Internet

Social Software - Some Thoughts

This article has been inspired by a chance comment during a conversation at the UKeiG stand at Online 2005, while viewing the new website. “Ooh! You’ve got a tag cloud!” I said, which elicited the response, “Is that what they’re called?”.

The drawing together of a number of recent developments leads me to think that a brief overview of social software would be valuable. For me, these trends include a project I’m working on that involves adding keywords to records of non-print based materials/objects; using del.icio.us to collect interesting bookmarks after my browser-based bookmark list failed, again, to move with me to another PC; and a recent discussion about which emerging technology trends might be adopted by the library service.

So, what is social software?

Social software is a phrase applied to web-based services that allow users to store, share, search and sort particular resource types. Two widely known ones are Flickr [1] for photo sharing, and deli.cio.us [2] for sharing webpage bookmarks. Both of these sites work in broadly similar ways – there are two aspects to them. Firstly, if you are a registered user you can collect photos/bookmarks, then add keywords (also known as metadata or tags) and share your collection; secondly, anyone can enter a keyword into the search box and see what comes back. It is also possible to browse a particular user’s tags, or browse by tags. Both these sites can provide a seamless experience while working on the web, since applets allow users to add to their collections while browsing.

Phil Bradley has put together an excellent list of this type of web service at http://www.philb.com/iwantto/share.htm.

Tags can also be displayed in a more visual form via Tag Clouds.

What does a tag cloud look like?

If you’ve never seen a tag cloud, the flickr one is here: http://www.flickr.com/photos/tags/.

Tag clouds are described in Wikipedia as: “A tag cloud (more traditionally known as a weighted list in the field of visual design) is a visual depiction of content tags used on a website. Often, more frequently used tags are depicted in a larger font or otherwise emphasized, while the displayed order is generally alphabetical. Thus both finding a tag by alphabet and by popularity is possible. Selecting a single tag within a tag cloud will generally lead to a collection of items that are associated with that tag.” [3]

All those tags without a controlled vocabulary!?

To those of us used to working in the defined world of thesauri and controlled vocabularies, the idea of the general public adding their own metadata and keywords to objects can seem to be inviting disaster and chaos. However, a recent research paper [4] by Marieke Guy at UKOLN examines tagging patterns in both del.icio.us and Flickr, and draws some interesting conclusions.

Their random sampling of tags from both sites indicated that only ten to fifteen percent of tags sampled were single-use tags. This probably contradicts what one might, intuitively expect from user based tagging. They conclude that there is now a movement within these sites for users’ tagging behaviour to start to converge on some frequently and heavily used tags. There are a range of issues with tags that will be familiar to readers – misspellings, use of plurals rather than singular terms, the use of punctuation to indicate hierarchy in multiword strings, synonyms and homonyms, and ‘nonsense’ tags used by groups of friends. Guy discovered that somewhere around a third of tags they examined were ‘malformed’. It is entirely possible that over time, these sites will develop mechanisms to overcome some of these issues while retaining the richness of language and ideas within the tag collections.

How does all this apply to libraries?

Connotea [5] is an example of seeing the potential of a site and adapting it for a particular audience, in this case scientists. Connotea is a free online reference management and social bookmarking service for scientists, developed by Nature Publishing Group’s New Technology Team. They took the open source code from del.icio.us and developed the Connotea site. Some of the features they have introduced, which extend the del.icio.us offering, include

recognising URLs from common archives and importing bibliographic data – resources supported include Blackwell Synergy, PubMed Central, Science, Wiley Interscience, Highwire press publications, Amazon, HubMed, D-Lib magazine and institutional repositories using the EPrints software.
enabling the creation of user groups, which allow a research team to manage their reference lists collaboratively and selectively view recommendations generated only from within the team.
enabling a researcher to use an OpenURL resolver, perhaps via a library service, to link to an institution’s holdings of a reference in their Connotea collection.
enabling the import and export of references in RIS format to other reference management software.

Services such as RSS feeds for any results, search and bookmarklets are also available. It is possible to create multi-word tags in Connotea, however the software does not distinguish between upper- and lower-case. This can be seen from their tag cloud on the home page, as both ‘Avian Flu’ and ‘avian flu’ are present.

Final thoughts

These types of sites are fascinating and have an appeal for easily sharing resources with others. Connotea has shown that these ideas can be reused in an academic environment and I would anticipate that the visual representation of tagging patterns may creep into other database-like offerings. Would it be useful to see the keywords or controlled vocabulary terms from the results of a bibliographic database search displayed in a tag-cloud-like manner?