On Tue, Oct 27, 2009 at 21:22, Bernhard Eversberg <ev_at_biblio.tu-bs.de> wrote:
> I'm with you as soon as you demonstrate it.
Demonstrate that the Web isn't records based? Surely you jest. :)
> (Our systems are not that
> simple as you seem to think, we do have linked records and hierarchic
> arrangements and so on.)
I worked as a senior developer at a national library for several
years. I do not think your systems simple. In fact, most of my ranting
here is because your systems are inherently complex, and needlessly
so. And so on.
...
> That model would still be old-fashioned though, according to your view
> of things,
Uh, I think you're jumping rather well ahead of things now, verging on
speculation, and not to mention that you don't have any good idea of
what "my view" is. :)
> because GBS is still record-based.
How do you know? I suspect your view is that anything that has an ID
in some sense it must be an ID for a record of sorts, and because a
book in GBS has an ID therefore it must have a record. Um, no. No, not
at all. There doesn't have to be any records as such. There doesn't
have to be a single RDBMS transaction involved. Check out tuple-stores
and cumulative key-value property pair stores, and clustering and
concurrency. The two latter (clustering and concurrency) is one of the
big reasons why people are urging semantic key-value stores as it
brings the task of doing and maintaining vast amounts of data like
this into the realm of functional programming, making optimization a
mathematical task rather than a traditional RDBMS hell. We're not
doing this stuff because it's new and fun and cool, and everybody is
talking about it, we're drinking the kool-aid and getting high on the
semantic buzz. We're *also* doing it because it's the next natural
step to take, and because it's bloody necessary and vital to the
future of all storage and computing, no less.
Alex
--
Project Wrangler, SOA, Information Alchemist, UX, RESTafarian, Topic Maps
--- http://shelter.nu/blog/ ----------------------------------------------
------------------ http://www.google.com/profiles/alexander.johannesen ---
Received on Tue Oct 27 2009 - 06:41:47 EDT