JISC Open Biblio 2 project – final report

**This is cross-posted from [openbiblio.net](http://openbiblio.net/2012/08/23/final-report-jisc-open-bibliography-2/).**

Following on from the success of the first [JISC Open Bibliography project](http://openbiblio.net/2011/06/30/final-product-post-open-bibliography/) we have now completed a further year of development and advocacy as part of the [JISC Discovery programme](http://discovery.ac.uk/).

Our stated aims at the beginning of the second year of development were to show our community (namely all those interested in furthering the cause of Open via bibliographic data, including: coders; academics; those with interest in supporting Galleries, Libraries, Archives and Museums; etc) what we are missing if we do not commit to Open Bibliography, and to show that Open Bibliography is a fundamental requirement of a community committed to discovery and dissemination of ideas. We intended to do this by demonstrating the value of carefully managed metadata collections of particular interest to individuals and small groups, thus realising the potential of the open access to large collections of metadata we now enjoy.

We have been successful overall in achieving our aims, and we present here a summary of our output to date (it may be useful to refer to this [guide to terms](http://openbiblio.net/2011/12/13/what-is-bibserver/)).

## Outputs

### BibServer and FacetView
The [BibServer](http://bibserver.org/) open source software package enables individuals and small groups to present their bibliographic collections easily online. BibServer utilises [elasticsearch](http://www.elasticsearch.org/) in the background to index supplied records, and these are presented via the frontend using the [FacetView](https://github.com/okfn/facetview/) javascript library. This use of javascript at the front end allows easy embedding of result displays on any web page.

### BibSoup and more demonstrations
Our own version of BibServer is up and running at , where we have seen over 100 users sharing more than 14000 records across over 60 collections. Some particularly interesting example collections include:

* [Cambridge Physics Tripos](http://bibsoup.net/edchamberlain/cambridge_physics_tripos) – a collection of 234 records extracted from a physics department MS Word reading list
* [Adrians bibsonomy bibliographie](http://bibsoup.net/acka47/adrians_bibsonomy_bibliographie) – a collection of 338 records extracted directly from Bibsonomy
* [Testing philosophy](http://bibsoup.net/epoz/testing_philosophy) – a collection of 21 records extracted directly from Wikipedia via the Wikipedia search collection builder

Additionally, we have created some niche instances of BibServer for solving specific problems – for example, check out ; here we have used BibServer to analyse and display collections specific to malaria researchers, as a demonstration of the extent of open access materials in the field. Further analysis allowed us to show where best to look for relevant materials that could be expected to be openly available, and to begin work on the concept of an Open Access Index for research.

Another example is the German National Bibliography, as provided by the German National Library, which is in progress (as explained by Adrian Pohl and Etienne Posthumus [here](http://openbiblio.net/2012/06/18/bringing-the-open-german-national-bibliography-to-a-bibserver/)). We have and are building similar collections for all other national bibliographies that we receive.

### BibJSON
At we have produced a simple convention for presenting bibliographic records in JSON. This has seen good uptake so far, with additional use in the [JISC TEXTUS](http://textusproject.org/) project and in [Total Impact](http://total-impact.org), amongst others.

### Pubcrawler
[Pubcrawler](https://bitbucket.org/sea36/pubcrawler) collects bibliographic metadata, via parsers created for particular sites, and we have used it to create collections of articles. The [full post](http://openbiblio.net/2012/06/13/pubcrawler-finding-research-publications/) provides more information.

### datahub collections
We have continued to collect useful bibliographic collections throughout the year, and these along with all others discovered by the community can be found on [the datahub](http://thedatahub.org) in the [bibliographic group](http://thedatahub.org/group/bibliographic).

### Open Access / Bibliography advocacy videos and presentations
As part of a Sprint in January we recorded [videos](http://openbiblio.net/2012/01/24/sprint-videos/) of the work we were doing and the roles we play in this project and wider biblio promotion; we also made a how-to for using BibServer, including [feedback from a new user](http://openbiblio.net/2012/01/25/bibserver-screencast-and-user-perspective/):

Setting up a Bibserver and Faceted Browsing (Mark MacGillivray) from Bibsoup Project on Vimeo.

Peter and Tom Murray-Rust’s video, made into a [prezi](http://prezi.com/), has proven useful in explaining the basics of the need for Open Bibliography and Open Access:

### Community activities
The Open Biblio community have gathered for a number of different reasons over the duration of this project: the project team met in [Cambridge](http://openbiblio.net/2012/01/19/thursday-19th-january-open-biblio-sprint-day-3/) and [Edinburgh](http://openbiblio.net/2012/03/14/day-3-of-the-march-sprint/) to plan work in Sprints; [Edinburgh](http://openbiblio.net/2012/03/14/opendataedb-the-results/) also played host to a couple of Meet-ups for the wider open community, as did [London](http://openbiblio.net/2012/06/13/bibliohack-meet-up/); and London hosted [BiblioHack](http://blog.okfn.org/2012/07/09/bibliohack-ed/) – a hackathon / workshop for established enthusasiasts as well as new faces, both with and without technical know-how.

These events – particularly BiblioHack – attracted people from all over the UK and Europe, and we were pleased that the work we are doing is gaining attention from similar projects world-wide.

### Further collaborations

* [TEXTUS](http://textusproject.org/) wants to integrate [BibServer FacetView](http://bibserver.org/2011/11/11/recent-updates-to-bibserver-adjust-your-facets/) and add a [‘TEXTUS’ field to BibJSON](http://textusproject.org/2012/06/21/bibliographic-references-in-textus/) – this is ongoing but [work to-date](http://beta.openphilosophy.org/) is available as a prototype;
* [Public Domain Works](http://www.publicdomainworks.net/) produced the [Open Metadata Handbook](http://en.wikibooks.org/wiki/Open_Metadata_Handbook) in collaboration with the Open Bibliographic Data Working Group;
* Mike Jones used his mobile application, [M-Biblio](http://mbiblio.ilrt.bris.ac.uk/), to deposit metadata to BibServer during the Hackathon – he writes about the trials and successes [here](http://mbiblio.ilrt.bris.ac.uk/2012/07/16/bibliohack/);
* [OSS Watch](http://www.oss-watch.ac.uk/), the JISC-funded organisation, looked at our project output to monitor good open-source standards. We shared their results [here](http://openbiblio.net/2012/05/29/open-source-development-how-we-are-doing/);
* [ServiceCORE](http://core-project.kmi.open.ac.uk/about-servicecore-project) tells you if a record within BibServer is available from any UK repository – we link directly to the results from our record pages.

## Lessons
Over the course of this project we have learnt that open source development provides great flexibility and power to do what we need to do, and open access in general frees us from many difficult constraints. There is now a lot of useful information available online for how to do open source and open access.
Whilst licensing remains an issue, it becomes clear that making everything publicly and freely available to the fullest extent possible is the simplest solution, causing no further complications down the line. See the [open definition](http://opendefinition.org/) as well as our [principles](http://openbiblio.net/principles/) for more information.

We discovered during the BibJSON spec development that it must be clear whether a specification is centrally controlled, or more of a communal agreement on use. There are advantages and disadvantages to each method, however they are not compatible – although one may become the other. We took the communal agreement approach, as we found that in the early stages there was more value in exposing the spec to people as widely and openly as possible than in maintaining close control. Moving to a close control format requires specific and ongoing commitment.

Community building remains tricky and somewhat serendipitous. Just as word-of-mouth can enhance reputation, failure of certain communities can detrimentally impact other parts of the project. Again, the best solution is to ensure everything is as open as possible from the outset, thereby reducing the impact of any one particular failure.

## Opportunities and Possibilities
Over the two years, the concept of open bibliography has gone from requiring justification to being an expectation; the value of making this metadata openly available to the public is now obvious, and getting such access is no longer so difficult; where access is not yet available, many groups are now moving toward making it available. And of course, there are now plenty tools to make good use of available metadata.

Future opportunities now lie in the more general field of Open Scholarship, where a default of Open Bibliography can be leveraged to great effect. For example, recent Open Access mandates by many UK funding councils (eg [Finch Report](http://www.out-law.com/en/articles/2012/july/government-details-open-access-funding-and-publishing-policy-for-taxpayer-funded-research/)) could be backed up by investigative checks on the accessibility of research outputs, supporting provision of an open access corpus of scholarly material.

We intend now to continue work in this wider context, and we will soon publicise our more specific ideas; we would appreciate contact with other groups interested in working further in this area.

## Further information
For the original project overview, see ; also, a full chronological listing of all our project posts is available at . The work package descriptions are available at , and links to posts relevant to each work package over the course of the project follow:

* [WP1](/tag/wp1) Participation with Discovery programme
* [WP2](/tag/wp2) Collaborate with partners to develop social and technical interoperability
* [WP3](/tag/wp3) Open Bibliography advocacy
* [WP4](/tag/wp4) Community support
* [WP5](/tag/wp5) Data acquisition
* [WP6](/tag/wp6) Software development
* [WP7](/tag/wp7) Beta deployment
* [WP8](/tag/wp8) Disruptive innovation
* [WP9](/tag/wp9) Project management (NB [all posts](/tag/jiscopenbib2) about the project are relevant to this WP)
* [WP10](/tag/wp10) Preparation for service delivery

All software developed during this project is available on open source licence. All the data that was released during this project fell under OKD compliant licenses such as PDDL or CC0, depending on that chosen by the publisher. The content of our site is licensed under a Creative Commons Attribution 3.0 License (all jurisdictions).

The project team would like to thank supporting staff at the [Open Knowledge Foundation](http://okfn.org) and [Cambridge University Library](http://www.lib.cam.ac.uk), the OKF Open Bibliography working group and Open Access working group, Neil Wilson and the team at the [British Library](http://bl.uk), and Andy McGregor and the rest of the team at [JISC](http://jisc.ac.uk).