On 4/6/11 10:27 AM, glenn mcdonald wrote:
Are you not able use the public instance for intelligent faceting
across the massive datasets that it hosts?
I think it's fair to say "Yes, I am not able to use the public
instance for anything I would consider 'intelligent faceting' of the
dbpedia dataset".
You can, but you refuse to see how.
As I said before, I don't mean this as a criticism of the technical
infrastructure of Virtuoso.
I am demonstrating and talking about what Virtuoso infrastructure enables.
But I do mean it as a criticism of dbpedia as a specific dataset, of
your data-explorer UI, of the granularity of RDF and the
expressiveness of SPARQL, and of the premise of trying to build one
big unified database of all the world's data.
First off, I've discuss a lot of these matters with Jeff, and
guess what, correct numbers isn't the point at all.
Nonsense. See http://jeffjonas.typepad.com/IRAHSS_Expert_Counting.pdf,
Again, I've discussed these matters with Jeff and this is not about
perfect numbers.
which summarizes itself like this: "This article suggests that the
single most fundamental capability required to make a sensemaking
system is the system’s ability to recognise when multiple references
to the same entity (often from different source systems) are in fact
the same entity." dbpedia as a dataset fails this test badly.
And how on earth does that have anything to do with Counting?
That's a comments about how you figure out that one or more Identifiers
share a common Referent.
Again, I actually not only talked to Jeff about this, I've actually
demonstrated the wonderment of OWL to him via this particular instance
in relation to these matters.
The only difference here is that Jeff doesn't approach these matters via
OWL and RDF. Naturally, since he started some of his work pre. Semantic Web.
Not sure what you mean by "exhibit" here. Your queries timeout,
so unless the needle happens to be in the first page of the
haystack, you're not going to find it.
No they don't and that's where we just will not connect. You've
already seen our browser pages that do just that, and your next
response will ultimately take us back to arguing about page
aesthetics.
Sorry, I can't follow this response. By "no they don't" do you mean
that your queries /don't/ timeout? They certainly do when I try them.
You can actually issue SPARQL with timeouts. Do you not remember the
conversation about partial aggregates in ad-hoc queries using SPARQL or
SQL? That's what I am talking about. What we call "Anytime Query" [1] as
a critical technique for ad-hoc queries at infinite scale [2].
Exhibit #2 -- how do we leverage faceted exploration and
navigation of massive data sets at Web Scale?
I thought I knew what "faceted exploration" meant, but your
"facet" example has nothing I recognize as a facet, so I'm not
sure what your claim is here.
What are you talking about? Using group aggregates to pivot data
across various dimensions is about what?
The link you said demonstrated faceting was this:
http://lod.openlinksw.com/fct/facet.vsp?sid=35044&cmd=refresh
<http://lod.openlinksw.com/fct/facet.vsp?sid=35044&cmd=refresh>
In this view I see a filter in effect on dbpedia-owl:HistoricPlace,
but "facet" usually means a little parallel list of counts along some
dimension, not just a filter. E.g., the "narrow your results" sidebar
here:
http://www.bestbuy.com/site/Digital-Cameras/Digital-SLR-Cameras/abcat0401005.c?id=abcat0401005
Instead of arguing, can you simply respond with a link to an
example of an endpoint that provides access to a massive data
corpus re. declarative queries.
See, this response just makes it seem like you didn't read my note for
comprehension.
What can I say to you, you just won't accept the point.
RPI have published RDF datasets. We loaded them. That's it.
Exactly. You loaded an artificially bloated dataset and then bragged
about its size.
I loaded a bloated dataset and bragged about the size. Hmm. Anyway, you
are 100% percent entitled to your opinion. I am not going to expend any
energy on your opinions as you refuse to work within any kind of context.
The folks on this mailing list understand why RPIs dataset is loaded to
the LOD cloud and what it means. I can't burn any cycles explaining that
to you, your comments are 100% self explanatory re. context infidelity.
Simple examples queries that you can perform against the LOD Cloud
Cache [1] that leverage faceted navigation and scrollable cursors:
1. Find all Entities associated with the Pattern: "New York" --
from the results page use Types or other Attribute filters to seek
your particular disambiguated needle in this massive haystack
I go to the link you provided (lod.openlinksw.com
<http://lod.openlinksw.com>), I type "New York" in the box and hit
Enter. The query times out with no results.
"Retry" means: you have a partial result set, if you can't pivot from
here, try again. DBMS basics for result handling when data is
partitioned horizontally. Our tweak is the support for partial
aggregates in this context.
2. Repeat the above with owl:sameAs inference context enabled
3. Repeat the above with owl:sameAs + a fuzzy InverseFunctional
property rule e.g. using foaf:name
Moot given that #1 produces no results, but since you already said
that you have inferrencing turned off on this public instance, how did
you expect me to do these?
Hit the "Retry" button, take a deep breadth, then try to fix your
context infidelity problem :-)
--
Regards,
Kingsley Idehen
President& CEO
OpenLink Software
Web: http://www.openlinksw.com
Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca: kidehen
