plenty of work and
> experimentation to do in this area, some of which might even overlap
> with the ResourceSync effort at NISO .
It certainly is a bold vision right now, that much is clear. I just
wonder what the timescale might be for the first steps to become reality
and implementable in affordable systems.
OTOH, all of that does not say much about the reality of RDA data
as they will come along by next year. How much of those concepts will
that stuff actually be able to support? If last year's test data will
be the de-facto standard (scenario 3!), then I just wonder.
The German National Library have just announced they're going to
do scenario 2, btw. (Which is what would emerge quite naturally from
the way we have worked already for a long time.)