Comments on: Name authority for dead people

By: Jamie Norrish

Jamie Norrish — Tue, 06 Oct 2009 20:11:54 +0000

Thanks for the kind words, Amanda (and sorry for the late response; I think the comment notification thing must have had a hiccup). I think you’re giving New Zealand too much credit – while there is a some degree of understanding regarding the importance of authority control/access, it’s apparently a giant leap from that to actually doing something.

That said, Te Papa Tongarewa (the Museum of New Zealand) is engaged in some authority control work with regards to parts of their collection, to be exposed on their site, and of course you know about People Australia. Libraries tend to have the information, but operate in splendid isolation even when the information is good.

More shameless self-promotion: those who aren’t familiar with the NZETC’s EATS system might like to read http://hdl.handle.net/10063/220 for a description.

In the context of your concern regarding duplicated work and the problem of retrieval from multiple sources, this is a obviously a difficult issue. Everyone has their own system and doesn’t want to change, so any solution has to either be forced on them, or involve a second system that links and coordinates the others. EATS is intended as the second; I’d love to hear if anyone has a third option that I haven’t thought of!

By: Amanda Hill

Amanda Hill — Thu, 10 Sep 2009 18:23:50 +0000

Thanks Jamie – I think your approach is exactly the right one. My concern with the situation in the UK cultural heritage context is that there is no easy way for practitioners to assign and share unique identifiers for individuals in the first place, so there is a lot of duplication of effort going on and consequent problems with retrieval of relevant information about the same individual from different systems. There is a lot of work now under way (by publishers and others) on uniquely identifying living researchers but my concern is that the cultural heritage sector is not involved in this in the UK. It sounds like this has received a bit more consideration in New Zealand!

By: Jamie Norrish

Jamie Norrish — Tue, 08 Sep 2009 00:58:47 +0000

Coming very late to this, I’d like to mention that the EATS system I’ve developed for the New Zealand Electronic Text Centre is designed to act both as a repository of entity authority data, and as a mechanism by which authority records from multiple systems can be linked — without the twin perils of privileging one system, form or piece of information over another, and without requiring any system to give up their existing identifiers.

I know that the Names project is aware of the EATS work (I believe it’s mentioned in one of the reports, and I’ve talked with Daniel about it), but it can’t hurt to mention it here. We (the NZETC) are currently in the process of incorporating records from the Auckland War Memorial Museum into our instance of EATS; we’re hoping this will be the start of similar collaboration with other groups within the cultural heritage sector in New Zealand (and Australia too, with luck).

By: Joy Palmer

Joy Palmer — Mon, 27 Jul 2009 12:51:10 +0000

I think we’re certainly on the same page, Amanda — this needs to be about identifiers and not biographical histories. My reference to wiki software might have been a bit of a red herring, I realise now — I agree with your point that what *is* needed is a mechanism for sharing/linking the peristent identifier data.

By: Amanda Hill

Amanda Hill — Mon, 27 Jul 2009 12:30:35 +0000

I appreciate your concern about deciding upon a definitive biography for an individual and putting the weight of archival authority behind it. The archival standard certainly allows for very full description. But the key point there is that although such full histories are permitted, they are not mandated. The core elements of an archival authority file can as simple as the names and dates of an individual, family or organisation, along with the identifier. My feeling is that it is this identifier that is the key, rather than lengthy biographical histories. I don't think archivists have a lot of time to create full authority records (although it will be interesting to read the results of the survey that was recently sent out to UK archivists on this topic). A bare-bones record with an identifier could link out to all sorts of other resources. The important thing is to have a mechanism for creating that identifier in the first place and for making it persistent and providing useful information when it is accessed (the linked data principles, in other words). There isn't an easy way of creating and sharing those identifiers right now, in the cultural heritage sector, as far as I can see. Wikipedia, for example, has 'notability guidelines' which would exclude many of the people we would want to identify. I didn't mention using a wiki in this post, although I did in our side discussion on Twitter! It just occurred to me that a way of making it possible for the community to create a shared resource would be to pre-populate a semantic wiki of some kind with the National Register of Archives' data, and build on it from there.

By: Joy Palmer

Joy Palmer — Mon, 27 Jul 2009 10:31:37 +0000

Hi Amanda,
A belated response to your post. As you know I have a somewhat different perspective on this, but I think fundamentally we’re in agreement — we need to be able to share and reuse data across the sector. The lack of a single naming authority does present significant issues for services and their users, and the example you give above from the Hub is an all too typical example. It’s clear that technically speaking we need to solve this problem. I have two distinct concerns, however. The first is a philosphical one — the archival standard for Name Authority files is about more than agreeing a naming convention (which I don’t have a problem with) — it is about providing biographical, historical and interpretive information about that person and the significance of their life. While we might all agree that this is simply an instance of historical interpretation, the fact is that is this becomes ‘The’ authority file, as written by Archivists, then it will carry a great deal of weight. Of course, such information could be very useful, but it’s the centralisation and standardisation of this ‘authority’ file as what might be perceived as a factual document that I have a problem with. I concede that a wiki could well allow for more flexibility and fluidity in this regard (if we could get archivists to embrace the mindset of multiplicity, that is). If such records could also link to related contextual information outside of the archive, then this would also be useful.

My second concern relates to the practicalities of getting archivists to collectively contribute to such a system, and whether we in fact need yet another centralised system to handle this problem. It would be an interesting experiment to see if we could get archivists (and non archivists) to contribute to a wiki to solve this issue, and whether a community could evolve out of this work, but this would require a great deal of effort, mobilisation, and handholding (but you’re right — we could try:-)). The ’simple’ solution of a wiki is not-so-simple if you want it to be actually used. In the meantime, I wonder if semantic technologies, better search, and also an open mind around the linking of archival content to contextually related content ‘out there’ and conversly, opening up archival data for use ‘out there’ might offer better solutions, and might be a better place to direct energies.

Lots to think about, certainly. I appreciate you forcing me to stretch my non-archivist brain a bit on this one!