My talk about general patterns in data was recieved well and I got some helpful input. I will write about it later. Steffen Hennicke, another PhD student of my supervisor Stefan Gradman, then talked about his work on modeling Archival Finding Aids, which are possibly expressed in EAD. The structure of EAD is often not suitable to answer user needs. For this reason Hennicke analyses EAD data and reference questions, to develope better structures that users can follow to find what they look for in archives. This is done in CIDOC-CRM as a high-level ontology and the main result will be an expanded EAD model in RDF. To me the problem of “semantic gaps” is interesting, and I think about using some of Hennicke data as example to explain data patterns in my work.

The last talk by Rita Strebe was about aesthetical user experience of websites. One aim of her work is to measure the significance of aesthetical perception. In particular her hypothesis to be evaluated by experiments are:

H1: On a high level, the viscerally perceived visual aesthetics of websites effects
approach behaviour.
H2: On a low level, the viscerally perceived visual aesthetics of websites effects
avoidance behaviour.

Methods and preliminary results look valid, but the relation to digital libraries seems low and so was the expertise of Strebe’s motivation and methods among the participants. I suppose her work better fits to Human-Computer Interaction.

After the official part of the program Vladimir Viro briefly presented his music search engine peachnote.com, that is based on scanned muscial scores. If I was working in or with musical libraries, I would not hesitate to contact Viro! I also though about a search for free musical scores in Wikimedia framework. The Doctoral Consortium ended with a general discussion about dissertation, science, libraries, users, and everything, as it should be

I soon realized that E-LIS uses a very old version (2.13.1) of GNU EPrints – EPrints 3 is available since December 2006 and there have been many updates since then. To find out whether it is usual to run a repository with such an outdated software, I did a quick study. The Registry of Open Access Repositories (ROAR) should list all relevant public repositories that run with EPrints. With 30 lines of Perl I fetched the list (271 repositories), and queried each repository via OAI to find out the version number. Here the summarized result in short:

76 x unknown (script failed to get or parse OAI response), 8 x 2.1, 18 x 2.2, 98 x 2.3, 58 x 3.0, 13 x 3.1

Of 195 repositories (that I could successfully query and determine the version number of) only 13 use the newest version 3.1 (released September 8th). Moreover 124 still use version 2.3 or older. EPrints 2.3 was released before the web 2.0 hype in 2005! One true point of this web 2.0 bla is the concept of “perpetual beta”: release early but often and follow user feedback, so your application will quickly improve. But most repository operators do not seem to have a real interest in improvement and in their users!

Ok, I know that managing and updating a repository server is work – I would not be the right guy for such a job – but then don’t wail over low acceptance or wonder why libraries have an antiquated image. For real progress one should perpetually do user studies and engage in the developement of your software. Digital libraries with less resources should at least join the Community and follow updates to keep up to date.

P.S: E-LIS has updated its software now (November 2008). A lot of missing features remain but those need to be implemented in EPrints first.

The DLRM is “defined” in a 213-page PDF-file – this is just not usable! Have a look at W3C or IETF how nowadays standards are defined. As DLRM is a conceptual model, you must also provide a RDF representation or it is just inavailable for serious applications on the web. And of course the silly copyright statement should be removed in favor of a CC-license. That’s the formal part. Summarizing the content of DLRM there are 218 concepts and 52 relations – which is far too much to start with. But: there are some really useful ideas behind DLRM.

The reference model includes a division of “digital library” into three levels of conceptualization (see image): first the visible digital library (the collection that users work with), second the digital library system (the software application that is installed and configured to run a digital library), and third the digital library management system (the software system that supports the production and administration of digital libraries). This division can be helpful to understand and talk about digital libraries – although I am not sure whether the division between digital library systems and digital library management systems is a such a good one.

Beside general talks about the Digital Library Reference Model the workshop provided some experience from practise by Wolfram Hostmann (DRIVER project) and by Georg Eckes (Deutsches Filminstitut) – never underestimate good real world examples! The most refreshing talk was given by Joan Lippincott (Coalition of Networked Information). She pointed out that much more then traditional repositories can be viewed as digital libraries. Especially user-generated content can constitute a digital library. A useful model for digital libraries should also fit for collections at Flickr, YouTube, Wikis, Weblogs etc. and user can mash up resources to create new digital library services, for instance the species search engine iSpecies. She is sooo right! In addition Joan mentioned initiatives to broaden the use of authority files and identity management. Another direct hit! If digital libraries only focus on interoperability with other “official” digital libraries they will not remain. Libraries are only one little player in the digital knowledge environment and their infrastructure is not defined only by them.

I enjoyed the workshop, I really like the digital library community and I am happy to be part of it. But some parts still seem to live in an ivory tower. If the digital library reference model does not quickly get adopted to real applications (both repositories like those build with DSpace, Fedora, EPrints etc. and open systems like YouTube, Wikipedia, Slideshare…), it is nothing but an interesting idea. The digital revolution is taking place anyway, so let’s better be part of it!

P.S: The slides will soon be available at the Workshop’s website.

Ich halte es zumindest für wichtig, in etwa auf dem Laufenden zu sein und bei Bedarf zu versuchen, Einfluss zu nehmen, falls die Entwicklung völlig an den eigenen Bedürfnissen vorbeigeht. So ganz verstanden habe ich Jangle allerdings bisher noch nicht. Statt eine neue Gesamt-API zu entwickeln, sollte meiner Meinung nach besser bei den existierenden Schnittstellen aufgeräumt werden – anschließend können diese dann ja in Jangle o.Ä. zusammengefasst werden.

Larry stellt Jangle und Primo gegenüber: auf der einen Seite der systematische OpenSource-Ansatz, in den man sich erstmal einarbeiten muss und der dafür insgesamt effizienter und flexibler ist und auf der anderen Seite das teure, unflexible Gesamtprodukt, das dafür schick aussieht. Leider geht es bei den Entscheidungsträgern meist primär um die Fassade, so dass sich OpenSource nur langsam (aber sicher) durchsetzt.

But if you see Europeana less as a monolithical project but as a network of participants (libraries, museums, and archives) that try to agree on standards to improve interoperability, the attempt seems more promising. In his introductionary speech Jill Cousins (director of Europeana) stressed the importance of APIs and ways to export information from Europeana so other institutions can build their websites and mashups with services and content from Europeana. I hope that open content respositories like Wikimedia Commons and Wikisource can act as both as source and as target of exchange with the European Digital Library. What I also found interesting is that Google and Wikipedia are seen as the default role models or at least important examples of portals to compare with. The talks I have seen so far give me the impression that the view on possibilities and roles for libraries, museums, and archives are more realistic than I thought. One of their strength is that they hold the content and are responsible for it – in contrast to user generated collections like YoutTube, Slideshare, and (partly) Wikipedia. But in general cultural institutions are only one player among others on the web – so they also need to develop further and “let the data flow”. If the Europeana project helps in this development, it is a good project, no matter if you call it a “European Digital Library” or not.

The next short talks were about CIDOC-CRM and about OAI-ORE – both complex techniques that you cannot easily describe in short time (unless you are a really good Wikipedia author , but at the least my imagination of them has been improved.

Ebenso beeindruckend wie die Kontinuität dieser Entwicklung, ist allerdings, dass bei den Beteiligten in (digitalen) Bibliotheken oft elementarste Grundlagen fehlen. Mein Standardbeispiel ist (immer wieder durch die Praxis bestätigt!) die verbreitete Unkenntnis der grundlegenden Unterschiede zwischen einer einfachen Suchmaschine und einer Metasuchmaschine. Um unabhängig von allen Hypes ein eigenes Gefühl für diese und andere Grundbegriffe zu bekommen (weitere ad-hoc-Beispiele: “Datenformat”, “Harvesting”, “Feed”, “Community”…) muss man sich immer wieder damit beschäftigen. Gut, wenn es also Fortbildungen, wie den Eingangs genannten Workshop gibt. Einen kleinen Beitrag liefert die Reihe Lernen 2.0 zum Selbststudium.

On the one hand I welcome if vendors try to agree on (open) standards and service oriented architecture. On the other hand the working group is yet another top-down effort to discuss things that just have to be implemented based on existing Internet standards.

1. Harvesting: In the library world this is mainly done via OAI-PMH. I’d also consider RSS and Atom. To fetch single records, there is unAPI – which the DLF group does not mention. There is no need for any other harvesting API – missing features (if any) should be integrated into extensions and/or next versions of OAI-PMH and ATOM instead of inventing something new. P.S: Google Wave shows what to expect in the next years.

2. Search: There is still good old overblown Z39.50. The near future is (slightly overblown) SRU/SRW and (simple) OpenSearch. There is no need for discussion but for open implementations of SRU (I am still waiting for a full client implementation in Perl). I suppose that next generation search interfaces will be based on SPARQL or other RDF-stuff.

2. Availability: The announcement says: “This functionality will be implemented through a simple REST interface to be specified by the ILS-DI task group”. Yes, there is definitely a need (in december I wrote about such an API in German). However the main point is not the API but to define what “availability” means. Please focus on this. P.S: DAIA is now available.

3. Linking: For “Linking in a stable manner to any item in an OPAC in a way that allows services to be invoked on it” (announcement) there is no need to create new APIs. Add and propagate clean URIs for your items and point to your APIs via autodiscovery (HTML link element). That’s all. Really. To query and distribute general links for a given identifier, I created the SeeAlso API which is used more and more in our libraries.

Furthermore the draft contains a section on “Patron functionality” which is going to be based on NCIP and SIP2. Both are dead ends in my point of view. You should better look at projects outside the library world and try to define schemas/ontologies for patrons and patron data (hint: patrons are also called “customer” and “user”). Again: the API itself is not underdefined – it’s the data which we need to agree on.

The “Grid Services” that OCLC is going to provide is based on the “OCLC Services Architecture” (OSA), a framework by which network services are built – I am fundamentally sceptical on additional frameworks, but let’s have a look.

The basic idea about services is to provide a set of small methods for a specific purpose that can be accessed via HTTP. People can then use this services and build and share unexpected application with them – a principle that is called Mashups.

The OCLC Grid portfolio will have four basic pillars:

network services: search services, metadata extraction, identity management, payment services, social services (voting, commenting, tagging…) etc.

registries and data resources: bibliographic registries, knowledge bases, registries of institutions etc. (see WorldCat registries)

reusable components: a toolbox of programming components (clients, samples, source code libraries etc.)

community: a developer network, involvement in open source developement etc.

Soon after social services were mentioned, at heavy discussion on reviews, and commenting started – I find the questions raised with user generated content are less technical but more social. Paul stressed that users are less and less interested in metadata but directly want the content of an information object (book, article, book chapter etc.). The community aspect is still somehow vague to me, we had some discussion about it too. Service oriented architecture also implies a different way of software engineering, which can partly be described by the “perpetual beta” principle. I am very exited about this change and how it will be practised at OCLC|PICA. Luckily I don’t have to think about the business model and legal part which is not trivial: everyone wants to use services for free, but services need work to get established and maintained, so how do we best distribute the costs among libraries?

That’s all for the introduction, we will get into more concrete services later.

After coffee break Haibo Jia presented A new Formal Concept Analysis based learning approach to Ontology building. He extracted 2201 different author keywords and 107 different classification terms from 900 documents of the ACM, added stemming and used the result to create an ontology for query expansion. I hope to follow his work with extraction and analyzing author keywords and classification terms too.

I could not follow the following presentation Formalizing Dublin Core Application Profiles Description Set Profiles and Graph Constraints by Fredrik Enoksson although I would have liked to get to now Alistair’s work. Dublin Core Application Profiles will surely become more important in the future but this presentation did not help me a lot – maybe I was just inattentive because of my own presentation.

So I switched to another session and saw one of my highlights: Panayiota Polydoratou presented the project Repository Interface for Overlaid Journal Archives (RIOJA). To prepare the establishement of an “Overlay Journal” a detailed survey was held among researchers in the astrophysics and cosmology community. The overlay journal concept was already coined by Ginsparg in 1996: a peer-reviewed journal is organized on top of a repository like arXiv.org. The survery showed that most scientists are not interested in a printed journal but in quality control and copyediting. Everything else can already be done with the repository (which the asked scientists already use a lot).

By the way: RIOJA is not the only project to prepare and create an overlay journal. I found the JISC founded Overlay journal infrastructure for Meteorological Sciences (OJIMS) and the Logical Methods in Computer Science (LCMS) which exists since 2005. Peter Murray pointed out that the blog TotallySynthetic.com and the digital library Crystaleye for instance can be called overlay journals. Looks like the concept of a journal is highly evolving (Wake up librarians! When will you finally start to collect weblogs? The Times They Are A-Changin’!). Ok, back to the conference, I will later think more about Overlay Journals, OpenAccess, Science 2.0, OAI-ORE…

In the next session I saw Fredrik Enoksson with An RDF Modification Protocol, based on the Needs of Editing Tools. He explainded that usually RDF is stored in triple stores and there is a need for a remotely editing RDF. In the luisa project a remote editing protocol for RDF is developed and the existing SPARQL update language needs to be extended to fullfill the needs of editing applications.

Beside the low number of participants my own presentation went very well – I got two good questions that helped me to think more about an SKOS encoding of country codes. The slides are at slideshare and a paper will follow.

Finally there were also some posters – but no real poster session to find the corresponding authors of a particular poster. A poster titled Combining Collaborative Tagging and Ontologies in Image Retrieval Systems attracted my attention because of the topic, but it looked very drafted and the explanation contained several gaps so I cannot say whether it contained some valuable research or not. I think the basic idea was to enrich a query with additional synonyms and query a combined index of automatically generated index terms, controlled terms assigned by the creator, and uncontrolled tags.

Video recordings of the presentations will be published in the next weeks and post-proceedings will be published in springer’s Lecture Notes in Computer Science (LNCS) series.

According to heise Zeno.org probably contains the largest free German fulltext collection on the Web with around 600 Million words. The German Projekt Gutenberg (not to confuse with the English project!) is not much smaller, but its quality much is less (missing sources and page numbers) and the publisher Hille & Partner, tries to put copyright restrictions on it. Zeno.org knows better and explicetely states if a work is in the public domain. However full downloads of large parts of the library are not allowed – so zeno.org is more free then Projekt Gutenberg (DE) and Google Book Search but less free then Wikisource and Project Gutenberg. I wonder how it would fit into the Open Content Alliance.

Somehow between in the jungle of German digitized material and digitization projects there is the Central Index of Digitized Imprints (ZVDD), a portal that was funded by the Deutsche Forschungsgemeinschaft (DFG) according to the principle “Too many cooks spoil the broth”: A large number of texts were digitized in Germany with much money and little use because no simple, consistent infrastructure was provided to distribute and archive the content. It could be so so simple with accessible metadata (OAI-PMH) and persistent identifiers…