Connecting Librarian

QR codes: do they provide the missing link between the physical and the digital? – Tristan Badham

Received the VALA Travel Scholarship to see how QR codes were being used in libraries, their implementation, reception from users and staff, further ideas etc.

QR codes are 2 dimensional barcode when scanned by a mobile device, you get linked to the resource the creator intended. Could be website, email address, phone number, coordinates on a map. You need a mobile device, with a camera, an Internet connection and QR code reader (an app) to make use of QR codes.

Being used a lot in advertising, real estate signs etc.

How they can benefit libraries:

• act as a bridge between the physical and the digital

• make access to information and resource easier

Providing information at the point of need:

• video guide on how to use the print management system

• a map of the library layout

• library audio tours

Catalogue records, with links to specific location information

QR codes within the collection, to link to online resources – particularly to the mobile version of them.

Social media – can communicate your library presences.

Contact the library and research help – particularly SMS reference services.

Other non-library uses included:

• videos by curators talking more about the art work being viewed

• Powerhouse Museum has built an app which the QR code refers to when scanned – each specific piece links to a specific page

Positives:

• Cheap and relatively easy to implement – time is in the staff and sign creation – also have good tracking

• Marketing appeal – makes the library look tech savvy

Negatives:

• Device can’t read the code – too big or too small or incorrect lighting

• Used the wrong way

• People don’t know what they are or how to use them

• Devices need a QR code reader which means downloading an app….

The more content your destination has, the more complicated the QR code, which loses pixilation in resizing. If that is the case, use a URL shortener first.

So are they really all that good?

• They are worth exploring as a useful technology, especially within the broader context of the move to mobile technologies

• one tool among many

• could complement rather than compete with other technologies

Awareness of the codes is high amongst young people, even if they don’t know what they are called.

QR codes could burst on the scene, but if they are used in the wrong way, they may disappear. (bit like Brendan Fraser as an actor).

Hacking the nation: Libraryhack and community-created apps – Margaret Warren and Richard Hayward

LibraryHack was created to foster re-use library data – a direct result of the NSLA Re-imagining libraries vision. QSL was responsible for Project 5 – community created content. It aims to make real the ability to help people to find, remix and create new content. Library hack was in four parts.

        1. Release of library data and digital content for re-use.

Data was to be made available on data.gov.au, so to ensure the data was discoverable where other public data was available and to add a presence for cultural data. All ten participating libraries placed their data in this central location. Fifty-three datasets were added, primarily images, but also search transaction logs, music and art. Data was able to be licensed for re- use, using Creative Commons. Copyright is an important consideration. Discovered that having geo-spatial data included, made the data more popular and re-usable and that most library formats are not re-user friendly. If we want to encourage more photo mash-ups, we need to make high resolution images publicly available.

Interestingly, Ancestry has taken on the public data and made good use of it.

2. Ideas competition

Discovering the sorts of things that people would be interested in

3. Hack days

Days for people interested in working with the data, to come and talk to the content specialists and to find out more about the datasets.

4. Learning

Offered a range of learning opportunities, focused on different topics, including animation and more on how to mash-up this data. Videos are still available at QUT for anyone who is interested.

Received 168 entries for the competition, as well as people creating new apps that were never entered into the competition.

Judging criteria: use of data/digital content, originality, quality, usefulness. Judging panel came from NSLA libraries.

Ideas category winner – Discovery by Diana Iles – included maps, images, manuscripts and map overlay integration. It delivered a visual message, but can be interactive when properly encoded with geo-spatial data.

Apps category winner – talking maps by Michael Henderson – walking West End multimedia tour (Brisbane suburb), custom built geographic interface, talkingmaps.com website, can listen to audio and explore images on the walk

Photo mashups category: Reflection of Time by Andrew Young – included historical images, with reflection of the artists own original work of a contemporary version of the same scene incorporated into it.

Digital media mashup category – Glorious image viewer by Mark Balandzic – projection of historical images on a variety of rotation lamps.

Collaboration was the key, between hackers and between them and the library. Mostly it was fun.

Also resulted in great staff engagement.

Next: More. Better. Easier. Collaboration.

Harvesting and semantically tagging media releases from political websites using web services – Peter Neish

Why are they interested in media releases?

• Play an important part in political process

• establish a party’s position on an issue at a particular time

• often used in time urgent reference request

• may go back many years (library has database back to 1992)

Number of political media releases released in Victoria has risen from just over 1000 in 1992, to over 6000 in 2009 and 5000 in 2010. The government puts out a lot more media releases than the opposition. The government keeps it own databases of these media releases. If it was online, the library stopped duplicating that work.

Due to the potential loss of this data when a change of government occurs, the decision was made to begin harvesting this data on the go. The aims of the project were to automate the process, combine the different databases together and to examine the possibility of automatically applying tags to media releases using web services.

Part 1 – Automation

• Key was RSS

• Political parties have websites, which had RSS feeds, which were used as a standard input to software.

• Built, in Java, a servlet which polled and returned the data from the political parties website – put the full-text and its associated metadata into the library database. It also produced and saved a pdf version of the media release.

It works, having harvested over 11000 media releases since July 2010, freeing up 2 days of staff time per week. Problems include having non-standard content in feeds (eg. dates), which they addressed with Yahoo Pipes and website’s changing their structure or CMS.

Part 2 – Semantic tagging

Manual tagging was no longer viable. After examining many options, went with Open Calais – from Thomson Reuters. Although business focused, it matched up with the type of data they had, gave a good number of tags (around 20), minimal false matches, good documentation sand community and generous limits on API calls. Unfortunately, their algorithm is a closely kept secret and not as much development is happening. Check out an example at http://viewer.opencalais.com/.

User Interface – did some useful user testing which helped inform the creation of the interface.

Review – of tagging – about 85% were correct – 4% were incorrect, 6% repeated and 5% redundant. One of the things they always got wrong was Victoria which it placed in the Seychelles – very frustrating.

Linked Data – get the info back in JSON and RDF. It links to its own ontology – which means that limited classes for government.

Media releases are now available as they are released – no backlog. Data is enriched by tagging and in future will link to other databases in the Linked Data ecosystem.

 

I’ve been pondering again. So I’m going to inflict it on you.

We have recently added Chili Fresh reviews to our catalogue.  I like it, its easy to use, easy to add reviews, we have the weight of Chili Fresh reviews from around the world to populate our catalogue and they have some really cool social networking features coming soon.

But Chili Fresh doesn’t have user tagging (at least not yet).  Which of course got me thinking about tagging.

A common problem experienced in public libraries and I’m sure in other libraries too, is when you have a user come up and say something like:

“Can you help me, I’m looking for a book that I have had before. Its about gardening, its green and its about this big (demonstration using hands).”

They can’t remember anything about the author, but sometimes they can remember more about the content.  In the above type of example, it would be something about vegetables.

Barring miraculous circumstances, (like you have read that same book), or the luck of finding said book on shelf or trolley in roughly the place you would expect it to be, the likelihood of finding it with that information alone is nigh impossible.

So my reasoning was that if users could tag our catalogue records with that sort of information, it we be of great use to both them and us in finding that same title in future.

A few problems with my reasoning as I pondered further.

First would be getting the users to tag the details in the first place. Although we have a few people (more than we expected) putting reviews on our catalogue, it is nowhere near critical mass.

Second, do you know how many green books about gardening, let alone vegetables, we have?

And finally, the piece de resistance.  You finally find that book that the user was so desperate for and the only thing they got right was that it was about gardening. As for the rest of it, the subject was hydroponics, the book was orange and it was a very different shape and size.

Now only if we could tag each item by its actual details as well as its perceived details, we might have something.  But by then, the tags will take up more than a screen of detail and would probably send the catalogue search feature into meltdown.

When we do get tagging (and we will somehow, someday), we won’t stop our users from adding this sort of information, but I guess only experience will be able to tell us if it will be of any help.

What sort of ideas have you had about new technologies, which might not work so well in a library situation?  Maybe we can help you resolve your problems around it. Would also love to hear your thoughts on this one.

Came in late, so I missed the start while I set up. Library Thing has a page for every work. Shows how many people share the book in their collections, how many reviews have been written, the top 32 tags that have been attached and more.

Has a recommendation feature. Those who have this also have that, similar tags, ratings, all the different versions of the covers that people have uploaded.

Library Things treat authors as an entity as important as a book, so the same features. Related tags, pictures uploaded by members, links, similarly tagged, also known as (ie. In different countries)

Tags – have own pages, what has been tagged, who has tagged with it, related tags – could be used as a reading list for a particular category. Eg paranormal romance has 1800 tags, yet it falls through traditional indexing systems.

Recommendations feature, includes a Unsuggester feature – if you like this, you will not like…… Using the data in the system in a multitude of ways.

What does it mean for social networking – not a niche site, as books are the centre of life for many, a Second Life for readers is books! Amazon is not the best book site forever, nor is Google Books. Books are not just items of commerce. Library Thing is what people kept, what books remain with people even after they no longer have it physically. Library Thing is about our identity which has been influenced by the books we have read. Books are objects of obsession.

Tags – every item is tagged. The bolder and larger it is, the more times it has been assigned that tag. Click on that tag and you will link to all titles that have that tag assigned. Tim did a comparison between the Library of Congress subject headings and Library Thing tags – very amusing and very demonstrative of the power of tags – not that he advocates that we no longer use subject headings, instead use both!

There is now Library Thing for libraries. Top is normal ILS data, but then all social networking tools are from Library Thing, included recommended titles, but only to titles that you hold. Integrates seamlessly to the back end of your ILS. This is not the solution – Tim says that in about 10 years we will be able to get all that we want from our vendors – getting it as XML if we want, and probably coming from Library Thing. Library Thing will offer these enhancements in the meantime.

Libraries need an OCLC type group for user generated data, which can be collected by libraries and then shared between libraries.

I know this is not blogging the Online Conference notes, but I had to take a break from my endless notes! So its back to Learning 2.0.

I had looked at Del.icio.us a while ago and although I have a lot of bookmarks, I didn’t want to share them. I use a lot of bookmarks, but they are all at home and link to things like banking, games etc. However, whilst I was distracted by other things recently (like a conference, organising a paper and a study tour etc), I was thinking about all the websites I come across (usually through RSS feeds) that I can’t use right now, but I know will be of use in the future.

Its been quite a few weeks since I did week 5 of this course, but I knew it was about del.icio.us, so finally those two pieces clicked in my mind and I thought why not use it to create a list of these useful websites, so that I don’t lose them. In discussing them with my computer programmer husband, we also talked about the possibilities of having topic bookmarks organised by topic and potentially private bookmarks in del.icio.us. So from here its explore and find out more!

As for Technorati, as a long term blogger I am well aware of this fabulous tool. But as with everything I have learnt from this Learning 2.0 program, there is always something new to learn, even from things you may already know.