Trough of disillusionment for microblogging and social software?

The IT research firm Gartner has recently published another of its technology reports for 2009: Gartner's Hype Cycle Special Report for 2009. This report is another in a long line of similar Gartner reports which do exactly what they say on the tin. That is, they provide a technology 'hype cycle' for 2009! Did you see that coming?! The technology hype cycle was a topic that Johnny Read recently discussed at an ISG research reading group, so I thought it was worth commenting on.

According to Gartner - who I believe introduced the concept of the technology hype cycle - the expectations of new or emerging technology grows far more quickly that the technology itself. This is obviously problematic since user expectations get inflated only to be deflated later as the true value of the technology slowly becomes recognised. This true value is normally reached when the technology experiences mainstream use (i.e. plateau of productivity). The figure below illustrates the basic principles behind the hype cycle model.

The latest Gartner hype cycle (below) is interesting - and interesting is really as far as you can go with this because it's unclear how the hype cycles are compiled and whether they can be used for forecasting or as a true indicator of technology trends. Nevertheless, according to the hype cycle 2009, microblogging and social networking are on the decent into the trough of disillusionment.

From a purely personal view this is indeed good news as it might mean I don't have to read about Twitter in virtually every technology newspaper, blog and website for much longer, or be exposed to a woeful interview of the Twitter CEO on Newsnight. But I suppose it is easy to anticipate the plateau of productivity for these technologies. Social software has been around for a while now, and my own experiences would suggest that many people are starting to withdraw from it; the novelty has worn off. And remember, it's not just users that perpetuate the hype cycle, those wishing to harness the social graph for directed advertising, marketing, etc. are probably sliding down the trough of disillisionment too as the promise of a captive audience has not been financially fulfilled.

It's worth perusing the Gartner report itself - interesting. The above summary hype cycle figure doesn't seem to be available at the report, so I've linked to the version available at the BBC dot.life blog which also discusses the report.

Extending the FOAF vocabulary for junkets, personal travel and map generation

As we know, FOAF provides a good way of exposing machine-readable data on people, their activities and interests, and the nature of their relationships with other people, groups or things. FOAF allows us to model social networks much in the same way as a social networking service might (e.g. Facebook). The big difference being that with FOAF the resultant social graph is exposed to the Semantic Web in a distributed way for machine processing (and all the goodness that this might entail…); not held in proprietary databases.

FOAF data has generally always been augmented with other RDF vocabularies. Nothing strange in this; this was anticipated, and reusing and remixing vocabularies and RDF data is a key component of the Semantic Web. My FOAF profile, for example, uses numerous additional vocabularies for enrichment, including Dublin Core, the Music Ontology, and the Contact, Relationship and Basic Geo vocabularies. The latter vocabulary (Basic Geo) provides the hook for this blog posting.

The need to provide geographical coordinates and related data in RDF was recognised early in the life of the Semantic Web, and the Basic Geo (WGS84 lat/long) Vocabulary website lists obvious applications for such data. Although including geographical data within FOAF profiles presents an obvious use (e.g. using Basic Geo to provide the latitude and longitude of, say, your office location), few people do it because few applications actually do anything with it. That was until a couple of years ago when Richard Cyganiak (DERI, National University of Ireland) developed an experimental FOAF widget (FOAF – Where Am I?) to determine geographical coordinates using the Google Maps API and then to spit it out in FOAF RDF/XML for inclusion in a FOAF profile. In his words, "there's no more excuses for [not including coordinates]". With coordinates included, FOAF profiles could be mapped using Alexandre Passant's FOAFMap.net widget (also from DERI), which was developed around the same time and extracts geographical data embedded within FOAF profiles and then maps it using Google Maps. Despite the presence of these useful widgets, FOAF profiles rarely contain location data because, let's face it, are we that interested in a precise geographical location of an office?!

More interesting – and perhaps more useful – is to model personal travel within a FOAF profile. This is consistent with the recent emergence (within the past year or so) of 'smart travel' services on the web, the most notable of which is probably Dopplr. Dopplr essentially allows users to create, share and map details of future journeys and travel itineraries with friends, colleagues, business contacts, etc. so that overlaps can be discovered in journey patterns and important meetings arranged between busy persons. It is also consistent with the personal homepages of academics and researchers. For example, Ivan Herman's (W3C Semantic Web Activity Lead) website is one of many which include a section about upcoming trips. There are others too. From personal experience I can confirm that many an international research relationship has been struck by knowing who is going to be at the conference you are attending next week! People also like to record where they have been and why, and the 'Cities I've Visited' Facebook application provides yet another example of wanting to associate travel with a personal profile, albeit within Facebook.

Of course, Dopplr and Facebook applications are all well and good; but we want to expose these journeys and travel itineraries in a distributed and machine processable way - and FOAF profiles are the obvious place to do it. It is possible to use the RDF Calendar vocabulary to model some travel, but it's a little itchy and can't really tell us the purpose of a journey. Other travel ontologies exist, but they are for 'serious' travel applications and too heavy weight for a simple FOAF profile. It therefore occurred to me that there is a need for a light weight RDF travel vocabulary, ideally for use with FOAF, which can better leverage the power of existing vocabularies such as Basic Geo and RDF Calendar. I documented my original thoughts about this on my personal blog, which I use of more technical musings. Enriching a FOAF profile with such data would not only expose it to the Semantic Web and enrich social graphs, but make applications (similar to those described above) possible in an open way.

To this end I have started authoring the Travelogue RDF Vocabulary (TRAVOC). It's a pretty rough and ready approach (c'mon, 3 -4 hours!) and is really just for experimental purposes; but I have published what I have so far. A formal RDF Schema is also available. Most properties entail being a foaf:Person and I have provided brief examples on my blog.

As noted, TRAVOC has been sewn together in a short order. It would therefore benefit from some further consideration, refinements and (maybe) expansion. Perhaps there's a research proposal in it? Thoughts anyone? In particular, I would be interested know if any of the TRAVOC properties overlap with existing vocabularies which I haven't been able to find. If I have time – and if and when I am satisfied with the final vocabulary - I may acquire the necessary PURLs.