Andraz Tori wrote:
On Mon, 2009-03-02 at 01:54 +0100, Giovanni Tummarello wrote:
Hi Andreaz :-)
I don't see the difference between the LOD model and the data
(including
links) itself. At least to us at Zemanta it is immensely
helpful to have
a lot of those links done. It brings down the cost of doing
really
innovative stuff to us and I believe to many others too.
We don't dereference them in real-time, but ahead-of-time to
produce
specialized datasets, but I don't think that makes a
difference.
to the risk of being pedantic..
Well the idea from the day 0 of the semantic web was that entity would
be interlinked by the reuse of the same URI. "linking" or putting a
"same as" is the same thing so nothing new.
the new part of LOD would be the publishing of a dataset as many
individual RDF description corresponding to the resolution of each
URI/URL. and that part is not being useful to you
I wasn't here at its conception, so I can only talk about current
situation. Currently this is only a part of a bargain, but not the whole
thing. The other part is actually having some concrete set of federated
generally-interesting datasets that are interconnected instead of
abstract vision or one-off efforts.
So LOD is actually people working 'together' to get something practical
out of larger SW idea.
I also think that LOD has started to cause the Network effect. Every new
dataset makes others more usable.
> That the bubbles continue to grown is however a sociological
> interesting phenomen :-)
And a good sign that something has gone right :)
Maybe :-) but people do things for many other reason that "they're
right". An alternative explanation i like is
http://inamidst.com/whits/2008/technobunkum
I disagree LOD is this kind of a beast. Services like ours are proving
LOD is important for 'mainstream'. While it is a stretch to say our
service is (already) mainstream or that LOD is the main enabler, it is
definitely meant for mainstream and LOD plays a certain part in making
it achievable (cheaply enough).
I think LOD achievement is enormous and this is only the
start.
:)
Said picture might have helped to get a lot of RDF data online. This
is undoubtly a great achievement
But sustenability and real growth comes only if we can prove real
reasons for people to publish this data, and in this way. While we
havent yet seen this, this doesnt mean that some application might not
exist
we're on exactly the same page here! Sustainability and growth depend on
working incentives for publishers to publish the data. And there is just
one thing (commercial) publishers care about - direct and indirect
traffic.
If a good incentive is getting more links from bloggers (which in turn
brings traffic and increases different rankings), there you have one in
Zemanta. Ok, actually it is only a potential, since we include datasets
on one-by-one basis. But when a company approaches us to incorporate
their links into our suggestion pool the first question we ask is: Do
you have proper connections into LOD? It makes everything so much
easier.
To date bloggers have created more than a million permanent hyperlinks
with Zemanta. For some links the ability to suggest them came from the
fact that the LOD data and links were available.
I am sure there must be some other LOD related services that in the 'end
consequence' bring traffic to the publishers. We are working on at least
another one - "Simple Semantic Tagging".
bye
andraz
Andraz,
I agree with your points and general sentiment.
One issue we need to address is the LOD cloud. Personally, there are
vital aspects of the big picture that is completely misses.
1. UMBEL - this is the data dictionary aspect of the Linked Data Web and
it meshes disparate ontologies (one aspect) and also provides a concept
scheme (its other aspect)
2. Dynamic Linked Data - our sponger cartridges perform lookups and
joins against other data providers (as you can see re. Zemanta cartridge) .
When you combine the main cloud, #1, and #2, you basically moot any
questions about the state of the "Linked Data Web". Instead, we move
over to more interesting and important issues such as quality of data
within the "Linked Data Web".
Quality is valuable and an opportunity for any "Linked Data Web" player
to innovate.
Show me an entrepreneur and I will show you someone who is knowingly or
unknowingly using DBMS technology to provide unique value to his/her
customer base via queries, lookups, and joins :-)
Kingsley
Giovanni
--
Regards,
Kingsley Idehen Weblog: http://www.openlinksw.com/blog/~kidehen
President & CEO
OpenLink Software Web: http://www.openlinksw.com
