Services/APIs/Systems/Technology/Data that we could use

[Owen Stephens]

Update: Roy Tennant has started to maintain a list of library-related APIs at http://techessence.info/apis/ (see his post below).
---------------------------------------------------------------
OK - this is just an attempt to list some useful things that people might want to look at using on the day - in no particular order:

• xISBN webservice (find all the ISBNs plus edition and other metadata related to a work)
• Worldcat Identities (get information about people/organisations listed in Worldcat)
• xOCLCNUM(find all OCLC Numbers plus edition and other data related to a work)
• Mirlyn API(University of Michigan Library Catalogue - search and retrieve metadata)
• OpenCalais (creates rich semantic metadata for the content you submit)
• ThingISBN (takes an ISBN and returns a list of ISBNs from the same "work" (ie., other editions and translations))
• ThingTitle (takes a title and returns a list ISBNs from the most likely LibraryThing "work," the LibraryThing title and a link to the LibraryThing work page)
• ThingLang (takes an ISBN and returns the language of the book)
• COPAC SRU Interface (search COPAC using SRU!)
• Library of Congress Subject Headings (LCSH as linked data using SKOS)
• Google Book Search API (link to items in GBS, and find out about their availability
• Any library system that you personally have access to an API
• There must be loads more - post any you know/care about here with a description - could be an API, a Service, a Dataset, a website - whatever you think will be useful.

[Dave Pattern]

We've got a few data streams available (OPAC keyword searches and hourly entrance figures), and I could certainly cobble together some more feeds and APIs.

Pewbot (which gives access to our borrowing suggestions) is also up and running again: http://www.daveyp.com/blog/index.php/archives/69/

There's also Ann Arbor's REST interface to their library data, which I blatently ripped off: http://www.daveyp.com/blog/index.php/archives/59/

[Chris Keene]

That's a good list!

I would add:

• Talis Platform (place to store RDF data and a ukbib store containing a lot of bib records, a crappy/broken example site using the latter can be found here )
• Wikipedia book sources (not sure how useful it is, but perhaps lists other resources we could use)
• Amazon Web services (well someone's got to mention them!)
• OAI-PMH / OAI-ORE / SWORD (protocols rather than specific services, and don't want to turn this is to a repository day, as groups like CRIG are already doing good stuff).

[Richard Wallis]

And of course the Talis Platform.

The documentation & APIs and development community forming around the use of Platform stores can be found on the N² Wiki.

For the kind of projects envisaged we can provide Platform stores both for somewhere readily available and visible to put/search data and for experimentation with some of the semantic web techniques such as sparql.

[Owen Stephens]

Thanks Richard - I was going to ask!

Would you be able to provide a Platform store with some data populated for the day (e.g. some bib records)?

Owen

[Owen Stephens]

For those who find the idea of sparql quite scary (like me), worth mentioning that there are also some simpler techniques (I think - perhaps Richard can confirm?) where you can push stuff 'through' the store using stuff like the 'Augment' service, which adds allows you to enhance (for e.g.) a result from one service, with stuff like image data or holdings data - which all looks a bit simpler than learning Sparql (these services also documented at the URL Richard gives above)

Perhaps that's just me being a scaredy cat.

[Owen Stephens]

Anybody know of a VLE or Learning Environment interfaces we could use? We could setup a Moodle instance for the day perhaps (or individuals could run it on their laptops)?

[Chris Keene]

well, not sure if this is useful (and sure someone can come up with something more useful) but one of the nice things with dreamhost is that they offer a oneclick install for Moodle. I did my one click a few months a go and it sits here: http://www.nostuff.org/moodle/. No content, but we can break it to our hearts content.

oh and I don't find sparql quite scary. I find it very scary :)

[Dave Pattern]

Not library specific, but our HIPpie spellchecker is fairly stable and can provide XML output (I'm also planning to add JSON soon): http://library.hud.ac.uk/wikis/hippie/index.php/Spell_checker

[Ben O'Steen]

I think it would be quite useful to have a couple of Apache Solr (search/faceting made easy) instances ready to roll - with a decent bibliographic set of fields set up. Additional fields can be added on the fly too.

Maybe a laptop/server/EC2 instance could be sourced for this? (Maybe an on-site device, as wifi could be flaky)

Who doesn't need a quick way to cache, facet and search through metadata when mashing up ugly services? ;)

[Ian Ibbotson]

Wotcher Ben, all.

I'll be along with a shrink-wrapped solr service that can index sword deposited metadata and expose over sru/z3950 and some other new OAI tools I'm mucking about with. Happy to leave the std solr urls open so people can throw records at it for indexing if thats helpful. Also have the shrink-wrapped meta-search portal if people want to plug in datasources to see along-side existing library resources.

[PeteJ]

Various Linked Data sources e.g.

DBpedia http://wiki.dbpedia.org/OnlineAccess
DBTune http://dbtune.org/

For getting started/playing with SPARQL, Leigh Dodds' Twinkle client is quite nice http://www.ldodds.com/projects/twinkle/

[Roy Tennant]

With Owen's permission I've taken his beginning list, as well as some of the items contributed in comments here, and created a list that I hope can be a somewhat permanent list to help unconferences such as this and other events happening elsewhere. It is on my TechEssence.info site at http://techessence.info/apis . Please let me know if you think something else should be on it. I'm trying to be somewhat selective but I'd like to include things useful to library developers. Meanwhile, Owen and I agree that any of these not already in the directory at http://programmableweb.com should be as well, to increase exposure of these APIs beyond the library community.