Dear Francesco,

in a paper published one year ago on International Journal of Digital Libraries 
written by myself, Achille Felicetti and Paola Ronzino we addressed a similar 
issue concerning Pleiades, i.e. the mapping of their geographic entities to the 
CRM.
I am sending you separately the paper as you can find there the details of the 
solution we proposed.

In short, the general solution we suggest is to consider “real” Pleiades places 
(as opposed to imaginary, unknown or non-existing places) as instances of E92 
Spacetime Volume that P161 has spatial projection on an E53 Place, which in 
Pleiades terminology is called a location. Notice that this E53 is the widest 
portion of space occupied by the related E92 in its life. If you want to 
consider the location extension at some specified instant, or during a 
specified time-span, like “Rome in the 19th century”, you need to slice the E92 
at that time-span and obtain a derived Spacetime Volume, which the CRM calls 
E93 Presence (in my opinion, with a poetic licence), whose spatial projection 
gives the spatial occupation in the specified time.

With geonames the situation is probably simpler as it considers, if I am not 
wrong, time-snapshots of places taken at the present. Actually in the paper we 
considered a more complicated case where the “place” may not be, or be known to 
be, a real one.

As regards the geonames definition of PPL populated place as "a city, town, 
village, or other agglomeration of buildings where people live and work”, it 
seems to me inconsistent, vague and contradictory. 
When is an assemblage of housings (and factories, as it seems necessary) enough 
agglomerated to be a PPL? For example, are a PPL the examples of “diffused 
urbanization” (called so by land planners), typical of some areas, e.g. the 
North-East of Italy, characterized by the lack of agglomeration? Is an area 
populated by pensioners only a PPL, as by definition its inhabitants don’t 
work? What about areas inhabited by commuters, who work elsewhere? How many 
people are necessary to rank a place as a populated one?
This definition may perhaps be used for practical purposes, and when such 
populated places are defined (and named) by administrative rules. 

Also, in my experience geonames is impractical as a gazetteer for cultural 
heritage, history or archaeology. Unfortunately the much more appropriate 
Pleiades covers only until Early Medieval geography - I don’t know if there is 
anything similar for later periods.

I think that the above answers to your questions; it may need some further 
refinement if you are interested in associating to the E92 its dwellings or its 
inhabitants. 
For the buildings B1, as they can rapidly change, be built or disappear, the 
link between the E92 and the B1s is through the space both occupy, the E53 
Place spatial projection of E92 and the location of each B1: a B1 Built Work (a 
building) P53 has former or current location E53 Place (the building lot) that 
P89 falls within E53 Place (the spatial projection of the E92, i.e. its 
location). 
For the people, a quick-and-dirty solution is P74 has current or former 
residence at the E53 Place, location of the PPL; but I would prefer the 
following construct: the E92 (the PPL) P11 had participant E74 Group (the 
citizenship), considering all inhabitants as participating in the existence of 
E92. 

It is interesting to note that with a little improvement the above would avoid 
anachronisms as buildings could be related to a town only for the time of their 
existence, which must be part of the E92’s time-span to which they are related 
through P160 has temporal projection. Interesting to think what about ruins 
e.g. in archaeological sites: they are still related in space, but not in time, 
as they survived to the town destruction; maybe also B1 should have a time 
dimension...but that’s anothe story. 
Same for people's participation (= living in) towns, with of course more 
mobility.

Best regards

Franco

Prof. Franco Niccolucci
Director, VAST-LAB
PIN - U. of Florence
Scientific Coordinator
ARIADNE - PARTHENOS

Piazza Ciardi 25
59100 Prato, Italy


> Il giorno 25 ago 2017, alle ore 23:54, Francesco Beretta 
> <[email protected]> ha scritto:
> 
> Dear all,
> 
> One of the basic place type in geonames.org is :
> 
> 
> PPL    populated place    a city, town, village, or other agglomeration of 
> buildings where people live and work
> 
> 
> A populated place in this sense means different things : 
>       • an agglomeration of buildings;
>       • a community of people;
>       • the projection of them on the Earth surface.
> 
> My question is: how should we model these entities and their relationship ?
> 
>       • an agglomeration of buildings -> E24_Physical_Man-Made_Thing or, more 
> precisely, B1 Built Work ?
>       • a community of people -> E74_Group or E4_Period
>       • the projection of both on the Earth surface -> E53_Place
> 
> "P156_occupies"  links  E24 and E53: but the spatial footprint of E24 (the 
> agglomeration of buildings) can change over time: new, larger city walls are 
> built, etc. 
> Are these different places (E53) which are related to different time-spans ? 
> Or the same place but in a more abstract sense ?
> 
> 
> And also: which class is suited for modelling 
> "FRST    forest(s)    an area dominated by tree vegetation" 
> or 
> "VINS    vineyards    plantings of grapevines" ?
> 
> And for mountain or valley ?
> 
> Can we model all this just with E53 Place and a vocabulary of types ?
> 
> Were can I find some examples for modelling these different kinds of 'Places' 
> using the CRM ?
> 
> Thank you for some hints !
> 
> Francesco
> _______________________________________________
> Crm-sig mailing list
> [email protected]
> http://lists.ics.forth.gr/mailman/listinfo/crm-sig


Reply via email to