Hi Crispin

There isn’t any work currently scheduled on bringing these together. DoB will at some point be subsumed by the thesaurus of Monument type (TMT) but we won’t be adding Aircraft types in as they are deemed to be separate concepts. At the moment if your recording Aircraft crash sites then the monument type is AIRCRAFT or WRECK if it’s underwater (I suppose we should add CRASH SITE as a monument type.). AIRCRAFT TYPE is used to classify the wreckage as it were.

 

We wouldn’t amalgamate them as MARITIME CRAFT and indeed OBJECT used to be part of the TMT but were taken out way back when as they didn’t fit the concept of what a monument was.

 

Also to include terms like SPITFIRE in the TMT would open a rather nasty precedent that would allow users to submit STONEHENGE or BUCKINGHAM PALACE as a monument type.

 

Phil

 

-----Original Message-----
From: The Forum for Information Standards in Heritage (FISH) [mailto:[log in to unmask]] On Behalf Of Crispin Flower
Sent: 08 February 2007 11:30
To: [log in to unmask]
Subject: Re: [FISH] His Dark Materials Thesaurus 3: The Amber Compromise

 

OK Phil, but don't forget maritime construction materials, cargo materials, source/archive materials, context and sample materials...!

Seriously, please think carefully about the implications for the EH thesaurus module in HBSMR, to ensure there is a logical pathway from the current situation to the new. Obviously the easiest path would be to keep the existing thesaurus classes but improve the object materials one to contain useful terms. If you do choose to unify them all, then yes, specific recording/filtering contexts could be given pre-filtered view of the thesauri, but that's not how it works at the moment (as that is what the classes were for). And we'll need to be sure all existing datasets can readily and automatically migrate to use the new thesaurus. I think the existing thesaurus upgrade routines will handle this, because we have allowed for the ability to switch thesaurus classes, but there will be lots of testing required to make sure, or a lot of people might get a bit peeved when several million records get un-indexed! And you don't want to meet a cross HERO on a dark night.

good luck!

C

 

PS has there also been any progress on integrating the Def of Britain, Aircraft and Mon Type classes? Now they really are artificially divided!

From: The Forum for Information Standards in Heritage (FISH) [mailto:[log in to unmask]] On Behalf Of CARLISLE, Philip
Sent: 08 February 2007 11:01
To: [log in to unmask]
Subject: Re: [FISH] His Dark Materials Thesaurus 3: The Amber Compromise

Hi Crispin,

 

The unified thesaurus would still allow you to only choose the specific hierarchy ie building/covering or object. You could do this in software by filtering on the top_term_uids.

As Nick said “For me, if I know its a covering building material I just want to be able to go in at that level and find relevant terms, not step back a few paces and work out if its organic or inorganic and if organic animal or vegetable.” My italics

 

Even if you don’t filter you’d still be able to only expand the hierarchy you’re interested in, which would allow Nick to disregard Object or Building materials and go straight to the covering materials bit (but with the added advantage that if it wasn’t in Covering he could check the other two and, for example, add a term from the building materials hierarchy).

 

I think there are several other good reasons for including them in a single thesaurus:

 

From a maintenance point of view it’s easier to update one thesaurus than it is to update 3.

 

It’s also easier to supply, especially to users who are designing their own small scale databases, where all they’re concerned with is having a list of materials or just interested in selecting terms to stick in a look-up list.

 

In addition a unified thesaurus would allow you to relate a covering material to a building or object material which you can’t do if they’re separate thesauri.

 

I think it should also be remembered that EH have to consider our own needs as well. We’re currently in the process of developing a new reference data system which will move away from traditional thesauri and into a Knowledge Organisation System (KOS) based on an ontological model. Whilst we will still be able to retro-engineer thesauri using EHKOS we will no longer be constrained by the current thesaurus structure and the new system will be able to define greater levels of complexity within the relationships (rather than just the equivalence, hierarchical and associative relationships which exist in thesauri) and as such create a knowledge network of concepts. This way you can associate monument types to periods to objects to materials without creating false relationships. Also as this approach is concept-based rather than term based you can add as many different labels in as many different languages as you like.

As such there will no longer be a need to say that a term is non-preferred as it will just be an alternative label for a concept.

 

So if I want to index the concept of a passage way between two buildings as Jitty I can do and someone in Yorkshire can index their database with Ginnel. A search engine powered by EHKOS would retrieve both records and the person in Yorkshire would no longer be upset when I tell them that their term is non-

preferred because it’s a dialect term.

 

I think although a unified thesaurus won’t be able to do all these things it will give a greater flexibility which is more compatible with the KOS model.

 

Anyway that’s enough form me.

 

Phil

 

 

-----Original Message-----
From: The Forum for Information Standards in Heritage (FISH) [mailto:[log in to unmask]] On Behalf Of Crispin Flower
Sent: 08 February 2007 10:03
To: [log in to unmask]
Subject: Re: [FISH] His Dark Materials Thesaurus 3: The Amber Compromise

 

Are you sure the concensus wasn't to keep the building materials and object materials separate? (Nick B and Nigel P among others).

These are used to index (or search for) totally different entities (monuments and objects) - and if it is sensible to distinguish between those, then surely it is also reasonable to keep the terminologies used for describing their composition separate.

I imagine someone thought about this a bit when they first came up with thesauri for umm object material and building/covering material !

all the best

C

 

 

From: The Forum for Information Standards in Heritage (FISH) [mailto:[log in to unmask]] On Behalf Of CARLISLE, Philip
Sent: 08 February 2007 09:31
To: [log in to unmask]
Subject: [FISH] His Dark Materials Thesaurus 3: The Amber Compromise

Dear all,

 

It’s day 3 in the Big Brother House.

 

I hope that unlike me you’re all tucked up snuggly-warm in your beds gazing out on a snow-covered Winter wonderland (apologies to any FISH folk from foreign climes who haven’t been deluged, overnight, by 5 years worth of snow).

If however, like me, you have the misfortune to live within 5 minutes walking-distance of the office, and thus no excuse to ring in saying that your car’s under 15 gazillion feet of snow, then I can only assume you’re reading this, you lucky person you!

 

Yesterday my brain went into meltdown – I think it had something to do with bricks.

 

Anyway I’ve decided to put the past behind me and forge on with yet another new and wacky plan to solve the materials conundrum (we have to get this sorted today as Philip Pullman only wrote 3 books).

 

I think I can safely say that the general consensus from the user community is that we need the following:

 

3 hierarchies:

 

Building Material, Covering Material and Object Material. These are the 3 Top Terms

 

So far so good.

 

Now we still have to get round the contentious issue of the B-word (ie. when is a material, a material and not a form of material).

 

What about if we have, below each Top Term, two grouping terms (facet indicators – call them what you will), something like.

 

Unprocessed material and Processed material (natural material/manufactured material might work just as well).

 

The point of this would be to try to differentiate between those materials which are used in their natural state and those which have had some form of processing.

 

I’ve not quite worked out yet how to get around the fact that stone which has been shaped by a mason is processed but maybe something like Dressed Stone (worked stone?) would work.

 

Thus a chambered tomb could be said to be made from Stone whereas Stonehenge (the trilithons at least) could be described as being made from Dressed Stone.

 

Terms which could fit into this category include brick, ware types, timber, plaster, mortar etc.

 

Thoughts?

 

Phil

 

Phil Carlisle

Data Standards Supervisor

English Heritage

National Monuments Record Centre

Kemble Drive

Swindon

SN2 2GZ

+44 (0)1793 414824

 

http://thesaurus.english-heritage.org.uk/

 

The information contained within this e-mail is confidential and may be privileged. It is intended for the addressee only. If you have received the e-mail in error, please inform the sender and delete it from your system. The contents of this e-mail must not be disclosed to anyone else or copied without the sender's consent.

Any views and opinions expressed in this message are those of the author and do not necessarily reflect those of English Heritage. English Heritage will not take any responsibility for the views of the author.