Sardis

[elton barker]

greetings all,

While Rainer has been hard at work updating the Pelagios database structure and API (news on this will follow in due course), I've just started working with The Open University's "OpenLearn" team (http://www.open.edu/openlearn/history-the-arts/history/classical-studies) to build some resources around the capacity that we now have to link between each other's resources, which will be of interest to the general public.

Basing our work on Hestia's resources, we will be creating a 'story' around the opening episode of Herodotus, the story of Croesus. Not only is it a fun episode to focus on in its own right and has themes that resonate with today's world; it introduces the reader to the 'geography' of Herodotus's world - especially the critical mediating role of the oracle at Delphi and the main players in the later conflict (Persia/Media, Sparta, Athens, etc.).

I have 2 questions/challenges to this group:

1. My general question/challenge relates to the way each of us present our data. This is a topic that has been raised before, but it's only going to become more important as the Pelagios network grows. Suppose, for example, that I'm interested in the place "Sardis" and I happen upon the Pelagios API:

http://pelagios.dme.ait.ac.at/api/places/http%3A%2F%2Fpleiades.stoa.org%2Fplaces%2F550867

I see that there are a fair number of datasets that reference Sardis - but it's not obvious to me (without prior knowledge of the projects) what those datasets hold. To take first dataset as an example (just because it's first and not because I wish to single this one out), what does "Arachne" mean to me as someone who doesn't know anything about it? 

So, the first challenge is: can we find a way of making it clearer to a general audience what the data are that each partner holds? (Would e.g. icons be possible, to indicate that the datasets are texts, coins, buildings, etc., or a combination thereof?) And, similarly, once I've clicked on the link, is the information that I find human readable?

2. My specific question/challenge follows on from my impression that it's still difficult to discover precise information, particularly temporally relevant information (I still struggle at least!). Given the activity that we'll be creating, do any of you hold in your data interesting information that might be of relevance? E.g. do you have coins showing Croesus's head, or any photos of Sardis or Delphi, etc.?

Thanking you in advance!

cheers

elton

[Ethan Gruber]

What if an annotated thing contained a property linking to a Getty AAT URI that expresses the thing's format, genre, or subject?

--
You received this message because you are subscribed to the Google Groups "pelagios" group.
To unsubscribe from this group and stop receiving emails from it, send an email to pelagios-proje...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[elton barker]

I'd like to see what that looks like. ('scuse the ignorance, but what's AAT?) But it sounds like the kind of thing I had in mind.

[Rasmus Krempel]

Elton is right. The Enduser will propably be less intrested in the "Data source" alone. There will be no one saying "If it comes to the Ancient World, I only use Ure Museum/ORBIS/Nomisma/etc!"

Arachne is a difficult case. It has multiple types of Resources stored. There would be a lot of Icons needed.

For a general audiance: you could tag the sources or Groups of Sources , namely Void Subsets, with DBPedia. Maybe there is a way to Express something like :

http://arachne.uni-koeln.de -"Contains"-> http://dbpedia.org/page/Ancient_Greek_sculpture 
http://arachne.uni-koeln.de -"Contains"-> http://dbpedia.org/page/Book 
http://arachne.uni-koeln.de -"Contains"-> http://dbpedia.org/page/Picture

Everyone could Lookup the "Definition" in Wikipedia. The Advantage would be :
1. Its commonly used and recognized

2. Its understandable for a general Audiance

3. You can easily dereference the URIs to Wikipedia

4. There is a Name for nearly everything (like there will be nearly everything in the Ancient World, except for cars and Airplanes etc.)

5. There are also example pictures

6. multi language support ;-)

7.I bet nearly every Project partner knows/has read the Wikipedia articles about the Data he holds. So Tagging will be easy.

This would be my advice for a general Audiance. For more specialized Audiences, a professional reference system would be the chioce. Like Getty AAT is, as far as i know.
The special tool and Reference system would be more a question of a cosensus that will also fit future Partners etc.

[Leif Isaksen]

I think this is a significant issue and a really important one to
tackle to make our every growing cloud usable at scale.

My only concern about Wikipedia and AAT (that's the Getty Art and
Architecture Thesaurus to you, Elton ;-)) is that they are so large we
are likely to see a lot of fragmentation in the way we each tag
content. On that level, a small, curated range of categories might be
the way to start. A big advantage of the AAT, however, is that it is
hierarchical, so that would allow people to be as fine-grained as they
want, while still allowing, e.g. simple search APIs to offer data
according to a shallower taxonomy. Is the AAT available as Linked Open
Data yet? (I imagine it must be in the pipeline).

The TaDiRAH Object Taxonomy also looks useful and might be more
tractable in the immediate term.
https://github.com/dhtaxonomy/TaDiRAH/blob/master/reference/objects.md

I think that Elton;s comment about icons is also extremely important.
I'm quite impressed with the Noun Project myself, but the Open Icon
Library might be useful and I'm sure there are others:

http://thenounproject.com/
http://openiconlibrary.sourceforge.net/

Is there a way we could match up icons to a subset of one the
taxonomies above as a first pass?

Cheers

L.

[Ethan Gruber]

Yes, the AAT was the first Getty thesaurus to be released in LOD, and the others will be released every six months: http://vocab.getty.edu/aat/