UK Web Focus (Brian Kelly)

Innovation and best practices for the Web

  • Email Subscription (Feedburner)

  • Twitter

    Posts on this blog cover ideas often discussed on Twitter. Feel free to follow @briankelly.

    Brian Kelly on Twitter Counter

  • Syndicate This Page

    RSS Feed for this page

    Licence

    Creative Commons License
    This work is licensed under a Creative Commons Attribution 2.0 UK: England & Wales License. As described in a blog post this licence applies to textual content published by the author and (unless stated otherwise) guest bloggers. Also note that on 24 October 2011 the licence was changed from CC-BY-SA to CC-BY. Comments posted on this blog will also be deemed to have been published with this licence. Please note though, that images and other resources embedded in the blog may not be covered by this licence.

    Contact Details

    Brian's email address is ukwebfocus@gmail.com. You can also follow him on Twitter using the ID briankelly. Also note that the @ukwebfocus Twitter ID provides automated alerts of new blog posts.

  • Contact Details

    My LinkedIn profile provides details of my professional activities.

    View Brian Kelly's profile on LinkedIn

    Also see my about.me profile.

  • Top Posts & Pages

  • Privacy

    Cookies

    This blog is hosted by WordPress.com which uses Google Analytics (which makes use of 'cookie' technologies) to provide the blog owner with information on usage of this blog.

    Other Privacy Issues

    If you wish to make a comment on this blog you must provide an email address. This is required in order to minimise comment spamming. The email address will not be made public.

The Search Session At MW 2008

Posted by Brian Kelly on 14 Apr 2008

On the final day of the Museums and the Web 2008 conference (Saturday !) I chaired a session on Search. There were only two papers presented at this session – and as the session was scheduled to last from 11.00-12.30 both of the speakers were happy for the session to provide an opportunity for general discussions after the papers had been presented.

Terry Makewell ‘s paper was entitled “The National Museums Online Learning Project Federated Collections Search: Searching Across Museum And Gallery Collections In An Integrated Fashion“. As described in a blog post by Nate Solas, the paper described the approaches to federated search being taken by 9 partner organisations in the UK. The two search technologies described were  OAI/PMH and Opensearch – and a decision was made to use Opensearch, due to its simplicity, the short timescales and the limited technical expertise and resources available by some of the partners.

Following Terry’s talk Johan Møhlenfeldt Jensen, Museum of Copenhagen, Denmark presented a paper on “Approaches To Presentation Of Cultural Heritage Information In The ALM-Area In Denmark And Scandinavia“. This paper complemented Terry’s paper nicely, and highlighted some of the challenges posed by federated search including the differing cultures across the archives, libraries and museums domains and the differing cultures across the Scandinavian countries.

The discussions afterwards focussed on whether a simple approach to federated search would be sufficient. Mike Ellis asked Terry whether used of Google search technologies, such as Google Coop, had been considered. It seems it had, but ruled out due to the complexities posed  by use of session IDs on some of the collections. In a subsequent tweeton the Twitter back-channel Mike pointed out his experimentation with Google Coop across a number of museums – and this was briefly tested by the two speakers after the session had concluded (as an aside I should note that this was the only relevant Tweet received during the session – however Terry and I were also interested in the football scores which I receive on  my Twitter account, including the flurry of goals conceded by Derby County!) .

The discussion on simplicity versus sophistication led to discussions on the user experience. Following a question on evidence of use of advanced search capabilities, data from an Australian example showed that a very low percentage of users (1%, I think) accessed an advance search capability – and, indeed, most users submitted only a single search term!  I pointed out that the importance of simple interfaces was likely to grow as use of mobile devices became more popular – a comment that was particularly pertinent to the MW 2008 conference, as the WiFi access problems conference delegates had experienced the previous day were apparently due to the large numbers of network users who were using an iPhone or Nokia N95.

There was a feeling, I think, that federated search may, in the future, be provided by mainstream commodity products – and, indeed, as collections management tools evolve and start to provide static URIs, the benefits of solutions such as Google Coop may become even more apparent.

Will there, I wonder, be a session on federated search at future MW conferences or will this area be, like institutional search, be addressed by mainstream solutions?

3 Responses to “The Search Session At MW 2008”

  1. Mike L said

    The NMOL fed. search was required primarily because we (I was part of this for a while) needed solid control over what was searched for, but could not impose much (occasionally any) standards on the partner’s content or search applications. Its entirely likely that more and better engines for this sort of thing will appear and they will need to be highly configurable for a long time yet. Plenty of Museum content is _still_ in the ‘deep web’ – unbrowseable – so a goal of such projects should also be to expose as much content as possible to Google et al via faceted browse until we have some lightweight semantic structures. I’ve been saying this stuff since 2000 ish? I need a new mantra.

  2. Spookily Google is now looking at the deep web. Not in a sensible way IMO, but anyway Mia blogged it…

    http://openobjects.blogspot.com/2008/04/its-wonderful-wonderful-web.html

  3. […] The Search Session At MW 2008 […]

Leave a comment