[OHM] Mapping what's on the ground and other good practices

[Rob Warren]

Call me crazy, but to have a hypothetical location for Atlantis on the  
map is ...intriguing.

My plan for Muninn is to do exactly what you suggest for maps. If you  
plan to use linked open data, can I ask for your opinion on hooks you  
would like to have. I've already added a sameAs tag [2] so that  
external triples can be referenced from within OHM. Similarly, bulk  
upload is something that should be encouraged by the adventurous and  
is a good way to keep the ball rolling; perhaps you want to create an  
initial upload from which your users can bootstrap their efforts?

My personal preference is for some kind of citation / source /  
documentation as a standard of verification. That does not mean that  
the data will always make sense or have a unique solution, but I think  
it is a good start.

I added a number of pages to the Wiki about possible solutions to  
concurrent hypothesis and date stamps according to the past emails in  
the mailing list. [1] Feel free to modify and add on to them.

best,
rhw

[1] http://wiki.openstreetmap.org/wiki/OHM/Plan
[2] http://wiki.openstreetmap.org/wiki/OHM/LinkedOpenData

> Date: Fri, 1 Mar 2013 10:08:31 -0700
> From: Sean Gillies <sean.gillies at gmail.com>
> To: historic at openstreetmap.org
> Subject: [OHM] Mapping what's on the ground and other good practices
> Hi all,
>
> I'm very excited about OHM developments, there's enormous potential
> here. I manage a site and dataset about places in the Greek and Roman
> world (plus some older Ancient Near East places, some Byzantine
> places) called Pleiades. Users continually ask me about adding
> detailed map data to Pleiades ? locations of monuments, buildings,
> walls, and streets ? but this is really outside the scope of my
> project. I think that OHM is possibly the better destination for such
> detailed data. And I think getting archaeologists and other
> researchers involved here could be good for OHM. Imagine the Penn
> Museum's maps of Ur
> (http://www.penn.museum/blog/museum/ur-digitization-project-february-2013/ 
> )
> in OHM. Or Eric Poehler's maps of Pompeii (http://www.pompeiana.org/).
>
> A major question for me: will OHM reflect past reality on the ground?
> And if so, what will the standards be? For example, say I create in
> OSM (the current OSM) a new continent in the Atlantic Ocean and name
> it "Atlantis." This is fiction, of course, but only determinable as
> fiction because we can visit that part of the ocean today by boat or
> plane, or virtually by satellite and falsify the assertion of its
> existence. Past features aren't so easily verified or falsified and
> their nature is essentially hypothetical, only approaching the
> factuality of existing features after much study. To restate my
> question: how good must a hypothesis about an ancient feature be to
> warrant its inclusion in OHM? Hypothetical lost civilizations of
> Atlantis abound despite lack of evidence ? including these in OHM
> would be a departure from OSM's principle of reality on the ground, at
> least in my view.
>
> I've assumed that OHM would adopt and adapt OSM's best practice
> rubrics. Looking at http://wiki.openstreetmap.org/wiki/Good_practice,
> I think it would be worth amending (for OHM)  "Map what's on the
> ground" to "Map strong and falsifiable hypotheses about what was on
> the ground" and develop a practice of citing research and historical
> documents. Nodes and ways of
> Old Babylonian Ur can cite published work. If I trace the hypothetical
> trail over the Alps that Hannibal's army left in its wake, I feel like
> I ought to cite evidence supporting it.
>
> I realize that showing is better than telling, and I'll try to do some
> leading by example when the OHM database is ready to go.
>
> -- 
> Sean Gillies