(T.1) Question: Do you provide Linked Open Data in RDF format?
Answer: Yes, we
(T.2) Question: RDF is about resource descriptions. What’s about resources?
Answer: The SymbolicData RDF store links both to
All that is available for download not only from our web site but also can be cloned form our github master repo. A DCAT catalogue description is on the way.
(T.3) Question: How do you link resources and resource descriptions?
Answer: Through fingerprints, a concept explained in more detail in our paper in Computeralgebra-Rundbrief 55 (October 2014)
(T.4) Question: How to set up a Local Sparql Endpoint?
Answer: Look at our wiki page LocalSparqlEndpoint.
(R.1) Question: Who is the team behind SymbolicData and what are your resources?
Answer: SymbolicData grew up in 1998 in a situation where heavily founded projects finished and people formerly not involved with the projects took efforts to secure part of the research infrastructure developed during the project. It was, is and remains (intentionally) a volunteer’s project without funding and lives on the resources that these (changing over the time) volunteers are able to sideline for SymbolicData.
(R.2) Question: You have no clear focus in your papers, resubmit material.
Answer: Papers about SymbolicData are not research papers but report about aspects and design decisions for Research Infrastructure. Hence they are “different” and, in some sense, even not academic. Since the topic of SymbolicData is clearly interdisciplinary we try to present the results to different communities, there is necessarily argumentational overlap.
SymbolicData follows a very Open Publication strategy - you find pdf versions of all papers under submission, accepted, rejected, and published in our wiki and selected reviewer reports in our git Repo.
(R.3) Question: What motivated you to publish your data as linked data.
Answer: All world is talking about Linked Open Data and its enormously growing importance during the last 10 years. There are, of course, also more substantial arguments …
(R.4) Question: How sustainable is your project? (asked by Stephen Watt at CICM 2014)
Background: We supplied to CICM-14 a paper that was not accepted for presentation in the projects slot but “downgraded” to a “Work in progress report”. Stephen Watt asked that question during the discussion of our talk. We supplied another paper to CICM-16 with the same result but decided not to present the project another time as “Work in progress”.
Answer: The project started in 1998, hence survived much longer as many other well staffed projects. During that time we had “dry periods” and learned to “survive in the desert” until the next rain (i.e., project money). The project is as long alive as the last person works on the last bits of the project.
(R.5) Question: What is the motivation for your choice of vocabulary?
Answer: Not every ontology on the internet has equal influence and it is hard to decide about their “weights”. We use foaf, dct, org, but also an agile approach that proved to be very fruitful within AKSW many times. This means, start collect RDF sentences (4 stars in the Tim Berners-Lee program) and look for the fifth star in a redesign process.
Our reference for potentially useful vocabularies is the standard reference http://lov.okfn.org of the Open Knowledge Foundation.
(R.6) Question: How do you describe mathematical software packages?
Answer: We decided to maintain within SymbolicData only a tall interface to descriptions of mathematical software packages with references to swmath. So all questions about ontologies should be discussed directly with the swmath people.
(R.2) Question: How information about conference announcements is collected?
Answer: This data was populated neither “manually” nor “crawled” but tool-based from several mailing lists to ensure quality of the data. This is a classical data architect’s approach.
We stopped that part of the project, since running it over two years we provided a “proof of concept”, but have no resources to perpetuate that without support from the communities. For the communities it seems to be enough to get informed along their mailing lists and to collect casual overviews of upcoming conferences at many not interrelated web pages.
(R.7) Question: Some of your arguments sound pretty standard, is this innovative?
Answer: Yes, this is very innovative, since many semantic web projects don’t use and even don’t recognize existing standards. Using standards reduces efforts and thus enlarges the scope of work that could be done with given resources.
(R.8) Question: You claim, that “practical use of data from a special field requires a minimum of semantic-awareness of the user itself”. Indeed?
Answer: Example: To use Wreath Product decomposition information from Transitive Groups you have to know what a Wreath Product is.
(R.9) Question: Some of the data is mentioned as not up to date - what are the steps that are being taken to keep the info fresh? This is crucial in the design.
Answer: We provide all data “as is”. To keep data up to date requires engagement of volunteers.
Take, e.g., the book list compiled by the German Fachgruppe for many years, that we translated to rdf. It were a good base to compile a common book list as a joined community effort but this requires volunteers to update that information.
The German Fachgruppe lost interest in that topic for several years and later on decided to start a new book list project neither taking notice of the old one nor our proposal, so both are orphaned (but yet available on the internet in a very structured way). The continuation of many undertakings depends on the priorities of (changing) board members.