Pat mentioned yesterday that JISC had published the report on usage of semantic tools in UK education. I obviously wanted to read this report, and with the help of Pat downloaded it from the JISC repository.
Here I’ll only focus on a few observations I made while reading it. First of all I must say that, as a researcher with interests in this field I can’t say how happy I am that a report like this is being made. It is an area of great importance, and often it seems to be forgotten, because it is a highly technical area, which many people have problems understanding. Additionally I do agree, that there should be a focus and indeed a timeline for getting semantic tools into education, because everybody would benefit from their usage.
Once that is said, I must say that I was reading some of the paragraphs with astonishment, and I just feel that I have to air some of the frustrating aspects of the report.
First of all the report introduces two levels of semantic technologies, soft and hard. Where soft is technologies like topic maps and Web 2.0 applications, and hard ones are technologies based on RDF or similar formats. This is a levelling which I can only object to. RDF does not constitute any hard semantic technologies, for instance it is the core technology behind RSS-feed, which is used for news delivery. RDF can be used almost as badly as XML and XML-schemas without linking to semantic relations, if the user chooses to do so. RDF needs to be utilised using RDFS, OWL, DAML or a similar ontology basis before it constitutes any hard semantics.
Soft semantic technologies on the other hand ought to be meaningless in the scope of the report. The author points out that semantic technologies ought to be machine processable. However I believe they are left in the report to aid the readers with understanding the path the semantic technologies are on. The problem then is that the definition include web2.0 tools, which hardly is a precise definition, as I’ve seen diverse technologies as blogs, wikis, flickr, youtube, google, in such definitions. Web2.0 is all about making content creation and sharing easy for the user, and then there might be semantics coming out of this process (such as tag clouds and FOAF relationships). It would have been much better if the report had explored some of these semantic relationships and how they can become hard semantics later on. This was only briefly touched upon.
The survey of semantic tools is an interesting read, which I enjoyed, as I did find a few tools I had not stumbled upon before, however it also seemed to lack a few. I do appreciate that a survey can’t include all tools, but there isn’t any mention of tools coming from Manchester University and/or Ian Horrocks (now Oxford) who definitely is one of the most active in this area, and have many different tools, which have and are being used in educational settings.
My last issue (at least for this blog post) with the report is the roadmap. As I said earlier, this is something I believe is needed, but I find it a bit naive. How are educational institutions and lecturers going to be encouraged to provide metadata and linked data? Where are the ontologies going to come from, especially in the timeframe mentioned in the report?
Ontology creation can be a long process, especially if they are agreed upon, and if they aren’t then the tools that use idiosyncratic ontologies are harder to make work, especially from a performance perspective. For lecturers to start providing metadata really needs a revolution, especially on the user interfaces of semantic tools, which usually can be perceived as a bit geeky. Here I can only say that I like what I read, however the timeframe is a bit optimistic, and will need a lot of funding, which I obviously will be ready to apply for.
