Offering large downloads of RDF is really easy. You just need a web server. But then all of the work to do queries is performed client-side, and a lot of
unneeded data flows to the client. Offering a SPARQL endpoint (especially a reliable, performant one) is much harder. But all the client has to do in that case
is fire queries, and only the results come back across the network.
Linked Data Fragments aims to offer something in between. The server does some, but not all of the work for complex queries, hence the restricted forms. The
client does some, but not all of the work. And the hope is that network traffic is decreased greatly without all the cost being laid on the server. Whether or
not that happens in practice has a lot (like always) to do with specifics. For example, rather than download a LDF result and then query it, it might be
streamed into a query engine and joined with other streamed LDF results, but only if the form of the query permits that. The argument is not that LDF is
inherently better or more efficient, but that it can be more efficient often enough to be a useful alternative. That's the nature of this kind of engineering
using networked information. It's full of trades. Rarely can you find an unambiguously universally superior technique.
The folks behind LDF put a number of posters and presentations into ISWC 2016,
so you might want to Google for those publications.
---
A. Soroka
Laura Morales wrote on 5/6/17 9:31 AM:
As far as I understand them, clients are supposed to submit only simple queries to servers in order
to retrieve subsets of the data. Queries like "?subject <ns:property> ?object", and
then download this snapshot (subset) of the data; in this case the list of triples for the propery
<ns:property>. The data is downloaded locally, and then the client can issue SPARQL queries on
the local copy of the data just downloaded.
Is this correct? Is this how "Linked Data Fragments" work? Doesn't this
generate a lot of traffic, a lot of data moving around, and very little improvement over
using a local SPARQL endpoint?
I really can't understand the benefits. Haaaalp!
For reference: http://linkeddatafragments.org
