Monday 26 April 2010

emtacl10 - Day 1

I may just blog selected sessions here, rather than all the ones I go to, but here goes for my (almost) live blog (so expect lots of typos and random ramblings) of emtacl10...

Mmm, lovely dinner and nice little delicate cakes for pud!

A welcome, etc., as you may expect, then into the first keynote:

Lorcan Dempsey - "The network has reconfigured whole industries. What will it do to academic libraries?: How libraries are going to have to change their services in the networked future. Has 64 slides to get through! Example of Netflix - knows we aren't in the US, so can't access it. This has changed the way people view films in America. Showed example of recommendations. Netflix however has recently moved much of it's infrastructure to a competitor - Amazon. Their core competence is recommending films - not interested in running a data centre that has to sit behind it, so happy for Amazon (a competitor in delivering films) to do that part of it.
Network encourages efficiencies of scale. Can analysise data much better when you are larger in scale, deliver better what people want. At a network level, makes stuff easily referenceable - use big networks as place to refer others to.
Lets us specialise where we can make an impact - outsource other stuff to other people.
Institutional scale issues? Every library duplicating effort at the moment - not specialising where we should / can make impact. So should library services be externalised? National? Collaborative? Third Party? Or should they be given up?
Sourcing /Scaling very important over recent history, so JISC for UK negotiating stuff, collectively letting us buy things, rather than individual libraries trying to do the same.
Talk about 4 things within an institution level (Discovery; reputation mgt, citation mgt, the whole social thing)
But first, collections! These influence how we think about our other services. Traditionally we;ve looked after scarce things (books) that we need to look after carefully.
Discover happens elsewhere - example of searching Google Books to find a particular passage cf flicking through a book. The norm is now online, external, no longer limited to physical access and familiarity. "We are no longer limited to the 150 people...." of our real life networks, we can scale up to the whole web. So we need to make sure our special, "unique" collections are easily discoverable by people who don't want to use our tools, but use Google, etc. instead. i.e. "Indirect discovery". "Disclosing" stuff to cloud services to allow this. Example of putting a link to a resource (article) in a wikipedia page as a way of SEO (though this was disputed by someone on twitter immediately!).
Why mess about with making nice catalogues, etc., when most discovery will happen through other tools on the web?
Have some scalar confusion here - things happening at all sorts of different levels, "presences" all over the place. Should we have lots of pages, or one thing that can feed into other services?
Also have scope confusion - what are the motivations for each presence and what are we aiming for?
We don't have a controlled environment, the "network" throws all these together.
Social stuff: (context - community - conversation. Return on attention). Manage demand not manage supply. Make people aware of what you have, not enough to put stuff on the web, you need to push things forward that people may be interested in.


Some directions - Thinks we will get rid of most print books except the latest most used stuff. Move to regional stores. 80% or more spending on materials will be licenced electronic content from a few big suppliers. Will have selective & targeted kocal engagement around scholarly materials.
Moving infrastructure to the cloud. complex systems will have to be simplified. Need to realise network effects in this move (recommendations, collaborative collection mgt, etc.).
Shift from space as necessary storage and infrastructure to social, Ad Hoc rendevoiuz space, showcase for particular expertise.
Relationship mgt - Roles is to make informational aspects of reserch and learning more efective. Need to "peel" expertise away from the collection to apply it to wider tools, practices and services. "People are entry points". Make expertise accessible and visible. Not exposes our expertise at the moment, we push to resources more than our own expertise. "If the library wants to be seen as expert, then it's expertise has to be seen".
Doesn;t like the phrase "Info Lit" (boo to him!), prefers "Scholarly literacy".
Has finished about 35 minutes late - sure the fire alarm wasn't that long!


Social Networks strand:

"New Applications derived from behaviour based recommendations" Spiering and geyer-Schulz: Talking about recommendations based on what users do, not what they say...
Sounds like BibTip is service they are talking about.
Bibtip is based on analysis of co-inspections (full views within same browser session). Recommendations reflect local user behaviour, not wider material. Language independant, any media in catalogue, etc. Purely browser history. Ex Libris have installed this on their cat system.
20-25% of user on any particular day click on recommendations. Sounds a bit dodgy to me - how do we know the pages people are viewing are useful? It may be that everyone clicks on the same rubbish book because it comes high up in the list of search results - this system will then recommend it to others, re-inforcing that choice as a recommended book? Or am I wrong?
Started saying it was local only - now saying they are bringing in sharing of data between libraries, but with priority to local recommendations. So (I think) recommendations from elsewhere only come in when local data not populous enough to generate recommendations.


"Mashup of REST-ful APIs only using RSS feeds to support research in a high demanding research environment" Chumbe: Telling us all about mash-ups. Using www.journaltocs.hw.ac.uk/api (is this the draft version of http://www.tictocs.ac.uk/ ? No, Dave Pattern tells me it's a seperate JISC funded project that builds on tictocs - no wonder it looked so similar, thanks Dave!) to make a friendly interface for staff to use.
Example - alerts for a repository manager when authors from their institution have published.
2nd Example - cross refs RSS feeds from publishers with university subscriptions. So, stuff you may be interested in (on a topic) that we subscribe to.
Lots of bits and pieces you can do with this.
Not sure I like this. Mainly seem to be turning RSS feeds into a search for latest stuff - shouldn't our search tools be doing this anyway? Perhaps that is why I feel a bit uneasy...


Lightening talks! It seems no-one put their name down for one on the first day!


"To face or not to face" Novosel - Nice to see someone from Croatia (Zagreb Uni). Largest Croatian Uni, 6,500 students. They have 28 subject librarians. That's 2 and a half times the number we have with 4 times the students. What are we doing wrong? Sorry, slipping into a rant about staffing there...
Talking about using Facebook (most popular social network in Croatia) - she suggests they are up to 10 years behind some other countries, so Facebook still up & coming there. Most academic libraries in Croatia lack even fairly basic web presence.
Put news type items from the wider country / world on their own wall, along with links to other pages - using it to push useful links / links to people? First 5 months went from 0 fans to 1,100 fans (now 1,400) - sounds like they really got in at the right point to surf the peak of interest in facebook in their country.
Dissapointed that people aren't using it as social space, just as a feed of information. There is probably more interaction going on than on our page though :-)



"Another look at personalisation" Haya & Stattin- Don't know what this one is about, slightly annoying I can't find abstracts anywhere!
Ah! Personalising library websites - inspired by BBC? Their "old" website looks quite nice at first glance! Now using Drupal to allow people to personalise their new pages - use Shibboleth to log in. This means they have all the associated info about groups, etc., people belong to. Their webpage now like how I think our portal should be? Can move stuff around, add elements to a pages, etc. Looks really nice. Have a search box that stays fixed in place, most other stuff (underneath) changeable. Don't care if people don't customise as "standard" site is still better than old one! Though nice if they do... Will be carrying on working on it to improve things.

That's the end of the first days sessions. (Organ?) concert in the cathedral is on offer tonight, followed by a reception of some sort.

2 comments:

  1. Ringtones plays a vital role in mobile downloading. Downloading cheap and free ringtones is one of the more popular ways we brand a phone to ourselves and identify our friends when they're calling.

    ReplyDelete
  2. I might be wrong, but I think JournalTOCs is a separate (JISC funded?) project that builds on the initial work done by ticTOCs.

    ReplyDelete