Boston airport

So I think the big guy upstairs is having a bit of a joke with me. Remember the US Airways ticket issuing saga?

Last night I rang Air Canada about today's & Tuesday's flights, and they confirmed the same problem on their end. I then rang two different Qantas numbers to try to get the tickets freed up. Both operators I spoke to told me that they couldn't see any problems from the information they were looking at, and that they couldn't do anything about it.

Skip to this afternoon - I turned up at the airport three hours before the flight departs just in case I had the same problems as the other day. So I got to the airport in record time in the taxi, checked in with no problems, and there was no one lined up at security (so I got straight through). Unbelievable!

Anyway, I'm now sitting in the [very small] Air Canada lounge at the airport. They have a mini-café where I was able to get some lunch (fortunately). There's not much else here. So now I'm basically stuck here for the next two hours. Well, at least I can catch up on some blog entries & relax instead of rushing to the flight at the last minute.

I headed out to the Prudential Center this morning for some shopping, and called into my favourite chapel in Boston (St Francis Chapel) while I was there. It was a nice morning - probably the first time I've had to relax since last Sunday. I definitely needed it. I'm also going to enjoy catching up on some R&R the next couple of days in Ottawa - preparation for another big week of meetings & travel. Meanwhile I remain hopeful that the rest of the trip will go smoothly.

Harvard University Library visit

This afternoon I visited Harvard University Library offices to talk to Stephen Abrams and other staff about GDFR, JHOVE2 and Rights Management.

Firstly, I had a chat with Stephen about GDFR to find out how things were going in a little more detail - of interest, as well as good preparation for the GDFR workshop next week.

We also discussed JHOVE2, along with Gary McGath, and agreed that there was overlap between it and AONS. We agreed we should look at some form of collaboration on these projects.

I then re-did my Rights Management presentation for Robin Wendler and Stephen, who had not managed to get to it at the DLF Fall Forum.

All in all it was a good visit. Stephen and his partner/wife(?) took me out for a lovely dinner at Legal Seafood afterwards. Thanks go to both of them for their hospitality!

Foundations of DSpace

Today, I visited the DSpace Foundation & those involved with DSpace at MIT to find out more about the Foundation and it's aims, as well as the planned development effort for DSpace 2.0.

Basically, the Foundation was a necessity to provide structure and governance over the redevelopment project. A pivotal role within the foundation is that of Technical Director (yet to be appointed) whose key role for the first 12 months of the project will be to draw consensus from the DSpace community and make informed strategic technical decisions. The person who fills the role will also keep a watching brief on standards and work in the community (e.g. Sun Honeycomb, WARC files, etc.).

The key goals/outcomes of the planned project are implement the following for DSpace:

• Scalability
• Interoperability
• Modularity
• Workflow - enhanced for ingest and preservation; submission and collection management workflows customisable for each collection type.
• Data Model improvements - metadata versioning is available now, and versioning for objects is planned for implementation in version 1.6.

The improvements delivered in version 1.5a:

• customised submission
• event mechanism
• Manakin integration
• Maven build
• More modular; service-oriented - moving away from monolithic; transitioning from individual to interoperable.

Version 2.0 development is planned for May to November 2008. The key changes are encompassed in the data model revision:

• Ability to handle external IDs (Handle, PURL, DOI, etc.) [pardon the pun]
  • feed in any ID an metadata
  • multi-IDs in versioning - internal & external ID
• Capability to store metadata at all levels of the hierarchy in the object data model.

Looking to the future there is also the intention to integrate with other projects and technologies such as GDFR, JHOVE, Bitstream Format Register, Honeycomb, SRB, Petabox.

Key issues I identified for NLA usage:

• No recursive parent-child relationship in data model. This could probably be circumvented quite successfully by relating objects through relationship references within the same Collection.
• Repository metadata is all completely accessible; objects can be restricted. Put simply metadata cannot be restricted and there is no plan to implement this at present, since the history of DSpace is as an access repository. Possible solutions:
  • record restrictions data and implement restrictions in discovery/delivery systems.
  • implement an second repository (DSpace or other) to store all accessible metadata & synchronise it on updates.

Coffee at COSI

I'm meeting with Michele Kimpton & the DSpace team at MIT this morning, so I walked down to Kendall Square and what did I find... a COSI cafe! This is a special photo for the APSR folks - they've been talking about COSI for ages, but in a completely different context - although I'm sure there's been a lot of coffee involved...

Pumpkin cheesecake

OK, so tonight I ventured out to the Cambridgeside Galleria Mall and had some dinner at the California Pizza Kitchen - evidently no longer a regional venture! I couldn't resist trying the Pumpkin Cheesecake that was their seasonal special.

So, I hear you ask, what does pumpkin cheesecake taste like? and should I avoid it if confronted with it? Well, it was actually quite delicious. Notably, it tasted more of cinnamon and cloves that pumpkin - that is, I couldn't actually taste much of a pumpkin flavour - but it was actually a very pleasant dessert! Therefore, I recommend that if you ever have the option, you should at least give it a try!

Beware international travel...

One of the METS Board warned that the wait to get through security at Philadelphia airport could be about 1 & 1/2 hours. I shared a taxi to the airport with Nancy Hoebelheinrich, chair of the group, and we arrived more than two hours before our flights left.

I spent the next 1 & 1/4 hours at the check-in counter waiting for US Airways to sort out my ticket, which apparently had been control-locked by Qantas (my originating/booking airline). In the end, they had to write a voucher by hand, which was the only way they could issue me an e-ticket.

I found out later that a colleague of mine at the NLA had experienced a similar problem when travelling in Norway recently for work.

I've contacted Qantas & hopefully the problem won't happen again - four more flights before I return home to Australia.

Oh, and by the way, it took me five minutes to get through security!

METS Board meeting

Yesterday & today I attended the METS Board meeting, which followed on from the DLF Fall Forum in Philadelphia. I mostly listened and occasionally contributed to the discussion. There was a fairly strong theme of making METS machine-actionable, which seems to be a major concern of the board members. One possibility that came up during the discussions was that of creating a more conceptual API for METS, which I believe would result in the long term machine-actionability (is that a word?) of METS. The METS board are going to set up a working group on this issue and would like someone from the NLA to be part of it.

Other things reported at the meeting which I noted down:

• Use of METS in web archiving (ARC/WARC)?
• BL to adopt METS as "the" metadata standard.
• LC has released MIX 1.1.

DLF Fall Forum - Fedora Commons

Sandy Payette gave a presentation on the Fedora Commons initiative. The presentation slides are available on the DLF website.

The key points I thought were of interest:

• Workflows, data curation, and enterprise SOA within and across institutions listed as some of the major challenges for the next generation or scholarly systems.
• Fedora Commons aiming to bridge web and enterprise architecture.
• A paradigm shift from the puzzle pieces analogy to a tangram analogy (many possible outcomes).
• In terms of software, a focus on the repository and key enabling services, as well as tools/applications for particular missions/scenarios.

DLF Fall Forum - NLA Rights Management

Time for me to wield my trusty presentation slides for a walk through the NLA's Rights Management project and software. This presentation also went down well with the audience, although the group was much smaller than for the AONS presentation (around 20 I think). This was a little disappointing - in some ways I originally expected more for this presentation than for the AONS one. I think part of the reason was another rights-related presentation that had been scheduled for the same time in another conference room. Anyway, better a small and interested audience than a large and bored one; or none whatsoever.

The small crowd was very interested in what we are doing with rights management. The main question I remember from the end of the presentation was about the use of METS Rights schema; to which I had to answer no - we aren't using it. However, the vibe that I got was that while many people were interested, they simply weren't at that point in dealing with rights information. I think a lot of people are still wrestling with other problems in the digital library arena, and haven't had the resources to look at the rights issue - or maybe have realised, as we have, that it's a can of worms - and a big one at that!

For those interested, the presentation slides are up on the DLF website and you can send me any comments/questions here on the blog.

DLF Fall Forum - JHOVE2 discussion session

Stephen Abrams from HUL ran a discussion session on JHOVE2 project this evening. The session involved a presentation on the key points about the JHOVE2 project, followed by input from the audience. The points that stood out most for me were:

• JHOVE2 will use DROID parts/method for file signature identification.
• The intention is to provide METS output, as well as the current JHOVE output, from the processing.
• Moving to a model where one object equals m files equals n formats (instead of 1:1:1).

There was also mention of trying to distinguish between well-formedness, validity, renderability and usability of a given file format, and a policy-based assessment of risk. I intend to discuss the latter point further when I meet with Stephen on Friday at HUL, since I think AONS addresses some of this. There may be a chance for collaboration in this.

DLF Fall Forum - AONS

This afternoon the much-anticipated moment came - or at least the moment that I'd been preparing for. I gave my presentation about AONS II at the forum. The presentation went down well; the smallest of the three function rooms packed out with an attendance of about 70 people. People were very interested in the AONS software and the fact that it is open-source, as well
as the focus on format obsolescence risk measurement in the project. Several people asked for the risk questions that we developed during the project.

For more information about the project and the software:

• APSR AONS website page
• Presentation slides
• SourceForge project
  

DLF Fall Forum - PennVibes

In this presentation, UPenn Libraries staff Delphine Hanna and Michael Winkler demonstrated some of the work they've been doing in content delivery and organisation with AJAX, XML and Java to deliver a flexible, configurable interface for users. Essentially they have a portal with function/tool widgets (RSS feed subscribe, catalogue search, etc.), which looks much like iGoogle homepage. The widgets are developed with Script.aculo.us and prototype.js, in a PageFlakes-like environment.

The main issue identified was a lack of standards for widgets like this. However, they are hoping that a pool of library-oriented widgets will be produced over time.

One of the really interesting things is that they are also using this same technology as a web publishing tool. Reference librarians are contacted by phone and obtain information from the caller. At the same time, they are using the information provided to create a widget-based web page which will help the caller with their research query. Once completed, the librarian published the composed webpage as a public website page for the caller to access and use, with the added bonus that the page is there for other website users to refer to in the future.

This creative use of the technology was of interest not only in terms of reference, but also web publishing. Assuming that all (or most) of the information you wanted to publish on the website was available in structured electronic form, you could potentially construct your entire website from a variety of these widgets which use that information, and then publish it as static HTML pages. Maybe we will see this emerge as a way of creating and maintaining websites over the coming years.

DLF Fall Forum - Zotero update

I remembered Mark Corbould raving about Zotero when he came back from his trip to the US about 12 months ago, so I decided to go along & get the latest.

Zotero 1.0 - New Features:

• export to Wikipedia citation templates
• create Simile timelines from data

Zotero 2.0 plans - Server-side Edition

• shared collections & notes
• scholarly groups in macro- and micro-disciplines, official groups
• recommendations
• bibliographic feeds
• APIs
• collection synchronisation

DLF Fall Forum - Library Catalogue as a Versatile Discovery Platform

Interesting presentation by Tito Sierra, Emily Lynema and Markus Wust of North Carolina State University (NCSU). They've built something they call the CatalogueWS API on top of their implementation of Endeca search server for their library catalogue. It has a fairly RESTful API with a custom response schema. It would be interesting to compare this with our RESTful implementations at NLA.

They've also implemented some nifty gadgets to demonstrate what they've done:

• a Google widget
• MobiLIB for mobile devices
• Facet Browser

on top of their web service API - available at www.lib.ncsu.edu/dli/projects/catalogwsapps/

Technical documentation is available at www.lib.ncsu.edu/catalog/ws/

DLF Fall Forum - ESA & the Library (Chris Mackie, Andrew W. Mellon Foundation)

This talk by Chris Mackie nicely complemented Richard Akerman's prior talk in the same session. This one focused more on Enterprise Architecture (EA) than on SOA. Here's my summary of the talk:

• Internal IT capacity is declining.
• Application overburden.
• What if funding runs out?

Hmmm...I wonder if this relates back to the old do more with less discussions we had at NLA many months ago? I think so...

• The value of SOA: malleable to organisational processes.
• Challenges:
• must understand business processes!
• service-oriented applications reinvent problems at a different level
• ESAs: the practice of SOA in a nurturing ecosystem that encompasses the enterprise
• so what is the enterprise?
• need to define it
• Word about the cloud: "use it, but don't let it own you... remember clouds are mostly vapour!"

Need for middleware. Examples of middleware for ESAs:

• ESB (www.kuali.org/student)
• Business Rules Engine (JBoss)
• Workflow
• FLUID UI (www.fluidproject.org)
• Projects:
• www.seasr.org
• Fedora
• DSpace
• www.sakaiproject.org
• www.open-ils.org
• www.bedewark.org
• Bamboo

Some of the examples above may be of interested to us at NLA in regard to the development of workflows for our physical formats digital preservation project, which is currently underway.

DLF Fall Forum - SOA for Libraries (Richard Akerman, CISTI)

This, for me, was one of the most interesting sessions at the forum. Also the one that might have most pertinence to the NLA's current work.

Richard first described enterprise architecture (EA) as vision translated into operational success (borrowed from an earlier presentation). This leads to the questions: who are we and what do we do as an organisation? The following points need to be addressed:

• Business framework
• IT Architecture framework
• Framework for implementation
• projects
• metrics
• buy-in
• processes
• people
• Implementation

The following needs must be addressed to achieve success:

• Tightly linking the architecture to the business.
• Use an approach that is disciplined and traceable back to the business.
• Develop an action plan to support implementation.
• Architecture outputs are integrated with each other and with other IT outputs (base analysis and design on te architecture).

SOA services, in the context of an EA, should be: autonomous; loosely-coupled; coarse-grained; have well-defined interfaces; and, allow reuse through composition.

CISTI has a five-person dedicated architecture group consisting of a senior manager/chief architects and four other architects. They have developed a litmus test to evaluate business functions in terms of SOA characteristics. They have an architect from the group involved in every project team to ensure the project achieves its goals within the EA and SOA environment.

I thought it would be great to have something like this at the NLA. I wonder if this is something we could consider along the lines of the Innovation division, particularly as this is really a problem for the organisation to solve; not specifically IT?

DLF Fall Forum - New Organisations for Collaboration & Development panel

This session was true to it's title, with emphasis on new organisations. I would have preferred more emphasis on informal organisation for collaboration - for example: how to set up collaboration between like-minded organisations to successfully achieve goals (without an Non-Profit Organisation [NPO] having to be formed). I accept that sometimes an NPO will be necessary (I don't think there's anything wrong with the directly both Fedora and DSpace have taken), but I think it would possibly have been more beneficial for the audience to hear about less formal collaborations and what made them successful. The latter of the panel presentations probably hit the mark more so that the others for me. My two cents...

My notes:

Vicki Reich (Stanford University) - CLOCKSS

• dark archive
• no/low cost of continued service in case times get tough
• board meets every two weeks by phone
• libraries and publishers working collaboratively
• subscription gives representation on the board
• board's role:
• raise money
• make decisions
• decide what content is made available to whom

Michele Kimpton (DSpace Foundation)

• 270 registered users of DSpace worldwide, but there are more
• 100 active developers
• 14 lead developers
• advisory board set up independent governance organisation
• owns copyright
• enters agreements/contracts
• raise funding
• 503c3 charitable organisation
• board members from both commercial & non-commercial organisations in the community (currently six people)
• foundation to direct, not participate in, coding
• funding: currently HP & MIT

Sandy Payette (Fedora Commons)

• Fedora needed long-term strategy
• set up board
• centralised approach
• outreach for more developers
• startup funding: Moore Foundation, $4.9m over four years
• see Fedora Commons proposal on website
• need to move away from central ownership/key people a little
• plans: a couple of pre-packaged solution bundles, plus roll your own option

Katherine Kott (DLF Aquifer)

• new organisations & initiatives:
• emerge as common problems are identified
• existing organisations have different priorities in future
• what to build/buy?
• what is specialised & not?
• collaboration
• distinction between it & cooperation
• common interest or commitment to other's interest as much as own
• innovation generator
• small team
• supported by individual enterprise contributions
• DLF
• consistent attention
• admin support
• working groups & resources from contributors

DLF Fall Forum - Public/Private Digitisation Agreements panel

This panel session didn't quite have the angle I was expecting. Instead of a real panel discussion about pros & cons of various approaches to contracts/agreements for digitisation, each panel member described the details of their arrangements with various digitisation service providers. While this was related and still informative, I had hoped for more open discussion about what to do and what not to do, and maybe some conflicting views between panel members; but it was not to be.

The key things that stood out for me were the hope that de-duplication can hopefully be done automatically, and the need to make clear who can do what after the term of the digitisation contract/agreement finishes.

DLF Fall Forum - Keynote 2 - Rick Prelinger, Prelinger Museum and Archives

It's Only A Moving Image

This talk was essentially about video content archiving and making it available. Rick Prelinger outlined some of his experiences with moving image. The bottom line of the presentation was that access to moving image content is still quite difficult, but advocating the value of making it available to the user community.

The quotable quote from the presentation was wonderful and unpredictable things happen when you give people access. I think this is true, but not just in the context of video content. I also think that these wonderful things happen at a much faster pace than they used to when we were all talking about access to physical collections.

The presentation slides are available on the DLF Fall Forum 2007 website.

DLF Fall Forum - Keynote 1 - Dan Gillmor, Citmedia.org

Dan Gillmore, one of the keynotes, gave an interesting talk titled From Lecture to Conversation, which was about the online environment and new technologies being used by everyday people for everyday journalism. Essentially he described what is a journalism revolution, and forecast the demise and disposal of spot-news journalists everywhere. He mentioned concepts such as: Read-Write Web, blogs as the Daily Me, community-based journalism (which I've dubbed open-source journalism), journalism's evolution from lecture to conversational style. He also predicts a move from the Daily Me to the Daily Us. From the library perspective, Dan spoke of a vital role for libraries (and others) to ensure media literacy in the upcoming generation.

The presentation raised questions in my mind about newspapers coming into the collection and their relationship with the online articles from the same newspapers' websites. Often these are identical - I wonder if there's a way of describing this relationship. I'm pretty sure we don't web harvest & archive the online news sites at the moment - maybe we should?

DLF Developers' Roundtable

This morning I attended the DLF Fall Forum Developers' Roundtable discussion. It's a technical forum intended for the discussion of common technical issues. The format of the meeting allows for each attendee to give an informal response to the prompting points: (1) cool new technology of interest; (2) over-hyped technology; and, (3) opportunities for collaboration.

I gave a response talking about the our IT architecture report, the single-business approach, service-oriented architecture, REST-based services, VuFind, AONS II, workflow analysis (BPMN) and Ruby on Rails; also mentioned that a variety of discovery systems is overhyped when a more single business approach would be better, and that there are opportunities for collaboration within the digital preservation community.

General themes were: the use of Ruby on Rails, OpenSocial API, SOA/enterprise architecture, and a tool called Twine; and Google overhyped, iPhone/gPhone overhyped, social networking tools overhyped. There was no real consensus on ideas for collaboration. It appears a number of people are using Ruby on Rails, and looking at OpenSocial API, some looking at enterprise architecture and SOA.

Sunday arvo in Philadelphia

I realised when I got here that I'd forgotten to bring any ties to wear. Not that they're a necessity, but they do keep you a little warmer & I quite like wearing them occasionally. Anyway, as a result, I spent part this afternoon shopping for ties, and the other part wandering the city. It was interesting walking around, but this was about the most exciting thing I saw. I didn't manage to get to any museums, etc.

A lazy day in Philadelphia

I got to "Philly", as the locals call it, about ten o'clock last night, checked in at the hotel, and got to bed about midnight. I awoke around seven this morning and vaguely remembered someone mentioning changing clocks back from daylight saving time - recollection confirmed by friendly note from the hotel which had been slipped under the door during the night. This was a little surreal, as we just changed our clocks forward for daylight saving last weekend back home in Canberra. With all the other timezone changing I guess another one isn't going to be too much problem!

I had breakfast & then headed off to mass at St John the Evangelist. I had planned going to St Patrick's, but since I was awake earlier than planned I just made it to 8:30am mass. The streets of Philadelphia were quiet at that hour - a little creepy in a strange place, but not surprising as virtually everything was closed. Apparently, St John the Evangelist church was the place of baptism of St Catherine Drexl and was where St John Neumann was installed as Philadelphia's fourth bishop (more details). Well, I wouldn't have found that out if I'd gone to St Patrick's.

I wandered back to the hotel & sat in the sun while I read The Philadelphia Inquirer, looking for some inspiration on places to visit this afternoon - none resulted.

On the flight from San Francisco I managed to get through some reading material, as well as final planning for my DLF presentations on Rights Management and AONS. I also managed to look out the window occasionally for some beautiful views. I figured someone would start complaining that there were no picture pretty soon, so I took this one of snow-topped mountains somewhere over the United States. I'd love to know where, so let me know if you have any ideas.

Timezone sticking points

Just realised that my posts are out-of-order due to timezone issues. In case you missed it, there's a blog entry from San Francisco airport.