Shared Approaches to Use of

I commented previously about a number of blogs which have been published by JISC services (and I should have included the CETIS blogs in my summary). I suggested that it would be useful to give some thought to ways in which JISC services (and also JISC-funded projects) could share best practices and explore ways of maximising the impact of their blogging services. This is a topic I’ll return to shortly.

On a related area, looking at the referrer logs for this blog I noticed a couple of visits from a bookmark. Following the link back I found a link from CCuran‘s bookmarks. I then noticed that Randy Metcalfe of OSS Watch was included on CCuran’s network. I work closely with Randy (we gave a joint talk on What Does Openness Mean to the Web Manager? at last year’s Institutional Web Management Workshop, and we’ve recently been working on a joint paper on openness)  so I suspected his bookmarked resources would overlap with some of my areas of interest. One of his tags Randy used to bookmark resources which are also of interest to me is “communications_strategy“. His page for this tag is shown below. page showing 'communications_strategy' tag

Myself and colleagues at UKOLN are looking at further developing our communications and marketing strategy – so Randy’s resources should prove useful to us.

The more general issue of sharing of bookmarks, using services such as, is an area I would like to develop further.  A few months ago I  realised that it would be useful for me to  bookmark  details of  venues I’ve used for events (or venues I have attended) and also good hotels I’ve stayed at.  So I’ve created a tag for recommended hotels (to remind me of the great, privately-run hotel in Edinburgh with free WiFi in bedrooms) and similarly for recommended venues.

With the JISC community many of us have an interest in finding good venues for events – and good hotels to stay in.  So rather than keeping a private copy of such information,  I think that would be a good, simple way of sharing such information.  There will obviously be a need to think about some of the limitations (e.g. the subjectivity of such preferences and concerns that one may be sued if negative comments are given) – but I think the reservations are minor and out-weighed by the benefits.

We’ll need to agree on the tags, I think.  But is this an idea worth pursuing?


  1. Brian-

    How about checking out, a ‘semantic’ review site?

    I’m suspect Tom Heath will pick up on the revyu link – and maybe he’ll give a rather more compelling argument (in fact, an argument: period) for why you should use it?

    Belated HNY, btw :-)

  2. I’m sure Tom would also be one of the first to draw your attention to not only the incredibly alpha status or revyu, but also the fact that it may not still be around in a few years. It is his thesis project isn’t it?

    In a final point, if there’s some JISC thing about finding venues and hotels etc., that sounds like a Google Maps project just waiting to happen doesn’t it?

  3. Before a more complicated (albeit more useful) tool gets set up, I’d be happy to tag venues that the RIN would recommend on my delicious account. We just need a unique tag..

  4. Brian, Hi, and great idea to collate this sort of stuff somewhere public; these are resources we will all find useful, and the more critical the mass the better. I’d second Tony’s comments that Revyu would be an excellent place to record these.

    By way of argument in favour, there are many advantages to doing it this way:

    1) Revyu allows you to “Review Anything”, meaning you are not constrained by someone else’s categories. This may make it more suitable for reviewing something less commonly featured on conventional reviewing sites, such as Conference Venues. Reviewed “things” can be tagged to provide a way of grouping related items.

    2) reviews and ratings given on Revyu are automatically exposed in machine-readable RDF/XML. Loading reviews and ratings into the Notes field on would make this data less accessible to other services wishing to reuse it.

    3) Revyu provides many ways in which third parties can syndicate their own reviews, or reviews of things associated with a particular tag. This can be done via RSS feeds, a little Javascript widget, and a SPARQL endpoint whereby people can write custom queries against the data held in Revyu. See the right hand bar of my homepage for an example that uses the JavaScript approach. See drewp' home page for an example that does custom SPARQL queries in Python.

    4) lastly for now, but by no means the final advantage, is that reviewing something on Revyu results in a URI being minted for that thing (and the review, and the reviewer). For the vast majority of things in the world that do not currently have URIs, this provides a vital hook for integrating data on a Semantic Web.

    I’d be delighted to talk more offline about how you might use Revyu to help this effort, and how the partner services I’m currently developing may address your concerns about subjectivity in recommendations.



  5. [responding separately to Phil’s comments…]

    Hey Phil,

    I think you’re doing the site a major dis-service. Sure, it hasn’t got the full complement of bells and whistles yet, but it’s developing at quite a rate, and everything has to start somewhere, right? Most importantly the back-end infrastructure is in place and has been for some time. User experience enhancements can be (and are being) added incrementally all the time.

    Is there such a thing as “incredibly alpha”? Labelling Revyu as “alpha” is our way of managing people’s expectations of the site, and helping to highlight the fact that feedback/feature requests/bug reports are all welcomed. I’d love to hear your constructive comments. You know where to find me; do get in touch :)

    As to the subject of longevity, come on Phil, this is the web!! Who is to say that Epinions, Ciao, or any of the other reviewing sites (even Amazon) will be around in a few years? Back in the mid-90s would you have said to people “don’t bother to submit your new web site to AltaVista, it won’t be around or worth using in a few years”?

    You’re right, this is part of my PhD research. Sure, that period of studying will certainly finish, but there’s no reason the site will not continue long into the future. Whether it does or not is partly in your hands; add some reviews, spread the word, syndicate the content back to your blog, build some Semantic Web mashups using the reviews, whatever. All these will help make the site a success and help move us closer to a Semantic Web.

    However, whether it lives for another one year or another one hundred years is totally immaterial. The point about exposing the reviews in RDF/XML according to the Review Vocabulary is that you can export the data, archive it, import it into other services, whatever you like. If Brian was concerned about the longevity of the data, he could take dumps of the RSS feeds of his reviews, or better still run regular SPARQL queries to collect and archive them. The key point is that the data is machine-accessible, machine-processable, and future-proofed. This is one of the beauties of the Semantic Web approach; but then you don’t need me to tell you that 😉


  6. Hi Tony, Phil and Tom
    Thanks for the comments.
    Phil’s point that Revyu is part of a PhD research project reminds me that many moons ago I attended a WWW conference in which a researcher at Stanford University described his research work. His project (the speaker was either Brin or Page) was called Google :-) So I don’t thing there’s necessarily a problem with using it.
    In terms of the service itself, I’ve registered and installed the bookmarklet. I don’t like the way the bookmarklet replaces the current page – the bookmarklet I use spawns a new window.
    Regarding the sustainability of the service itself and getting people to use it (outside of a small techie community) I would suggest toning down the rhetoric of the technology. Advising users to run regular SPARQL queries will turn many people off, I suspect!
    Another issue is the whether people who might be willing to use services like this will want another service to use. From my point of view, I tend to use for a variety of purposes. Social bookmarking services clearly get better the more users there are, so a well-established service will have advantages over new services, even if the newer services have technical advantages. If that is true generally, might it not be woirth looking at ways in which you can leverage the service? Perhaps a Greasemonkey scripot which allows me to check a box saying add to Revyu when I add a resource to, for example.

  7. Sorry Tom, I just thought it was fair to put things in perspective and your defence certainly helps to round out that view.

  8. “We’ll need to agree on the tags, I think.”

    are you sure?

  9. Hi Phil
    When you ask if I’m sure that we need to agree on the tags, I assume that you are referring to the debate that tags should not be imposed, but should reflect an individual’s preferences?
    If so, I disagree. Folksonomies should be lightweight and avoid the need for centralised agreement. However if a small group of consulting adults can agree to use a tag, this this can provide the intergation I’m suggesting.

  10. So long as it’s clear you won’t be tagging, you’ll be classifying with a declared vocabulary 😉

    Anyway, that’s not *quite* what I’m implying.

    I’m saying that coming to the agreement in the first place can be a waste of time (depending on the size of the group and I’d suggest that “the JISC community” is too big, but UKOLN is probably OK), instead, just do it and see what tags emerge and standardise on those (not quite the individual preference free-for-all you suggest).

  11. Hi Phil – I suspect we’re not disagreeing. In a situation like this I can’t mandate a tag, but I can suggest one. And if this works (i.e. it becomes widely used) then that’s the one that becomes the ‘standard’. If it doesn’t, then something better will hopefully emerge. I’m suggesting a process for starting the tagging, along the lines of tags for previous UKOLN events (at IWMW 2006, for example, we suggested the iwmw2006 tag – and most people were happy to use this, which meant aggregating blog postings, photos on Flickr, etc was easy to do).


Submit a Comment

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>