About Me

My photo
Web person at the Imperial War Museum, just completed PhD about digital sustainability in museums (the original motivation for this blog was as my research diary). Posting occasionally, and usually museum tech stuff but prone to stray. I welcome comments if you want to take anything further. These are my opinions and should not be attributed to my employer or anyone else (unless they thought of them too). Twitter: @jottevanger

Saturday, May 10, 2008

National Collections Online: why, what, how...in fact, ?!?

Friday afternoon we gathered at the Science Museum's Dana Centre, perhaps two dozen people sampling a variety of disciplines but all, for one reason or another, interested in the National Collections Online project. This was a workshop for the feasibility study being run by Flow Associates (Bridget McKenzie and Mark Stevenson) following up the discussions they've had with, presumably, all or most of those there. NMSI and the V&A had good representation, not just techy and new media but curatorial and from the collections data side. The National Maritime Museum, the third museum partner in the project, had a couple of representatives too. Culture24 is another partner, and Jane and Jon were there. The NMDC, who I take to be the initiating partner, had no direct representation but wil hopefully learn the lessons anyway. The rest of us included e-learning experts, a couple of people from other museums and the Wellcome Trust, Ross flying the flag for academia, Ben Hayman of Lexara that for the commercial sector, Nick of Collections Trust, Jill Cousins from EDL (which I thought was brilliant, unfortunately she couldn't stay until the denouement but I have a feeling that her presence may have swayed some sentiments in the direction of EDL), and finally Tom Steinberg, director of mySociety. A very interesting crowd.

The biggest problem with the session was, I guess, exactly the same as that which it was trying to address, and perhaps will have helped us to break through it before the next gathering. That problem is the vagueness of the project, which made it almost impossible to talk incisively about it. Someone mentioned a 10,000lb gorilla at one point; all I could see was a big cloud in the middle of the room, who knows what was at the heart of it. I've not talked about NCO before so I should do the basics: Flow's inquiry is into the "viability of an online resource to integrate national museum collections", and perhaps we can orient ourselves as to what this might mean by the well-known points of reference that they have mentioned: CHIN/CCO, CAN, culture.fr, Artstore, Powerhouse. It seems that many people said "not a portal", which was the common response at the session too. Given this, we both needed to decide what we were talking about if not a portal, and know that in advance to make much progress on questions like whether a "single point of access" would be of use to various audiences. This isn't a criticism of how the session was run, it was sure to be tricky as a natural consequence of the fact that, even before the project properly starts, it maybe engaged in a major change of direction (and this is the time to do it!).

Anyway, it was a useful process and it did seem to bring out a good degree of consensus on the unsuitability of an old-skool "portal" approach. I took away a couple of key points though, including a useful reminder:
  1. Fiona Romeo made a remark that inspired a "doh!" moment in me: why not talk to Google about them ingesting our content directly? This time last year we were talking about this very prospect at the UK Museums Semantic Web Think Tank, and it still seems like a really good plan (Yahoo! too, especially since they seem at least as keen on adopting semantic technology). Still, somehow, I'd stopped thinking about this line of attack having been concentrating on the opportunity (or threat, if done badly) offered by EDL. And in the session I didn't shy away from pushing EDL as the obvious place in which collections data should be aggregated, so as to scale to all museums and not duplicate efforts. I argued that NCO should wait for this to firm up before deciding what to do that would compliment or build upon it. However when Fiona mentioned the prospect of working with Google, I realised that over the last 6 months of talking to EDL about how to make sure that it wasn't seen as irrelevant by museums, I'd started to forget that it needn't be the only path down which we go to achieve our goals - just the one that grabbed my attention late last year. I think that the two approaches can be complimentary, and in fact EDL itself would be well advised to talk directly to the search providers about their ingesting structured data. NCO could, in theory, provide something of a breakthrough that would be genuinely extensible and scalable. This would also put the lie to one of my contentions, which was that there was little point in doing something that only involved a small number of nationals. On the contrary, if it opened a very wide door for cross-collection search as this approach might, it would be very worthwhile.
  2. I had a moment of clarity, which followed on from the recent hooha on the MCG list concerning the disappointment or otherwise of the NOF Digitise programme. One of my arguments was that it would be better to make distinctions in projects between those that do digitisation, those that build functionality, and those that build user experiences. I realised that I'm talking basically about slicing funding differently, changing from vertical to horizontal slicing, and that it's not unlike talking about markets. I'm going to post separately with more thoughts on that.
  3. We had useful input from Tom. Though he was kind of preaching to the converted about the idea of making our content as widely available as possible, it's not surprising, and he also furnished us with some useful parallels and metaphors. From our later chat at the pub it's clear enough that he's as keen on the lightweight dissemination of semantic data as I am (albeit sceptical about the Semantic Web - but then in a sense so am I, it's semantic technology that is making the headway, and there are riches to be found there that do appear to be speeding us in the general direction of a more semantic web in any case)


sebchan said...

Hi Jeremy

The issue that sounds like it is bubbling away underneath all this NCO discussion is the - for whom? (which I gather might be bound up in all sorts of class and 'worthiness' issues that are particularly pronounced in the UK)

It is one thing making collections available for traditional museums research (and to some degree, education) audiences through projects like EDL etc . . . but it is a totally different thing making them available and usable for non-traditional museum audiences.

Whilst everyone sounds like they are against a portal approach, sometimes a well designed and, importantly, marketed approach is the way to go.

For other audiences - say those wanting to find out more about the antiques they are buying and selling on eBay then Google and SEO methods are the way to go.

Then it becomes a question of resources and the success metrics for the organisation/s doing the aggregating.

It may not be coincidental that the colonies (canada and australia) seem to have been the early movers.

heh heh


Jeremy said...

Hi Seb, thanks for the comment. GA tells me I have a regular visitor from Sydney and I had my suspicions it might be you...
You're right, the question of the target audience was the starting point for our huddles on Friday - a group each talking about the "general", "learning", business and specialist (including academic) audiences, and what they might want from....whatever this is meant to be. It's true that various programmes and KPIs are tied to angst-ridden (and justifiable) concerns about ensuring broad involvement in our work. In a way my point is that, precisely because we are keen to address multiple audiences now and in the future, the processes of digitisation, of building base functionality, and of creating a user experience should as a rule be separated. At least, funders should not require all parts of this value chain to be built together. Like you say, there may well be strong reasons why museums would still want to do this, though, and if it doesn't mean spreading resources too thinly then yes, fine. But I can think of multiple examples just from MoL where we have been obliged to commit to doing things for the sake of a funder that, let's just say, didn't make best use of resources. With the UK's HLF explicit about not funding digitisation per se, only projects that have a user-facing product and may involve some digitisation along the way, we're going to carry on seeing lots of efforts diluted and divided. And in the end we do want a web output, of course, but an "architecture of commissioning" as well as technology that worked on a larger scale might be more effective. I think getting too hung up on audiences is a problem - if something is worth accessioning into our collections, it's worth having it online. Only getting this out there in a free and easy way (as you suggest) will let our audiences define themselves and use the stuff as they will.

I'd take issue with what you seem to perceive the point of EDL to be, though. The whole reason I've got involved with that project is to ensure that it has as a strong focus the middle layer of the architecture - the functionality built around aggregation, search, translation, and using semantic technology - and exposes this as a public API. True, the first generation may only be Europeana, the portal-style website, and this just might in fact hit those targets you mention of being well designed and marketed. But I'm pinning my hopes on a decent API to help it address the myriad other audiences in a million other parts of the Web. It would be an obvious step to talk to the main search companies about how they could not simply index the pages, but ingest the data in a structured, meaningful way.

CHIN really has led the way, and you guys too, and who knows if we'll catch up here in Blighty. I'm not convinced that a portal looking at the collections of just national museums (in fact just 3 of them to start with) would get us far in that direction, but we have other things we need to get right too, if we are to get all our museums' collections in one place - political and structural barriers, possibly problems of scale (given that there are considerably more museums here than in either Canada or Australia).

Cheeky get! ;-)


Frankie Roberto said...

I wasn't invited to the NCO meeting, so it's interesting to read your notes on the discussions.

I'm pleased to read that people like Tom from mysociety were there, as well as museum people.

It'd be good if as much of the discussion was as open as possible.

Jeremy said...

Hey Frankie. Yup, there were a couple of folks from the Science Museum there (Robert Bud, as you know, and Sarah Winmill) but your team wasn't represented. Perhaps it's worth getting in touch with Flow about participating in the next session. I did think it was really useful to have the views of people like Tom and David Mason from the London Grid For Learning as well as the usual crowd.
Glad what I wrote was useful, though I know it's pretty garbled and mainly reflects my own preoccupations! I hope at the next session we can start to see the direction NCO might head in. Right now I haven't a clue!