| 
View
 

problems open data could fix

Page history last edited by Mia 12 years, 6 months ago

AKA 'finding the gap' in order to fix it...

 

While thinking about the barriers to open, re-usable data from museums, libraries, archives, it's important to realise that (in the UK, at least), cultural heritage organisations are operating under difficult circumstances, and this tends to focus their collective minds on proposals that either solve problems they already know they're facing, or that require no resources (attention, time or budget) from them.

 

What problems do cultural heritage organisations face that open data could help with?

 

From Europeana's 'Yellow Milkmaid' report (The Problem of the Yellow Milkmaid: A Business Model Perspective on Open Metadata - pdf):

We distinguish ten major potential benefits:
1. Increasing relevance: open metadata can be used in places where online users congregate (including social networks), helping providers to maintain their relevance in today’s digital society.
2. Increasing channels to end users: providers releasing data as open metadata increase the opportunities that users have to see their data and their content.
3. Data enrichment: open metadata can be enriched by Europeana and other parties and can then be returned to the data provider. Opening the metadata will increase the possibility of linking that data and the heritage content it represents with other related sources/collections.
4. Brand value (prestige, authenticity, innovation): releasing data openly demonstrates that the provider is working in the innovation vanguard and is actively stimulating digital research.
5. Specific funding opportunities: releasing metadata openly will potentially grant providers access to national and/or European funding (European and most national governments are actively promoting open metadata).

6. Discoverability: increased use and visibility of data drives traffic to the provider’s website.
7. New customers: releasing data openly offers new ways to interact with and relate to customers.
8. Public mission: releasing metadata openly aligns the provider with the strategic public mission of allowing the widest possible access to cultural heritage.
9. Building expertise: releasing metadata openly will strengthen the institution’s expertise in this area, which will become a marketable commodity such as consulting services.
10. Desired spill-over effects: institutions and creative industries will be able to create new businesses, which in turn will strengthen the knowledge economy.

 


Some earlier responses from twitter...

 

 

And in support of trying to find the answers...

http://twitter.com/homebrewer/statuses/95672721026985984 @mia_out Tweet the answer once you've got it! It would help rally support from up the ladder.

 

What other barriers exist to publishing open, re-usable data from museums, libraries, archives?

Comments (6)

Rachel Coldicutt said

at 12:29 pm on Jul 28, 2011

A problem that open data could fix is sustainability. Lots of arts organisations (particularly in the performing arts) use slightly bespoke and/or legacy systems, and their data formats may or may not conform to wider standards. This leads to more bespoke wrangling to create web (or other) applications, which then become expensive or difficult to sustain. It also prevents white labels or shared applications. In my experience, arts organisations often concentrate on the 2% that makes them distinctive, rather than the 98% of shared need across the sector.

in terms of barriers, I think fear (both of potential infringement and of existing staff being made irrelevant) is a major obstacle. Also, in lots of places I've worked, building e.g. an open API would be thought of as too boring to invest in: if there's money to be spent, senior managers often want to spend it on something immediately visible that will yield, e.g., a marketing result - they might not have the expertise to realise the importance of more infrastructure-type projects that will seem long, expensive and irrelevant. When I wrote my first digital strategy (I'm not saying who for!) I was told that a three-year view was "too long term". So, without wishing to generalise, I think short-termism may be a problem.

My fear is that all of the above will result in the WWW being a very sad place: fear and short-termism meaning that suddenly there will be little or no discoverable art, as it will all be hidden away or badly formatted. And after unfindability, irrelevance will surely follow...

Jasper Visser said

at 1:50 pm on Jul 28, 2011

For another project, over the past 6 months I've asked 10s of people working in or with cultural institutions what they considered the main problem the sector faced. (I should make the final report of this investigation...). The problem, not surprisingly, is cooperation between institutions. With this, people meant a variety of things, such as the lack of insight in what other institutions are doing and have done (so everybody keeps repeating the same mistakes and investing in the same research), the lack of willingness to share valuable information and best/worst practices and the trouble it sometimes takes to do something as simple as arrange a loan.

Open data, obviously, can solve some parts of this issue. For instance, Digital Heritage NL has an open project database that combines best/worst practices, details etc. of many of the digital projects done in the Netherlands over the past couple of years. Project managers should be encouraged more to add their work, the data can be made more accessible, etc., but it's a start and it's open.

However, I believe there's an underlying problem that cannot be solved with open data (as the true impact of technology is always limited): trust. Successful cooperation is built on trust, and somehow I feel that's the thing lacking. Open data can increase trust, but probably it will first have the adverse effect.

If you're looking for the gap, I (and many people with me) believe it's cooperation. The exact role of open data in this, however, is more complex than that.

jeffreynewman said

at 5:00 pm on Jul 28, 2011

Dear Jasper and Mia,

I appreciate Jasper's comment. I remember some time ago reading a blog on the etiquette of social media - essentially, of course, courtesy, generosity, empathy and especially taking the time to communicate in a way that recognises the humanity of the other. Someone said that when you meet someone with whom you have been in contact over the internet, what is surprising is: that you recognise the person as you expected them to be!

Building trust is partly about the other but also about oneself: do we trust ourselves? If so, others can trust us?

Hope that is on the right lines...

Jeffrey

Nick Poole said

at 11:08 pm on Jul 31, 2011

Hi Mia,

I just wanted to say a few words by way of expansion on my tweets. It is interesting how similar the discussion about Open Data is to some of the earlier discussions about Digitisation, and before that Cataloguing (if you go way back...) In all 3 cases - Open Data, Digitisation and Cataloguing - there is a fundamental economic problem: the upfront investment required is disproportionate to the perceived benefit in the medium to longer-term.

Because making data openly available as #linkeddata requires actual money to be spent, and because there is a general perception that doing so is an active *rejection* of any economic potential it might have had, the underlying Value Proposition is not compelling. As things currently stand, you have to be something of an evangelist for quite a complex point of principle to publish your museums' records as Open Data.

What I was trying to say in the tweets was that it might be better to look at applications which make use of #linkeddata to add value to areas of museum practice that are currently difficult/expensive. Interpreting and recording structured information about objects is expensive. Research is expensive. Even the manual re-entry of repetitive descriptive information is expensive. The use of dynamic co-referencing based on #linkeddata to support augmented cataloguing might provide a use case which helps to 'sell' the idea of Open Data to a community beyond the evangelists and early adopters.

It seems to me that if we can help a critical mass of museum professionals to become *consumers* of Open Data, and to regard it as actively beneficial to their work, then it would become much easier to sell them on the exciting possibilities of making their own resources in the same way to others.

Mia said

at 12:35 pm on Aug 1, 2011

Hi Nick, thanks for commenting.

I should probably have been more clear in my text, but this is about open data rather than linked datal - I think there's a whole extra set of issues when thinking about publishing linked data (unless you use someone else's service). Open data, on the other hand, can be published in almost any format that can be exported from a collections management system - CSV being a classic example. There are issues with the subsequent disconnect from the internal source database, but at least it gets the data out there in a re-usable format, and at very little cost to the organisation.

So, having said that, can you expand on the idea of publishing open (or just linked?) data as an active rejection of the economic potential of the data?

Museums often exchange records about objects when doing loans, etc - this is well-established as a use case, but apart from conversations at Museums and the Web 2011 about using Open Search as a potential format, I'm not sure if there's ever been much work on this.

Cheers, Mia

Mia said

at 6:13 am on Apr 15, 2012

There's a collection of real questions people have asked that linked open data in cultural heritage might be able to fix at http://ldch.da.ulcc.ac.uk/index.php/People

You don't have permission to comment on this page.