Resource graph - Related resources for node

[Vincent Meijer]

Hi all,

I am working on an idea I have to record all interpretations made regarding a specific (heritage) resource. 

I would like to relate the assignment of e.g. a cultural period or the use/type of an object (hide scraper, spear head, vase) or an object category (defensive structures, tools, jewelry) to an authority.

This authority can be an actor or a (to be created) academic reference, both resource graph types, rather than values in a regular dropdown/authority document. 

I want this in particular because it is easier for my users to create a new resource than it is to add values to dropdowns via RDM. Also it seems to me that using the type of search queries CIDOC encourages (find unexpected relationships) would work better this way. 

I came up with the following addition to HERITAGE_RESOURCE.E18 (new nodes are the ones sticking out in the top) for HERITAGE_RESOURCE_USE_TYPE_ASSIGNMENT.E17:

(ASSIGNMENT_AUTHORITY.E39 would hold a foreign key to the related resource.)

(Full res: https://drive.google.com/file/d/0B5r8wxmmppbQNTdGczg3cXZMLTg/view )

My question is: has anybody done this type of relations before or has an idea how to do it? 

Arches is currently written to have only relations between two complete resources, not between a node of a resource and a complete resource.

I get the feeling I should override the create_resource_relationship() function in Arches' resource.py, but perhaps I'm looking at it the wrong way. 

Thank you in advance for any suggestions,

Vincent

[Edmund Lee]

Fascinating request Vince, thanks for raising it - I'd be interested to see how Arches handles this use case as well. It's a common heritage issue - either competing or equally valid interpretations of a given artefact or site based on the (always) limited evidence that archaeology provides. From my recording experience (way back when) a typical problem might be that aerial photography has identified a site as a crop-mark, but we don't have any dating evidence (so in the U.K. it could date from anywhere between say 4000 BCE or 2000 CE). It could be an IRON AGE / DEFENDED ENCLOSURE, or it could equally be a MID TWENTIETH CENTURY / ANTI AIRCRAFT BATTERY (a rather extreme example, but possible). For a user coming to that record, it would be helpful to have insight into which authority had suggested which interpretation.

Ed

[Carlisle, Philip]

Hi both,
I think this is an example where the CRM-Inf extension would come in handy. See http://www.cidoc-crm.org/crminf/

 

Phil

 

Phil Carlisle

Knowledge Organization Specialist

Listing Group, Historic England

Direct Dial: +44 (0)1793 414824

 

http://thesaurus.historicengland.org.uk/ 

http://www.heritagedata.org/blog/

 

Listing Information Services fosters an environment where colleagues are valued for their skills and knowledge, and where communication, customer focus and working in partnership are at the heart of everything we do.

--
-- To post, send email to arches...@googlegroups.com. To unsubscribe, send email to archesprojec...@googlegroups.com. For more information, visit https://groups.google.com/d/forum/archesproject?hl=en
---
You received this message because you are subscribed to the Google Groups "Arches Project" group.
To unsubscribe from this group and stop receiving emails from it, send an email to archesprojec...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

We help people understand, enjoy and value the historic environment, and protect it for the future. Historic England is a public body, and we champion everyone’s heritage, across England.

Follow us:  Facebook  |  Twitter  |  Instagram     Sign up to our newsletter     

For the first time, we are opening up The List asking people to share images, insights and secrets of these special historic places to capture them for future generations. Can you help us #ListEngland?

This e-mail (and any attachments) is confidential and may contain personal views which are not the views of Historic England unless specifically stated. If you have received it in error, please delete it from your system and notify the sender immediately. Do not use, copy or disclose the information in any way nor act in reliance on it. Any information sent to Historic England may become publicly available.

[Dennis Wuthrich]

Hi Vincent,

Great question.  We are thinking about a very similar use case and have identified an extension to the CIDOC CRM (the inference extension http://www.cidoc-crm.org/crminf/ that Phil Carlisle mentions) that will be included in version 4 of Arches.

The basic summary is that the drm-inf extension will let you create a branch that says "Actor X assigns a value of "Iron Age" to the cultural period with a certainty of 80%".  The certainty value can be numeric as in the example, or a concept (such as "likely", "very likely", or "uncertain").

To support this, Arches version 4 will support the ability to reference another resource instance in a node, as you rightly point out isn't currently possible.

Hope this helps,

Dennis  

[Adam Cox]

Hi Vincent, I've had a similar idea before. In some cases, it would be nice to make a dropdown in a Heritage Resource form that was dynamically populated with all of the, let's say, Actor resources. When a user chose an Actor from the dropdown, a relationship would be formed between the Resource and the Actor. This is essentially just a subset of the existing related resource menu. You could further create a new relationship type ('assigned ____ attribute") and use that whenever relationships are made through this dropdown. Of course, that wouldn't work for various assignments of the same attribute by different entities.

To get closer to what you are thinking, this may be hacky, but here's a thought: Use your altered resource graph with a similar dropdown as described above. When a user chooses an Actor, instead of making a relationship (or perhaps in addition to making a resource) the UUID for that actor is saved as a string in your new node. Then, in the report for the Heritage Resource, you would list the assignment, and below that use the UUID to construct a hyperlink that points to the Actor resource.

Great question overall, looking forward to hearing what you end up trying out.

Adam

--
-- To post, send email to arches...@googlegroups.com. To unsubscribe, send email to archesproject+unsubscribe@googlegroups.com. For more information, visit https://groups.google.com/d/forum/archesproject?hl=en

---
You received this message because you are subscribed to the Google Groups "Arches Project" group.

To unsubscribe from this group and stop receiving emails from it, send an email to archesproject+unsubscribe@googlegroups.com.

[Vincent Meijer]

Thanks very much everybody for your replies. 

The Inference extension looks great. I do think that at the moment it is overkill for what I am trying to do, but for expansion in the future it seems very suitable.

@Adam, As for the foreign keys as strings in nodes, for me that sounds too much like a hack :) 

Not only will strings as IDs cost more resources when searching (compared to integers), but you will also loose the awesome universal relationship discovery promoted by CIDOC and the like (since you will need to know where to look to find these relationships).

For finding actors/references related to a single decision node in a heritage resource, speed wouldn't be a problem. However when you want to see all decisions an actor/reference is involved in, you would need to get all resources in your database, look in all nodes where you make these kinds of foreign key references and see if the string matches your actor/reference. Trouble! :)

Would it be a better solution to allow these kinds of relations in the resource_x_resource table?

Let me know what you think!

Oh by the way, is there a more concrete release date for arches v4?

I would love to skip what I am doing and start fresh from v4, since I have not done that many modifications to my v3 code (mainly ripped off HIP). 

However, I do need to get something up and running sooner rather than later.

Vincent

--
-- To post, send email to arches...@googlegroups.com. To unsubscribe, send email to archesprojec...@googlegroups.com. For more information, visit https://groups.google.com/d/forum/archesproject?hl=en

---
You received this message because you are subscribed to the Google Groups "Arches Project" group.

To unsubscribe from this group and stop receiving emails from it, send an email to archesprojec...@googlegroups.com.