Information Apartheid - Can a Database be Prejudiced?
OK, hang on to your hats for this one. Corresponding with James Grimster of the wonderful Orangeleaf, we came across one of those deep conundrums which you don’t often have time to stop and think about.
The point is broadly this - the underlying mores of social media and the emerging practice of open, democratic interpretation of collections both depend on software systems. All of this information - user-generated Interpretation, folksonomic tags, wiki contributions - has to be kept somewhere. So is it possible to reinforce prejudice and exclusion by keeping it in a separate place from the ‘authoritative’ information about our objects?
Take the Amazon model - so often cited as the poster-child of smartly contextualised interfaces which help the user make subtle distinctions between ‘authoritative’ voice (the ISBN number, title and publishers blurb) and ‘everything else’. This model is often held up as the way in which we will enable users to differentiate between authoritative object record and the response of external ‘users’ to this record.
But this isn’t democracy, nor is it a free outpouring of public self-expression. Like all things in Social Media world - the reality of Citizen Curatorship, Collective Interpretation or any of the other emerging models is that it depends on a radical re-defining of the balance of authority.
Put it this way - if I said to a particular socio-ethnic group ‘I want your help to revisit these collections and build new knowledge about them, but I’m not going to put what you create in the ‘clean’ core information. I’m going to put it off to one side and call it metadata’ - how do you think they would react.
I have a feeling that this is going to become a new frontline in our ongoing skirmish between command-and-control and socially democratic culture. It seems (and is quite likely from the forthcoming MLA ‘Leading Museums’ document) that the argument for open, democratic and diverse interpretation is won. There is, however, a rearguard action to ensure that it is a transitory, as opposed to a lasting change by closing it off in a walled garden marked ‘inaccurate stuff generated by the public’.
And so the question is - can those who would wish to defend curatorial primacy of interpretation and authority use our Collections Management Systems as a honey-trap to deflect the risk of user-generated views? I posited two models:
- A user looks at an object record on the London Transport Museum website and, using their UGC server, makes a comment about that record.
- A student participates in a Revisiting Collections project at the Cutty Sark and creates a new narrative interpretation of a figurehead.
My question is - are both of these pieces of information equal? Is it reasonable to hive one (the LTM example) off as ‘metadata’? Can we accept the other (Cutty Sark) into the core object record on equal terms with the original documentation - which after all may have been created by an untrained volunteer in the first place.
I’m not sure I have the answers, but I do know one thing - it’s no good rewriting the social contract and perception of the function of museums, archives and libraries if our systems perpetuate the old centrist model of authority.
July 21st, 2009 at 12:41 pm
Nick - just quickly as I’m in a dash
I actually think that users *like* the way in which we (at least visually, on-page) separate the “real” reviews/editorial/etc from the UGC stuff. I think at present this is one of the ways in which we come to build up a picture of the way that resource / book / collection item / insert other thing here “is”. We see a YouTube video which is the “original” and then scroll through the pages of inane commentary, or read a publisher review before working out the stars given by readers. Personally, this works for me. As soon as you start bringing in information and making it equal, you lose the edge that UGC brings. I think.
July 21st, 2009 at 9:20 pm
An interesting question, Nick, but I think it’s going to be a variable picture, with different models for how the data are held and what’s accepted into the sanctum of the Collections Management System, and how the UGC is presented and reused over the long term - does it end up reinvigorating the institution’s own understanding of its collections?
This may have little to do with architecture. Yes, the LTM has a UGC server, and perhaps this suggests something of a firewall between that data and the CollMS data; but perhaps it was really just an easier way of adding the capacity (data structure, functionality) than extending the CollMS to take in UGC. In other examples where a separate system (such as Flickr Commons) is the locus for the UGC creation, it’s later ingested into the CollMS. And if the internal architecture places the UGC into separate a database or tables, well, does it really mean it’s isolated? Surely wherever there exists a link between the object and the comments etc., you can reuse and display that network of data as you please. I don’t see that even if your system led to users adding new notes directly into your CollMS it would necessarily be different: in your multi-valued note table in MimsyXG or whatever each record would have its source attached to it, and then it’s a question of how you choose to interpret and represent the source (as authoritative, inferior, equal, comic, whatever).
In other words, it’s all reversible as long as the link is there. What might be more insidious than the issue of whether UGC is in the same database/table/server as the “master” record is whether the data structure allows for people to express what they wish to with as much power as the curator can. Can they represent their knowledge and their mental map of relationships as they wish to? If not, perhaps they’re subservient to the museum voice after all.
July 21st, 2009 at 11:12 pm
I think there needs to be separation of this and that citizen curators are generally accepting of the separation. The museums play a role of checking content in the same way that information sent to a news organisation should not be treated as authoritative until someone has checked the source.
What museums do need to do is to have processes in place to ensure new information from social media sites is checked and incorporated into the authoritative record. This adds additional workload to the museums, but it also harnesses the public to gather information about collection items that would otherwise remain undiscovered.
June 14th, 2011 at 2:41 am
73. I truly appreciate this post. I’ve been looking everywhere for this! Thank goodness I found it on Bing. You have made my day! Thank you again