Is fuseki a simple local sparql endpoint to start with? My instinct is
cwm, phpld, and gold are for the more experienced. Maybe not cwm....I
just want to sparql query data.
I will try Fuseki first:
I will start by Installing Java.
http://shinephp.com/install-jdk-7-on-ubuntu/
Then I will follow the tutorial, by first installing Jena and Fuseki.
http://jena.apache.org/tutorials/index.html
I started looking here:
https://jena.apache.org/tutorials/sparql_data.html
I have a fuseki server running at: http://localhost:3030/fuseki.html
I wonder if it will work. Thus far it has proved to be rewarding.
Tuesday, September 30, 2014
Monday, September 29, 2014
Chat on #swig for September 28, 2014 - Melvin's Discussion of Bitmark
It all started when Melvin Carvalho was using a hubot called klaranet on IRC which was awarding kudos to those in the room. It was another day on Freenode's #swig channel. He talked to me about his excitement about bitmark, a new form of crypto-currency, which is used by klaranet.
Bitmark's aim is to integrate crypto-currencies and RESTful principles with 5-star linked data. In its implementation, it uses URIs, which can be integrated with Payswarm, WebCredits (see also wiki), and Bitcoin. Bitmarks can be sent from URIs to other URIs, with the record of their movement stored as RDF. These URIs can then be used in IRC with klaranet. Bitmarks can also be sent to IRC nicknames (supposedly identifiers as Melvin described them), which can be authorized through registration with the network (at least on Freenode). In addition, Melvin believes that personas, which can be URIs describing people, will be tied together with owl:SameAs. From the discussion, it seems unclear if IRC nicknames can also refer to URIs.
In short he says the key of what he said was, "Apply the principals of Web Architecture and Fieldings REST to cryptographic currency, in order to become truly decentralized and scale without limits"
The Bitmark project started only recently, as indicated by the Bitcoin talk forum, and so far it has been doing quite well. A recent interview about Bitmark is also available at Bitcoinist.
Bitmark's aim is to integrate crypto-currencies and RESTful principles with 5-star linked data. In its implementation, it uses URIs, which can be integrated with Payswarm, WebCredits (see also wiki), and Bitcoin. Bitmarks can be sent from URIs to other URIs, with the record of their movement stored as RDF. These URIs can then be used in IRC with klaranet. Bitmarks can also be sent to IRC nicknames (supposedly identifiers as Melvin described them), which can be authorized through registration with the network (at least on Freenode). In addition, Melvin believes that personas, which can be URIs describing people, will be tied together with owl:SameAs. From the discussion, it seems unclear if IRC nicknames can also refer to URIs.
In short he says the key of what he said was, "Apply the principals of Web Architecture and Fieldings REST to cryptographic currency, in order to become truly decentralized and scale without limits"
The Bitmark project started only recently, as indicated by the Bitcoin talk forum, and so far it has been doing quite well. A recent interview about Bitmark is also available at Bitcoinist.
Sunday, September 28, 2014
Things to do (September 28, 2014)
To do (concerning views in http://bshambaugh.org/eispp3.pdf)
(1)
On page 16 there are OPM Structural Links:
http://www.er.byu.edu/er2003/slides/ER2003PT1Dori.pdf
These will need to be considered and added so the REA and partOf relations to look something like the example on pages 92 to 93.
(2)
When reading James Dylan Hollenbach's thesis (http://dig.csail.mit.edu/2010/rdf-widgets/thesis.pdf), he mentions that Exhibit (http://people.csail.mit.edu/dfhuynh/research/papers/www2007-exhibit.pdf) is great for new developers, but is not very customizable. This led to search for other sources about Linked Data
(http://people.csail.mit.edu/dfhuynh/research/papers/www2007-exhibit.pdf).
To do (concerning Provenance in http://bshambaugh.org/eispp3.pdf)
(1) rawbase (R&WBase)
I read the R&WBase paper:
http://ceur-ws.org/Vol-996/papers/ldow2013-paper-01.pdf
But I still have yet to check out any code on their project site. Eventually
http://rawbase.github.io/
To do (concerning conversation with Markus Sabadello):
(1)
See if you can implement the NRP Logging your work tutorial as linked data
https://docs.google.com/presentation/d/1YxPuis1_jbcSxUTqntd4ltBLg9IZP2oscE0Kovd15vo/edit#slide=id.p
(2)
Look at the Linked Data Platform, and see if there are actual applications that you can look at? http://www.w3.org/TR/ldp/
Can you run them on a FreedomBox (with each person's triples on a separate box)
edit: okay I found some implementations from the dataversity video below:
http://www.w3.org/wiki/LDP_Implementations
Look at introduction the Linked Data Platform:
http://video.dataversity.net/video/the-w3c-linked-data-platform-ldp/
edit: mentions http://open-services.net/
Comment on the list perhaps:
http://lists.w3.org/Archives/Public/public-ldp-comments/
(1)
On page 16 there are OPM Structural Links:
http://www.er.byu.edu/er2003/slides/ER2003PT1Dori.pdf
These will need to be considered and added so the REA and partOf relations to look something like the example on pages 92 to 93.
(2)
When reading James Dylan Hollenbach's thesis (http://dig.csail.mit.edu/2010/rdf-widgets/thesis.pdf), he mentions that Exhibit (http://people.csail.mit.edu/dfhuynh/research/papers/www2007-exhibit.pdf) is great for new developers, but is not very customizable. This led to search for other sources about Linked Data
(http://people.csail.mit.edu/dfhuynh/research/papers/www2007-exhibit.pdf).
To do (concerning Provenance in http://bshambaugh.org/eispp3.pdf)
(1) rawbase (R&WBase)
I read the R&WBase paper:
http://ceur-ws.org/Vol-996/papers/ldow2013-paper-01.pdf
But I still have yet to check out any code on their project site. Eventually
http://rawbase.github.io/
To do (concerning conversation with Markus Sabadello):
(1)
See if you can implement the NRP Logging your work tutorial as linked data
https://docs.google.com/presentation/d/1YxPuis1_jbcSxUTqntd4ltBLg9IZP2oscE0Kovd15vo/edit#slide=id.p
(2)
Look at the Linked Data Platform, and see if there are actual applications that you can look at? http://www.w3.org/TR/ldp/
Can you run them on a FreedomBox (with each person's triples on a separate box)
edit: okay I found some implementations from the dataversity video below:
http://www.w3.org/wiki/LDP_Implementations
Look at introduction the Linked Data Platform:
http://video.dataversity.net/video/the-w3c-linked-data-platform-ldp/
edit: mentions http://open-services.net/
Comment on the list perhaps:
http://lists.w3.org/Archives/Public/public-ldp-comments/
Constructing triples for rea and partOf relations
The figure is an attempt to map part-whole relations, the rea ontology, and Wordnet terms to triples using the part-whole and rea model diagrams earlier.
Please note the other urls: http://www.w3.org/2006/03/wn/wn20/instances/word-automobile.rdf (wn20instances:word-automobile), http://www.w3.org/2006/03/wn/wn20/instances/word-item.rdf (wn20instances:word-item)
I would like to expand the word item with lexical file number 06, frequency 16, offset 03593583 and definition "a whole or individual unit; especially when included in a list or collection". (from: wordnet.princeton.edu)
However, I am not sure how to query the WordNet data on the w3c site. The website http://w3.org/TR/wordnet-rdf/#queries gives some example queries. I will have to figure out how to load these into a triple store and query them. Perhaps I will use Neo4j (http://www.neo4j.org/) and set up a sparql endpoint with this. (or just figure out how to run a sparql query of the one at w3.org). Later, I would like to use a tuplestore such as Linda.
Earlier I thought about using regular expressions to go through the wn20full.zip [download available at (http://www.w3.org/2006/03/wn./wn20/download/)]. I'd want to use something like grep (http://www.gnu.org/software/grep/manual/grep.html), which may be made easier by looking at the regexone site (http://regexone.com/).
Please note the other urls: http://www.w3.org/2006/03/wn/wn20/instances/word-automobile.rdf (wn20instances:word-automobile), http://www.w3.org/2006/03/wn/wn20/instances/word-item.rdf (wn20instances:word-item)
I would like to expand the word item with lexical file number 06, frequency 16, offset 03593583 and definition "a whole or individual unit; especially when included in a list or collection". (from: wordnet.princeton.edu)
However, I am not sure how to query the WordNet data on the w3c site. The website http://w3.org/TR/wordnet-rdf/#queries gives some example queries. I will have to figure out how to load these into a triple store and query them. Perhaps I will use Neo4j (http://www.neo4j.org/) and set up a sparql endpoint with this. (or just figure out how to run a sparql query of the one at w3.org). Later, I would like to use a tuplestore such as Linda.
Earlier I thought about using regular expressions to go through the wn20full.zip [download available at (http://www.w3.org/2006/03/wn./wn20/download/)]. I'd want to use something like grep (http://www.gnu.org/software/grep/manual/grep.html), which may be made easier by looking at the regexone site (http://regexone.com/).
Wednesday, September 24, 2014
P2P Computing - Architecture
This is meant to illustrate concept that users should be able to choose nodes that will perform peer to peer computing. These nodes may be participants in a project. In reality, this choice may be a bit more difficult, and may depend on what is available at the best price. In addition, this illustration shows a geographical distribution of nodes. In addition, it may be useful to specify HPC variables of the compute architecture.
Tuesday, September 23, 2014
Monday, September 22, 2014
Simple Part Whole Class Diagram (checking with Protege)
I guess I made the mistake of not putting Faults for the wheel, motor, headlight, and Car as subclasses of fault. In addition note that wheel points to wheel_of_car, motor points to motor_of_car, and headlight points to headlight of car.
Simple Part Whole Relations Diagram
A class diagram above, and a property diagram below of simple-part-whole relations v1.3 (with personal extensions guessed from the document).
http://www.w3.org/2001/sw/BestPractices/OEP/SimplePartWhole/simple-part-whole-relations-v1.3.html
Sunday, September 21, 2014
Figuring out REA
Presently I am looking at how to fill the empty box in the top figure with the word contribute. It's meaning defined by WordNet.
http://www.w3.org/TR/wordnet-rdf/
The rea namespace rea: is defined by the diagrams below.