- Tech Sharecase, 8 December 2017
- Technology Sharecase, 10 November 2017
- Tech Sharecase, 20 October 2017
- Tech Sharecase, 8 September 2017
- Tech Sharecase, 5 May 2017
- Tech Sharecase, 21 April 2017
- Tech Sharecase, 18 November 2016
- Tech Sharecase, 23 September 2016
- Tech Sharecase: 28 July 2016
- Tech Sharecase on 10 May 2016
- December 2017
- November 2017
- October 2017
- September 2017
- May 2017
- April 2017
- November 2016
- October 2016
- July 2016
- May 2016
- October 2015
- September 2015
- May 2015
- April 2015
- November 2014
- June 2014
- April 2014
- March 2014
- February 2014
- January 2014
- October 2013
- September 2013
- August 2013
- July 2013
- May 2013
- April 2013
- March 2013
- February 2013
- January 2013
- October 2012
- September 2012
- July 2012
- June 2012
- May 2012
- March 2012
- February 2012
- January 2012
- November 2011
- October 2011
- September 2011
- August 2011
- July 2011
- June 2011
- May 2011
- April 2011
- March 2011
- February 2011
- January 2011
- December 2010
- November 2010
- October 2010
- September 2010
- August 2010
- July 2010
- June 2010
- May 2010
- April 2010
- March 2010
- February 2010
- January 2010
- December 2009
- November 2009
- October 2009
- September 2009
- August 2009
- July 2009
- June 2009
Tag Archives: Federated search
I’m not sure sure if there’s a name for these homegrown federated search services that let a library offer a single search box that serves up search results bento-box style from the catalog, a discovery tool, the library website, etc. The libraries at the University of Michigan and North Carolina State University have had such search tools for a while; now it looks like Columbia University is testing one out, too: CLIO Beta.
Announcing CLIO Beta (video intro)
For the “Articles” search piece, it looks like Columbia is using the API from their Summon subscription. For the “Catalog” search and the “Academic Commons” (the institutional repository) search pieces, they seem to be using Blacklight, an open source discovery layer developed by the University of Virginia.
Stephen Francoeur, Ellen Kaufman, Louise Klusek, Rita Ormsby, Mike Waldman
Attendees were asked to come to today’s meeting with something related to search (a new search engine, a new search feature or interface, an article or blog post about search, etc.)
Guide to Searching
We looked at a video tutorial and companion website from the library at the University of Massey (NZ) that walked users through the basics of search. We liked the website’s screenshot and the way the video had a table of contents that let you jump head to a specific section.
The library at the New York Law School has a search tool called DRAGNET that lets you find laws and other legal materials on various free legal databases. It was built using Google Custom Search. More details about how the service was put together can be found on this ACRL page. We wondered what it would be like to do something like this ourselves that searched a collection of open business-related databases on the web. We also talked about the plans for the Law.gov website, which are underway and will assemble a free resource of the nation’s laws.
One Search Boxes on Library Websites
Following up the discussion of search tools built with Google Custom Search, we looked at a Jamun, project being developed by the Leddy Library at the University of Windsor by Art Rhyno and Mita Williams. This tool will offer users a single box that searches across a number of different key resources. We also looked at the single search box (QuickSearch) that the library site at North Carolina State University features. We tried a bunch of different searches to see what comes up (notes of our searches didn’t get recorded, but you can try this one for “market share honda” as a useful example).
A Model for Teaching Search
We talked about librarian Iris Jastram’s model for teaching search, which she calls “exploding an article” and outlines in this blog post at Pegasus Librarian. In the classroom, students are introduced to the concept of being able to take one scholarly article that is relevant to them and use it to move in different directions to find others like it:
- using Web of Science, you can move forward in time by looking for articles that have cited the one in hand
- using the bibliography in the article, you can move back in time by tracking down the sources that the author used
- using key terms in the article or in the descriptors for that article in a database, you can move to the sides to find articles that are about the same things
We looked at a draft of a Newman Library toolbar that was built using the free LibX service. The toolbar features a search box for the library catalog, for the e-journals lookup tool, and for Bearcat. It also turns ISBNs, ISSNs, and DOIs into clickable links that will run lookups in relevant search tools from the library. Finally, it places an icon on the pages describing books in Amazon and other online booksellers; when the icon is clicked, the toolbar runs a search for that item in our catalog. This toolbar for our library is still being finished up and will be available soon.
I’m intrigued by libraries that are trying to find creative ways to cut down on the number of places users have to go to search for information. First, we had federated search tools that sent out searches to a bunch of places and aggregated the results; now we’re seeing discovery layers (like Summon, Primo, EBSCO Discovery, WorldCat Local, VuFind, Project Blacklight, etc.) that aggregate content into a single index and offer search results more quickly than federated search typically can. Still, these new discovery layers may not be the ideal search tool for every silo the library owns or licenses.
Our students, trained by Google to look for a single search box to rule them all, are probably looking for unified search tools on our site. While discovery layers are great and all the rage now, it seems like some libraries have accepted that discovery layers may not be the ideal solution for searching all the silos or presenting results from those silos. If you take a look at what the libraries at North Carolina State University have done, you’ll see that the “Search All” box on the library home page presents results from different silos and spreads those results out on the search results page into different clusters. Check out what happens if you do a search there for “market share.” Notice the way that the articles are presented via the discovery tool (Summon), the books via the catalog, and other categories of content via other search tools:
This federated “search all” feature on the NCSU libraries web site actually pulls together a number of different search tools, which are explained more fully elsewhere on the site.
Librarians at the Leddy Library at the University of Windsor are at work now on a tool called Jamun that does something similar: it gives the user a single search box that will do a federated search across several different silos and present those results spread across the page in different categories. This project is being built using a Google Custom Search engine. You can learn more about it from the video recording the presentation about it at the recent Code4Lib North event and by checking out Mita Williams’ blog post about it (which includes an embedded slidecast) and Art Rhyno’s blog post from April.
[kml_flashembed movie="http://www.youtube.com/v/aZGqCFSZGAg" width="425" height="350" wmode="transparent" /]
Robert Drzewicki, Stephen Francoeur, Gerry Jiao, Ellen Kaufman, Louise Klusek, Wilcina Longdon, Jin Ma, Kannan Mohan, Ryan Phillips, Linda Rath, Mike Waldman
Database Trial Reviews
We discussed briefly the pros and cons of writing up reviews of database trials on the reference blog and the library’s internal mailing list.
Bike Route Maps and Directions
In celebration of Bike to Work day, we looked at a number of options for finding bike route maps and directions:
- maps can be downloaded and added to Google Earth
- recently added biking directions
- it would be nice if the maps displayed info from Twitter feed for different bridges in NYC
- missing some of the human element (insider knowledge about safest, easiest routes, etc.)
Call Numbers in Catalogs and Library Floorplans
We tried in vain to recall what library has a catalog that lets you click on a call number in the catalog to show you the location of that item on a floorplan. As we tried to remember which library has this feature in its catalog, we looked at the catalog from the library at the University of Huddersfield (UK), which offers on the item record a visual shelf browse feature, a QR code for the book (which probably leads to the permalink for the item), and circulation stats for that item (see for example this record for The Iliad of Homer).
At the recent OpenSciNY conference at the Bobst Library at NYU, a group of librarians, scientists, and publishers got together to talk about open access publishing, open source software, and opens notebook science. Among the more interesting things talked about were:
- Flickr and Astrometry.net: amateur astronomers are uploading images they’ve taken with their telescopes to Flickr. One of the presenters at OpenSciNY, David Hogg, worked with some colleagues to put together a service that uses the Flickr API to identify any images that have been recently tagged with “astrometry.” Once tagged in this way, an image on Flickr will be analzyed by the Astrometry service and a comment appended to the image that details the celestial objects visible in the image (see this one from Flickr as an example)
- ChemSpider: Antony Williams from the Royal Society of Chemistry talked about the problem of finding reliable and comprehensive information on chemical structures on the web. ChemSpider describes itself as a “chemistry search engine” that “has been built with the intention of aggregating and indexing chemical structures and their associated information into a single searchable repository and make it available to everybody, at no charge.”
Substitute for EtherPad
EtherPad, a recently shuttered free service that allowed for collaborative editing of documents, released its source code, thereby allowing a number of clone services to be created. One such service is Sync.in
We talked a while about the difference between discovery tools (like Summon and EBSCO Discovery Service) and federated search tools (like 360 Search, which we use for our own Bearcat Search). It was noted that with the new discovery tools, the thing that takes the longest to set up is getting your catalog records into the system. What makes a discovery tool different from a federated search one is:
- With a discovery tool, you are searching one, centralized index of records that the vendor has assembled; with federated search, your query is being transmitted simultaneously to all the vendors that you can connected to your fed search tool. Search results are returned faster in discovery tools because of this difference.
- The vendor of a discovery tool can normalize all the data stored in its index, making results more consistent (and helping to speed up the delivery of search results) and manipulable (the facetting of results works better in discovery tools).
We wondered if many faculty outside the library use Bearcat as a means to identifying databases that were previously unknown to them but might be useful for their research needs.
We might have demos of two different ERM (electronic resource management) products this June.
Citation Management Tools
We wondered to what extent faculty and students are aware of and maybe use citation management options available to them:
Jin Ma, Mike Waldman, Matt Haugan, Ellen Kaufman, Mike Waldman, Stephen Francoeur
EBSCOhost Integrated Search
Mike Waldman showed us this tool that CUNY is looking at and asked us to think about how it compared to Bearcat Search. EBSCOhost Integrated Search will search everything that CUNY Central pays for (as well as the unique EBSCOhost databases we’ve subscribed to). We can customize the display of search results so that the databases are grouped into first and second tiers. We took a look at how Brooklyn College has set up their instance of this tool. One difference that we noted was that Bearcat doesn’t have search field for “source” but the EBSCOhost Integrated Search does.
We talked about how the free xFruits web services can be used to repurpose RSS feeds or to create new ones. For example, you can convert email into an RSS feed using this tool.
In light of this week’s release of beta invitations to Google Wave, we talked again about what this new tool might allow us to do.
Software for Collaboration and Communication
Ellen Kaufman talking about technology at her old job and how they used Microsoft Sharepoint, portals, and Confluence.