Jan 252011
 

Today I will be live blogging Haggis and Mash, a mashed library event taking place at the National eScience Centre in Edinburgh. This means this post will appear incomplete/updated all day. Formatting tweaks and the odd fix of typos and names coming soon. Apologies for picture quality – largely taken on my phone at handy opportunities.

The day kicked off with a lot of hellos and registration.

I’ll be blogging all day – for more information on the programme have a look at the mashed library wiki here: http://www.mashedlibrary.com/wiki/index.php?title=Haggis_and_Mash.

Owen Stephens introduced the day, opening with Chris Keene, University of Sussex talking about VuFind.

Started 2007 at Villanova University – it’s a free open source resource discovery or library catalogue service. What is a catalogue? Chris has a chart comparing information. Search records and show stuff – that bits easy. Showing availability, holdings and shelfmark – it can normally be done but there is no standard way to do that. My account, reserve books etc. harder but doable. Inter Library Loans and other functions

Really works well – like any other catalogue out there. It works right out of the box and really drives the MARC record hard – uses all those things that usually just sit there in the background. That’s good and bad  – it shows up your bad records very quickly but at least that information is being used.

Cools URLs – they are reliable and consistent “probably the biggest innovation in the library world”.

Apache Solr is VuFind’s Engine, it drives the functionality. LAMP is native home but it can run on Windows. It’s easy to install (even though it sets up Java stuff that is usually rather trickier). Can take as little as 8 minutes to install VuFind from scratch on Amazon Web Services (there’s a script that makes it even easier as well).

People use VuFind to replace other catalogues but there is a growing need for web catalogue interfaces which search a subset of a Library’s LMS catalogue in another context – on the careers office website or the rare books or the DVDs or a subject base etc. That’s one of the things that is also easy to do with VuFind. When you have those discussions you can avoid lots of small systems and instead use the big library system. Question: how does this deal with non MARC? Chris says that it can work fine but that he was mainly thinking of materials splitting off from the main catalogue that would already be there.

Connecting VuFind to the library management system can be tricky – both for loans, holdings and my account type stuff but getting records into it and getting an installation up and running is easy and well worth a try – go experiment!

Now onto:

Boon Low, NeSC, University of Edinburgh talking about VuFind vs Blacklight.

Boon works on the UX2.0 project, with his colleague Lorraine Paterson who’s along in the audience, and they have been working on building a repository and user interface.

What is Blacklight? A discovery interface for the library catalogue, digital repository front end, single search interface. Similar to VuFind it is essentially a front end for Apache Solr. The UI is cleanly separated from the repository or data store. There is some synergy between Blacklight and Solr (as names may suggest) – it is very simple to install both together and indeed that will be demoed later this afternoon in the developer session (1.30pm – bring your laptop).

Features: faceted search (Solr), user services such as bookmarking etc., saved searches and search history. Boon is now going to demo and compare both types of system on live sites.

In VuFind the UI rather buries the saved search and search history links. In both Blacklight and VuFind you have to login to access the saved search. Naming of site areas vary in VuFind. Bit confusing. Implemented differently in Blacklight but also problems here too. I’m not sure I’ve entirely gotten the hang of the major differences here but think others on the hashtag may have much to offer here.

Bookmarking wise both systems offer bookmarking for logged in users. Blacklight also offers a new Folder feature for non logged in users to mark items. But this is sort of competing with the existing bookmarking system. Blacklight has single and multiple exports, currently VuFind only has single exports but an audience member adds that multiple exports will be in the next version of VuFind.

Both systems have usability issues (though different) and both are implemented in different ways. Boon is now showing us VuFind (London School of Economics – LSE) and Blacklight (University of Virginia – UVa) installations. UVa Blacklight front page highlights content a little, faceted search comes from left hand menus. LSE VuFind is more straightforward. Blacklight has breadcrumbs for faceted search available – very helpful for users. Display shows lots of the MARC record. But the UVa breadcrumbs UI layout isn’t Boon’s favourite. LSE VuFind shows breadcrumbs in very different ways – high on left hand menu – it frees space for other content (in this case suggested topics related to the search). Boon thinks this is one of the better implementations.

Finally Boon gives us more examples of the Blacklight UI.  Most include breadcrumbs which are promininet as are left hand facets. Metadata is shown with holding information. The latest version of Blacklight has a smaller breadcrumb display. Stanford University also use Blacklight but with very different look and feel. Some visual hints around facets to let user know that they can open menu. Their breadcrumbs UI is perhaps too cluttered. There is floating element for pagination and Boon likes this as it has the look and feel of Mobile phone apps. Users don’t have to scroll to do pagination. Think that UI such as BlackLight and VuFind should both add in this type of functionality.

And now UX2 Blacklight mobile web for Edinburgh University. Boon and Lorraine are currently doing research on designing a UI for such a small screen. Even designing for the iPhone 4 with it’s 900 oixels width can be challenging.

Contact Boon at: boon.law@ed.ac.uk

Questions:

Q1, Mandy Phillips at Liverpool John Moore’s University: looks easy to set up but what’s the staff time to develop and design this. So if I’m on a paid LMS and have 3 staff on it and I want to move to Open Source how long will it take?

A1: (CK) Easy to start, much longer to get perfect. Skills wise those already working on library systems and php and mySQL or has installed ePrints then they should be able to do this. The customisations and integration with logins will vary depending on different systems

(BL) the expertise of your staff you should already have. For Blacklight you probably need RubyonRails for instance. At Stanford they have a team of 8. They went from 1 to 8 developers in the team without new hires – retraining. And a claim that you can learn Ruby in a week – though Boon doubts that speed! For UX2 project we have one developer but we’re struggling.

Q2, Iain from Glasgow Caledonian: Does all that functionality get updated by the library management systems automatically?

CK: With Voyager there is a module that acts like an API and that does update. For Talis there is something called Project Jangle and a driver to connect to the library system. But it is a real time connection. Obviously there’s a risk and dependency there: your LMS goes down the UI goes down for instance.

Next up:

Matt Machell, Talis and the Juice Project – Open Source Interface Enhancement

Matt has been inspired by current Talis projects but also by having an 18 month year old and a lot of Dr Zuess!

Open Source for Matt is all about building blocks – things that can be combined together and work really well. So aren’t UI’s where open source often falls down? Well no, there is “the little language that could” – Javascript makes a lot of web 2.0 sites look superb. However there are browser issues – differences in which different implementations work, what changes are allowed. It can be a bit of a headache for working across browsers. The DOM is a mess. The solution is jQuery – it fixes all the inconsistencies across browsers. You can focus on what you want and making it work where you like. Released in 2006. Now bundled wit Microsoft and Nokia tools. Estimated use on top 10,000 websites. Twitter use it for instance.

Why so popular? It’s easy. It solves a common problem in an elegant way. There is a really stromg community. And really easily extensible. So things that don’t work get fixed and recontributed. Really easy to add functionality to pages. Plugins let you add etra functionality.

How does this help your library catalogue? Well you can rewrite your interface. Cue “Broadminster University” – the Talis test library name. So you can add hovering help text very nicely as you hover over an item for instance. The code for this is straightforward – script at top of page and then they are just used wherever needed in the page.

You could… pull in data from elsewhere (like Google Docs, Yahoo Pipes, Flickr, etc.). Or to hide excess catalogue data until needed. And it lets you enhance your catalogue without neccassarily touching core code of your library system.

But… it’s easy to unintentionally slow a page down, break usability or reduce accessibility.

Juice! juice-project.org

Wouldn’t it be ace to put together lots of different library specific functionality and combine it, share it and have it in one central place. Juice sits on top of jQuery and in current version is more like a plugin. Lots of functionality here so now for a demo of some fun stuff!

For instance: it’s easy to grab the code for the Google Preview of a book on the page for the relevant record. Or you can add in some information from a blog via RSS. You just have to provide API key and RSS feed for instance. Code shown again and it’s easy to use and tweak.

Great for bookjackets/alternatives, for geolocation or Google Maps, Facebook Like Buttons, Tweets, Delicious bookmarks, QR codes etc. But you have to think what is going to be useful for your main users – that’s key.

Why do this Open Source?

Many hands make light work, sharing innovation, don’t reinvent the wheel, basically build lots of small reusable components to slot into an existing system.

So… JavaScript is a pain, jQuery makes Javascript less painful, plugins can solve common problems and Juice helps with common library problems.

jquery.com

eclecticdreams.com / @shuckle

http://blogs.talis.com/prism.

Q&A

Q1: I’m impressed – thank you. My question is about personalisation. Most libraries do not deliver at just one level – whether educational level or accessibility etc. Is there something that identifies needs at front end? Something to identify this user only needs ebooks for instance?

A1) Obviously at the moment not in place. Need access to information about the user. You’d need some sort of webservice to do that but certainly theoretically doable. You’d need to know all that core stuff about personalisation needs though

Q2 – Andrew Jackson, NHS) any examples of libraries using jQuery on their interface

A2) Yes, have done plugins for ?? County Council, Manchester University has a plugin for Juice to pull in alternative book jackets (on Prism 3).

Q3) One of the things I’m always a bit cautious about is what’s the long term support issue. what sort of community is there for a long term guarentee for the future?

A3) like any open source project it lives or dies on community. For Juice we have a team at Talis, some people in Sweden. But the fact that it’s open source and the code is available from Google code it is at least entirely possible to grab the code and work onwards yourself but you can’t guarantee anything.

Coffee break – with Burns’ night cupcakes:


Burns’ Night Cupcakes

EduApps – Martin Hawksey, eLearning Advisor HE and Sara Brown, eLearning Advisor FE

EduApps just won an award from the Scottish Government for innovation.

JISC RSCs are “stimulating and supporting innovation in learning” and there are 13 centres across the RSC network. The RSC N&E supports institutions all over the North and East of Scotland

What is EduApps?

There are loads of little bits of open source software you can intsall on your machine and use. JISC TechDis has been promoting these for a while but to install them you need admin rights for your machine. So the idea was to install apps to a memory stick and that can be plugged into any Windows PC and you don’t have to set anything up (idea also on the web at Portable apps – not a new idea just a good one).

Originally EduApps was AccessApps. There are 8 software collections. Free to download and use. Open Source and Freeware. Portable. Whilst we have packaged everything on a memorystick and handed them out today that’s not a sustainable model for us. Hence our website has a way to do this yourself onto a memorystick. Portable Apps is justopen source apps but EduApps includes a lot of accessibility apps that rather break that model.

So here’s what it looks like. We have some special tools for today as well as the standard packages. There are screenreaders, magnifiers, Gutenberg stuff is a speciality item – complete work of Robert Burns!, Accessible web browser web IE. Can not only add stuff to the collection but you can brand it for your institution and fully customise and personalise the memory sticks – whether individually or on behalf of your students.

How AccessApps used at Dundee College

Dundee have a strong accessibility focus and interest. They have set up a Portable Access Device Development Initiative (PADDI), SLIC funded prokect (innovation and development fund). Ran August 09-June10. They wanted to roll out AccessApps across the college. Involved trianing for college staff and students – including adapting training materials that the RSCs already have available. Dundee also wanted to brand AccessApps for Dundee. The RSC staff ran “train the trainer” type sessions. There were 3 pilot groups identified with about 120 students in those groups. The college identified apps and resources they wanted on the stick. AccessApps were set up on the network for various security reasons. They were keen to roll out memory sticks but what happened was that students tended to use these at home and then use the network version on uni machines. This is actually great as Dundee have now rolled out the Apps to all of their computers.

To support the roll out there was training and USB pens provided. Training and drop in sessions were set up and AccessApps was added to the library induction. And the custom content included library links etc.

Some of the AccessApps

RapidSet – sets colours for applications to meet your needs.

Virtual Magnifying Glass – does just as it sounds. Been used in interesting ways though – not just for accessibility but also by art students to look at detail etc.

VuBar – identifies a line of text at a time for those that struggle with large blocks of text.

Evaluation

Questionnaires went to the pilot groups at the beginning and end of the project. Focus groups ran as well.

Over 80% og students felt that AccessApps had helped them in their studies

Main message was flexibility – being able to use on and off campus was great and that range of 60 types of software was great for them – could pick and choose useful things as needed.

Chart of top applications used. Very widely spread applications. Three just mentioned all in there. The other two in the top 5 was TypeFaster – a typing tutorial – and TheSage – a dictionary and thesaurus. Students reported more use at college than at home, not sure how that will be reflected now fully rolled out. A lot of users were accessing the catalogue, induction materials, podcasts etc.

Really good feedback hence wider rollout. Now a key part of library induction, Staff CPD etc.

www.eduapps.org

Sam Stirling (s.stirling@dundeecollege.ac.uk) – re: Dundee College project

Sara BRown (@sasab/sbrown@rsc-ne-scotland)

Martin Hawksey (@mjawksey/

Q&A

Q1) Just Windows?

A1 MH) Well Mac and Linux are so accessible that they don’t need toughening up (lols). It’s really a numbers game but there are links to Mac apps on the websites.

Next Up

Julian Cheal of UKOLN, University of Bath – Exploiting Linked Data in Libraries

Archives Hub and CoPac and LinkedData and Julian makes it all pretty.

Some disclaimers: Julian is not a librarian. And he doesn’t really know much about libraries. Cue awesome Star Wars/SPARQL Wars intro!

First there was the web, created by Tim Berners-Lee, and organised. Beautiful sites like the BBC. Lovely. Then people got hold of it. GeoCities. Oh Dear. But now we have tools like WordPress and such. Things are moving on to look much better and we’re all adding data and information to the internet. And we have things like Wikipedia. I was lucky enough to see Jimmy Wales talking in Bristol recently – the whole world has built it. And you can edit it if it’s a bit wrong fantastic. And we have moved onto a time when the web is mainly full of funny cats. But users contribute this stuff.

Back in the olden days of the web people at Cardiff University started building stuff on films, IMDB. It started many years ago to the scale of site and information – all user contributed or at least user driven (and data driven). All data driven and though the site (nice Internet Archive stuff) has changed a lot that data remains the core.

Who here has used Google maps? You all did to get here today I reckon! It’s easy to use – you don’t need to be an expert to find things or set up your own maps. Similarly you can now buy houses on the internet with lots of special criteria. All driven by data. People can add, companies can add. Sharing leads to these great services. Last.fm is 100% user driven. They have a huge catalogue of artists and albums from users just listening to their music. All user and data driven. They have an API and you can autotweet your favourites etc.

This event is called Mashed libraries. The books on this slide are all about taking data and mashing it up into something new. The fact that we have had quite a few mashed events shows that this is quite popular (though not many of us in the room have mashed up data). Many of those that have mashed stuff up have used Yahoo Pipes! which is very easy and drag and drop but you can also do it in code – theres a huge range of how you mash.

So how does this relate to Linked Data?

Triples. RDF. They are not cupcakes. What are they?

Lets look at Star Wars.

Here’s Luke Skywalker and he knows Hans Solo. But how. Express that as a triple. Need a URI for that relationship. So Luke needs a URI – he’s THAT Luke Skywalker with context etc. Same for Hans Solo. And the last part if FOAF:knows. That’s all machine understandable. It can all get complex quickly.

RDF – data structure. Linked Data is a bit different. Back in a second.

So now that we know who that Luke Skywalker is we can ask DBPedia about him – information sucked in from elsewhere around that known entity. So going back to IMDB they now have RDF data that powers the site. So if you click on a link about Luke Skywalker you can see he’s played by Mark Hamill. Storing that information in RDF lets your computer analysis and access and understand that information much more richly.

The BBC have invested a lot into Linked Data for their wildlife pages – not only can you see a picture of a bird but connected to information about that bird etc. Or for the sports page to create stats. Wonderful mash ups but instead of having to go out and work out how to compare data in your mash-ups your URIs will tell you if these people/entities are the same and can put it all together accordingly.

The Guardian and the New York Times are both putting out information as Linked Data as well. And DBPedia has a super set up combining not only Wikipedia but also council sites, Ordnance Survey etc. The government is also opening up their data as linked data. For instance to interrogate expenses.

So, back to universities and libraries. This is an application created by Ben O’Steen to turn all the British Library data into Linked Data and adding things like the location of authors/articles etc. Which lets you see the map of authors

benosteen.com/timemap/index

A great example that linked data isn’t scary stuff that no-one understands but instead a way to do cool and fantastic mashups in an easier and more elegant way than before.

Previously we were talking about Luke Skywalker and knowing who that is does mean comparing and looking

Locah – SameAs

blogs.ukoln.ac.uk/locah

e.g. Ernest Shackleton – connecting to Archives Hub page, the viaf (OCLC) page – and confirm that it’s all the same etc. And build up links.

Q&A

Q1) RDF vs Linked Data

A2) RDF describes data. Linked Data (or Semantic web) says this is information that is linked in a certain way, usually by RDF. Sort of the implementation of the linkages made possible

Q2) Ben O Steen: Correction. Data on that map is from Crystalography not British Library.

Q3) Understand a bit about Linked Data. Think I’d recognise RDF if I saw it. Less sure of the tools I’d need to do things with RDF?

A4) What is the Yahoo Pipes of RDF/Linked Data? No ready made tools just now. Chris? has a tool to upload exel spreadsheet and change into linked data. Not so many user friendly bits. Quite a new technology and not quite there yet.

Owen: Tony Hirst has done some stuff using YQL for querying linked data. Not a native tool but can be used to manipulate and query data – have a look at his blog OUseful.

Julian: RDF is just a concept. Triples – subject, predicate and object. It’s not a thing but a concept.

Q4 Gerard Bennett, University of Westminster: How do you work with linked data that isn’t publically available – subscribed or private data?

A4) If data is hidden away it’s not really going to work because of the URI concept.

OWen: I would disagree there. At the heart URIs provide identification. Like you can have an intranet and the internet – there’s no reason you cannot have an intranet of linked data. Not a technological issue though lots of benefits to open data (and LOD – Linked Open Data). Paul Walk says there is an important

Semantically RDF structures nothing use – relational data structure. Can’t do anything with RDF that couldn’t do with relational data base. the difference is that it can sit in lots of different places – and you can use open and private data together if you want to. And you can publish private data structures even when you are not sharing the data itself. There is an expectation that it only works on the web if the data is open but nothing structurally that requires that.

Q6) I read articles and go to events where BBC and Guardian very enthusiastic about linekd data. I try to persuade colleagues but still struggle – any good arguements

A6) Read a good blog post the other day – solving problems that haven’t come up yet! But if we link across private and public data or just make your data public then we can create powerful aggregations. Connecting data to make your data better. Sum better than parts

Matt: Ultimate mash up API. Systems and tool neutral so very very powerful. Lets me connect up reading lists, government data, literature extracts etc. Anything that’s useful in a single platform neutral API.

Owen: has been working on neutro? project at the OU. We have lots of audio and visual materials that are poorly represented by MARC. Fields you would want for audio visual material can go into MARC but they get a bit lost. RDF/Linked data means you don’t have to commit to a single structure/schema. You can share elements between formats say but you can represent the data in more than one way. Useful to split these things up. Don’t get page numbers where you expect video length. That’s been a really useful lesson for me.

Comment: differences between trad relational database and RDF. Table structures can be a lot of work – often has to be flattened and exported at CSV. The triple format of RDF lets you get around that and have common structures.

Burns Flash Mob at St Giles at 1pm should you wish!

Various spaces and options available. Short talks (that’ll include me) in main theatre. Also video of setting a mashup challenge with a prize. In addition to Ben’s challenge. Open Bibliography was working with the British Library. Released at CC0. 3 million records of book holdings. Challenge is to do things with this data. It’s in RDF but we have some other outputs, JSON outputs etc.

openbiblio.net has full info. Two top prizes of £500 and more prizes for ideas etc. For nice visualisations, for identifying issues in the data etc.

Owen emphasizes that there are two top prizes: one for the best mashup, one for the best concept/idea. So don’t be put off if you aren’t confident building your ideas.

And an upcoming pitch:

Pancakes and Mash

Sponsored by RLUK (very generously – hoping for funded places for student places etc)

Shrove Tuesday, 8th March 2011, University of Lincoln

£12 admin fee (and no accomodation in that!) – can book online from the event blog: http://lncn.eu/cef/. About 40 places left at the moment.

There will be social stuff and pancakes!

And finally for the morning Owen thanks sponsors for Haggis & Mash: SCONUL and JISC.

12:41: We’re off to lunch. More liveblogging later and definitely more from Open Edge – Open Source in Libraries tomorrow as well!

Lunch unexpectedly included not only tasty catering, remaining cupcakes and @jaffne’s excellent Tablet but also a delicious chocolate cake celebrating the super open library work going on brought along by JISC’s Dave Flanders (sorry for the blurry picture here – I was grabbing for a piece of cake as I snapped it!):

Matthew Phillips of Durham but formerly of Dundee will be talking about a project he undertook there with Aleph/SFX.

Migrated from Dynex to Aleph in 2004/5. Ex-Libris volunteered to convert records to MARC. Worked well for our journals with electronic and print on the same record.  Matthew is describing how SFX has been used to grab outside data on coverage, various access paths etc. Used AJAX to add very appealing visualisation/graphics to display journal coverage. Code on EL Commons. website.

Next up video from Gary Green of Voices for the Library (voicesforthelibrary.org.uk).

The campaign is also on Facebook and Twitter. They wanted to stand up for libraries. They started the #savelibraries hashtag and have had lots of support back. Lots of appreciation for libraries emerging. The challenge: Tweets are there and have been archived to. We’d like a creative way to use those tweets and mash them up to highlight why public libraries are so important. Prize is £60 Amazon voucher. Info at http://bit.ly/MarDixonMash.

Owen Stephens now talking about the Sir Louie Project

This is about Oxford, which uses SAKAI, and the library using GEAC with Primo as a UI. Most readers and lecturers have materials, reading lists etc. that they want to share. They have been doing that with Citation Helper in SAKAI. That Citation Helper was developed by universities in the US. There were standard ways to enter data to avoid mistypings etc. They set up integration with Google Scholar for instance. You click on the stuff you want to add basically. They had a connector for MetaLib but not for Primo. They wanted to search with Primo. But they really liked the Google Citation interface.

Not all students will have all the same borrowing priveledges but there is only so much one can do with that at the moment based on what the different systems make available for reuse. Oxford wanted to display availability. Firstly the task was to have a search from SAKAI that lets you search Primo and add new citations when suitable. So firstly by adding a cookie – in Primo you can pick up that cookie and display the Citation Helper. But not the best solution. Instead they named a browser window – you can fit almost 2MB of info into the window name. We wanted “it’s this list and we want to show them CitationHelper”. Then we can use jQuery that says if you see this window name then open this CitationHelper. And we use an OpenURL to search appropriate sources. This uses Juice as discussed by Matt.

There was always an SFX button in the search. Other sites may have other OpenURL resolvers there. We added COINS: a way to add that information in a SPAN tag (COINS = Context Object IN Span). So information is on the page but not shown. e.g. libx toolbar in FireFox – when you browse a page with an ISBN it adds a link to look that book up in your library catalogue (you can add other info etc).  Use Juice to read COINS (VuFind does this too) and looks for an item ID to link back to Primo – and then it allows it to be shown in the interface. Not adding lots to SAKAI, just to web interface. Using the DAIA standard to display availability (“this item: wot’ve we got?!”). Can tell you what’s there, if it’s open source, if it is available online or from the store, etc. Added Juice extension that understands DAIA. We’ve collapsed that information (via jQuery/Juice) so that you can show/hide with a button.

When Matt talked about adding functionality in the library earlier we have to remember that library information is useful in lots of other places and that increases the likelihood of sustainability.

The SAKAI stuff I’ve shown you maybe doesn’t look too pretty but it’s easy to add book covers or previews etc. It’s one line of code maybe. The availability stuff was hard as not done before but the stuff that’s been done before is easy to implement.

You can extend and reuse code and we share a lot of this in EL Commons. We grab OpenURL and paste it with the code etc. It’s easy to grab and reuse metadata etc. You can do more with Primo as full API with web services and xml etc. But what we’ve done is lightweight and should survive interface changes.

> There was a short break while I gave my own presentation on EDINA APIs – available here.

Bob Kerr of OpenStreetMap is now up talking about the project which is free and entirely user generated and maintained. There are about 85k users and about 5% contribute data. This is a data project NOT a mapping project. More info on how the data is collected, what events and projects are taking place etc. can be found on the project documentation at http://wiki.openstreetmap.org. There are also derived maps – such as the cycle map showing routes as well as pubs and toilets!

So how good and how consistent is our data? We’ve compared our data with the names of the roads in the Ordnance Survey and it’s really not bad. It’s not as good as MasterMap but we’re getting there. So roads wise. We have 850k roads, 250k missing. We think the Open Street Map will cover all roads in the UK this year. It’s moving really really fast. The OS MasterMap goes down to house detail and we’re really getting there. This is very new. We’ve just had permission to trace from Microsoft Bing images.

And there is a new lovely geeky feature: Lat and Long have been taken and turned into hex to create a shortlink (e.g. http://osm.org/go/e6aIC2Y7) and we’re looking at using that for Linked Data.

Next Up:

Richard Cross from Nottingham Trent University

A very small mash up now that can’t compare with what we’ve just seen but it is my first mash up and I figure mashed library is about celebrating mash ups of all types and scales.

Resource list item checker (RELIC). Put in ISBN/ISBN-13 then you bounce your request to Open Library (large data set) and you can populate your answers with vendors etc. Which resource lists does your ISBN appear – is it on a lot of them? What are the holdings on your catalogue for that ISBN etc?

So resource books on the library with ISBN can be looked up and you can see where it appears for instance. Nice. You can see related ISBNs from LibraryThing as well. Proper use case for this for acquisitions and subject librarians. It’s my first mash up.

Would like a clever solution for when Open Library gives no match. Would still like to take ISBN forward etc. Haven’t had time to play with Google Books preview. Want to look at OCLC xISBN service as alternative to LibraryThing. Also would like to have more on DOI and LCN – to pick up more references to items etc.

Currently being tested by Acquisitions staff and liaison staff. Hopefully it will go far!

Next up:

Nick Dimant, PTFS Europe – who is talking tomorrow – will Look at Koha, one of the Open Source library systems.

So this is one of our demo systems – a pretty vanilla Koha system. The opening screen is entirely configurable. Editing of text is simple text/WYSIWYG stuff, you don’t need HTML. So if we search for the word spin for instance. When searching you can configure all sorts of things. You can show all sorts of information as well – you can see bookjackets, there’s facet searching here, etc. Some nice stuff linking LibraryThing tag browsing etc. to catalogue. Normal but useful functionality – place on reserve, print record etc. Can also export for citation. If you use Zotero and you have it installed there is a firefox plugin that shows an extra button on the record page.

Koha also lets you create any number of RSS variants for searches, titles etc.You can add things to your cart then download or email or request etc. any of them. A link to lists of various sorts – both reading lists and most popular items (most borrowed) items etc. A nice way to explore content. And tag clouds etc. also let you take a different view on data.

So far I’ve been “joe public” but now I will login. Koha supports multiple logins as well as LDAP etc. for corporate authentication.

Once logged in you can see what’s on loan, any personal messages (e.g. you left behind an umbrella) etc. You can manage your own personal tags for items – so you can create and view and access content in custom ways. You can view your search history (forever!). You can also view your reading history (complete again) – to see what you’ve borrowed etc. You can also enable “purchase suggestions” if you want to and there are two ways you can do this – one is an email to acquisitions, the other is a formatted request for acquisitions so that it tracks and can be ordered. Nice functionality.

Q: Can you suggest things for academics only? No, not currently supported but could be done with a spare afternoon (says Colin, chief software engineer at the back of the room).

In terms of messages you can opt in/out of some messages – not required ones but early notice of items etc. Messages can be by email, RSS, Twitter etc. You can configure the medium. And the lists look rather different now – you have personal as well as global lists. And sharing options here as well – can tweet things from the system for instance. You can also get RSS of new books etc. Can place reserve etc. from links in the RSS as well.

We’ve been looking at out of the box but you can control your own stylesheet or use your corporate stylesheet to style your OPAC.

Staff client type demo of Koha:

Browser based – all the staff client through the browser. Very easy to roll out and support. Can access all modules from here. All the standard stuff. You will only see modules you have the rights to see when you login as a staff member to the client. Everything throughout the system is hyperlinked – very appealing looking interface though it looks enough like traditional systems to be familiar in terms of layout etc. And a good record of messages and alerts – handy for disputes. One user database for staff and students. if you are staff you can set up functions for other users – quite a lot of granularity here. So that was a quick look at circulation there.

Reporting is quite nice – you can look at busiest/most active users for instance. You can analyse your catalogue by item type. Lots of nice snap shots of the data. You can also get into stats wizards for the next layer of details. Lots and lots of options which basically makes SQL statements (you can paste in your own or others’ from the community reports in directly as well). Big active community so loads of reports that you can run and borrow from others. Also Crystal Reports, php etc. can all search the database (schema are all published).

Global settings very simple – half a day’s training apparently. Can change holds settings etc.

And with that we would up for coffee and chatter and demos of weird and wonderful bits of kit from Ben O’Steen and dinner plans etc. And I headed home for my own version of haggis and mash (veggis haggis and baked potato)!

Some useful links/other reports for the day (more to be added as I see them):

  One Response to “Open Edge – Open Source in Libraries Day 1: Haggis and Mash”

  1. [...] This post was mentioned on Twitter by Karen Blakeman, ostephens, Nick Sheppard, Guardian Edinburgh, Nicola Osborne and others. Nicola Osborne said: Am liveblogging a bit today btw – notes from #haggisandmash will be appearing all day here: http://bit.ly/gL5ckA [...]

 Leave a Reply

(required)

(required)

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=""> <strike> <strong>