I was one of the authors of the report --thanks for this very helpful contribution. You are right that for NGC4LIB much will sound familiar but, as you point out, there are some themes that are worth further thought and action. If people want a quick way in to what is a fairly detailed report I suggest taking Section 7 ('Making Decisions') first.
I'm glad you picked up on the 'Context/user behaviour' issue. I couldn't agree more with your comment that this is an 'untapped concept in Library Land'. We believe there is great opportunity for the domain and we'll be picking this up for further work at a follow up workshop on 27th June in London. User context and usage data can drive a number of valuable services. In particular a 'recommender' service (a la Amazon). I've noted the work being done at CSU. Marvin Pollard gave a presentation this month at the EUSIDIC conference in London. He shows (screenshots) how they are using data to drive a recommender service. I belive he uses data collected from the URL reolvers and this seems to me to have huge potentail value. In The UK Dave Pattern at University of Huddersfield is also doing this but I thin his work is based around circulation (loans) data. Clearly the bigger the aggregated dataset the better (potentially at least) the service--and that's a real challenge for a very fragmented UK library landscape. So you can bet LMS (ILS) businesses like OCLC and Talis--that also have search/cataloguing 'Platforms' are looking at how they can make use of aggregated search clickstreams etc. You've got to bet too that other LMS (ILS) vendors are working on doing something.
In the report we use the term 'Vertical Search' to encompass the Primo, Encore, AquaBrowser, Endeca, WorldCat local, VuFind etc etc approaches. We do this simply to show that what is happening in libraries is part of a much wider trend in search. Vertical search is needed because in some circumstances Google isn't the answer.
At the recent EUSIDIC conference Jens Vigen from CERN (birthplace of the Web) described a 2007, a survey that was conducted to understand which tools High Energy Physics (HEP) scholars use to find the information they need. Over 2000 answers, representing about one- tenth of the active HEP community, were collected and show that community-driven resources largely dominate the landscape. Commercial services such as Google and Google Scholar serve only a small proportion of the users (8% for Google and 1% for Google Scholar). Whilst the HEP domain is very specialised the survey results prompt some questions about the supposed universality of Google.
This then puts the product responses of the library systems vendors into a wider context (and explains why vertical search providers like Endeca are looking at libraries). This is where they hope they can compete with Google......they’ll be better able to compete if they made use of usage and context data.
We are grateful for all feedback on the report.
Ken
Ken Chad, Director, Ken Chad Consulting Ltd
Tel +44 (0)7788 727 845
Email: ken_at_kenchadconsulting.com www.kenchadconsulting.com
Registered in England & Wales - Registered No. 6099834
I was one of the authors of the report --thanks for this very helpful contribution. You are right that for NGC4LIB much will sound familiar but as you point out there are some themes that are worth further thought and action. If people want a quick way in to what is a fairly detailed report I suggest taking Section 7 ('Making Decisions') first.
I'm glad you picked up on the 'Context/user behaviour' issue. I couldn't agree more with your comment 'untapped concept in Library Land'. We believe there is be a great opportunity for the domain and one which we'll be picking up for further work at a follow up workshop on 27th June in London. User context and usage data can drive a number of valuable services. In particular a 'recoomender' service (a la Amazon). I've noted the work being done at CSU. Marvin Pollard gave a presentation this month in London which shows how they are using data to drive a recommender service. In The UK Dave Pattern at University of Huddersfield is also doing this. Clearly though the bigger the aggregated data set the better (potentially at least) the service--and that's a real challenge for a very fragmented UK library landscape. So you can bet ILS businesses like OCLC and Talis--that also have Search/cataloguing 'Platforms' are looking at how they can make use of aggragated search clickstreams etc. You've got to bet too that other vendors are working on doing something (we'll we hope so).
In the report we use the term 'Vertical Search' to encompass the Primo, Encore, Aquabrowser, Endeca, WorldCat local etc etc approaches. We do this simply to show that what is happening in libraries is part of a much wider trend in search.
So in some circumstances Google isn't the answer.
At the recent EUSIDIC conference Jens Vigen from CERN (birthplace of the Web) described a 2007, a survey that was conducted to understand which tools High Energy Physics (HEP) scholars use to find the information they need. Over 2000 answers, representing about one- tenth of the active HEP community, were collected and show that community-driven resources largely dominate the landscape. Commercial services such as Google and Google Scholar serve only a small proportion of the users (8% for Google and 1% for Google Scholar). Whilst the HEP domain is very specialised the survey results prompt some questions about the supposed universality of Google
This then puts the responses of the Library systems vendors into a wider context (and explains why vertical search providers like Endeca are looking at libraries). This is where they can compete with Google...
We'd be grateful for all feedback on the report.
Ken
Ken Chad, Director, Ken Chad Consulting Ltd
Tel +44 (0)7788 727 845
Email: ken_at_kenchadconsulting.com www.kenchadconsulting.com
Registered in England & Wales - Registered No. 6099834
> Date: Tue, 22 Apr 2008 15:37:48 -0400> From: emorgan_at_ND.EDU> Subject: [NGC4LIB] "next generation" library catalogs and a jisc report> To: NGC4LIB_at_LISTSERV.ND.EDU> > "Next generation" library catalogs were more or less the topic of the> recently published report by Sero Consulting called "JISC & SCONUL> Library Management Systems Study: An Evaluation and horizon scan of> the current library management systems and related systems landscape> for UK higher education". [1] I thought I would share a few of the> salient points here in order to spread the wealth and foster discussion.> > First, the report is intended to recommend courses of action> regarding integrated library systems (ILS) considering the current> networked environment. To this end the consultants surveyed> librarians, conducted telephone interviews with vendors, did their> own environmental scan, and bounced ideas off sets of experts -- a> "reference group". In the end the report does not suggest purchasing> any new ILSs at this time. The environment is too turbulent. More> specifically, the report recommends, and I paraphrase:> > * investing in systems with caution but not complacency> since the role of the "conventional" library may appear> unclear,> > * seeking increased value in ways to improve services> around the ILS,> > * focusing on breaking down barriers to resources such> as implementing single sign-on, unifying workflows, and> especially liberating metadatg for re-use,> > * emphasizing service oriented architectures for> de-coupling ILS components,> > * re-exploring the role of consortia to exploit social> networking effect, and> > * working more collaboratively in order to develop and> enhance the potential of Library 2.0 functionality.> > Second, the report echoes many of things that have been said here,> such as: the changing nature of user expectations, the need to make> library content more accessible, the use of Web Services computing> techniques, the non-integratedness of "integrated" library systems,> the mature library market and how it is dominated by a small number> of software providers in turn owned by private equity firms, the> viability or non-viability of open source software, challenges in the> process of change, and the need for Web standards in library systems> (not library standards). Nothing too new.> > On the other hand there were two things in particular I found> interesting. First, while many people use Google to find information,> Google does not have enough context to be helpful/useful at the next> level. Yes, these search terms result in the following list of URLs,> but if Google knew more about you -- your context -- then it might be> able to create an even more relevant list of URLs. This, IMHO, is an> untapped concept in Library Land. Libraries always exist as a part of> a larger group. By knowing more about the constituents in the larger> group an information system -- be it a computer or a person -- can> provide better, more relevant service. It is not all about objective,> dry applications and systems. It is also about building relationships> that can only be done only at the smaller institutional level. This> represents an opportunity for librarianship. From the report:> > Google isn't presently solving the entire problem. It> doesn't yet provide enough 'context', for example to> students and researchers. What's the best stuff? What> stuff should the institution licence? If we knew this> it would result in an improved use of resources. It's> about the user's context (e.g. university student).> Your profile gives better results. The better the> context is defined the better results will be. The> context problem is not solved in HE, which has an> opportunity here because its users are (relatively> anyway) well defined—with a particular goal anyhow.> --Dave Errington, CEO at Talis [page 70]> > Second, in this changing environment it might be a good idea to> rethink the role of librarianship, at least a little bit, and I liked> the three complimentary information service models outlined in the> report:> > 1. Aggregation - Collecting content as opposed to> providing federated search against it. Amazon.com is> good at this technique.> > 2. Network Effect - Maximizing value and reducing unit> costs by exploiting the knowledge of the masses, open> data & platforms, and clickstreams. From a member of> the reference group, "The changing nature of> information and user behavior gives rise to some> serious professional questions about the business we> are, the nature of our expertise, and how much we are> actually in a position to judge what is good for the> user. All this prompts questions about how systems> should develop." [page 84]> > 3. Long Tail - Seen as an opportunity for subject> specialists, this is about focusing library services> against locally developed and unique content. Digitize> that stuff in your "special collections". This is akin> the ideas of Ray Bradbury's novel Fahrenheit 451 where> individuals memorize a single book. [2]> > Finally, I recently asked for greater input and perspectives from> vendors regarding "next generation" library catalogs but the response> was meager. Since much of the report was gleaned from telephone> interviews with library software vendors, the report includes almost> 40 pages worth of vendor profiles where folks can learn a bit more> about vendor perspectives.> > > [1] http://www.jisc.ac.uk/media/documents/programmes/> resourcediscovery/lmsstudy.pdf> > [2] This idea is far from new, and I was certainly not the first to> articulate it, but I did elaborate upon it in a column for Computer> In Libraries in 1997. See: http://infomotions.com/musings/unique-> collections/> > --> Eric Lease Morgan> University Libraries of Notre Dame
_________________________________________________________________
Be a superhero and win! Play the Iron Man Mashup Game
http://www.ironmanmashup.co.uk
Received on Wed Apr 23 2008 - 11:20:36 EDT