Maybe TransforMap is a Gazeteer, too?

(Jon Richter) #1

In thinking TransforMap as a spatial data infrastructure (SDI) with specific requirements (Geometry storage, Metadata description, Commoning process, to only name a few), we’re constantly confronted with de facto standards of the digital geo domain.

From Who’s on first?

An interesting way to think about a gazetteer is to consider it as the space where debate about a place is managed but not decided.

Long version is long. You might want to get a cup of coffee or maybe a drink if you’ve been thinking about this sort of thing for as long as we have (or maybe longer).

What is a “gazetteer” anyway?

Especially the first sentence of this excerpt led to repercussions in my mind, considering the geospatial semantics we are picturing for alternative economies and the associated discussions we are proliferating.

After reading the whole piece, have you had a look at their collection of place types, which would be our classes, if any?

Offering pixelHumain's Communecter to Shareable for their Map Jam needs | Evaluating open data of cities in the United States of America
2015-12-03 unsceduled Monkey mumble-meeting
2015-09-28 | Mapping the Commons + Commons Institut | Wikimedia
Geodata storage evaluation
(Thomas Kalka) #2

Würdest du bitte nochmal kurz versuchen in deutsch zu erklären, was du hier denkst ?

(Thomas Kalka) #3

Ich verstehe nicht genau, was die da machen.
Das sieht erst einmal aus wie eine Hierarchie von Gebieten.
Von einer Sammlung von “place types”, die wir für transformap bräuchten, ist das noch weit entfernt, oder ?

(Jon Richter) #4

Linkdropping. Getting back later.

(Matt Wallis) #5

Very interesting.

The canonical source for a place is a text file, specifically GeoJSON with a unique 64-bit numeric ID. This is because all computers speak “text files” and “numbers”. Text files can be inspected or updated in any old text editor. Text files can be printed. Numbers are fast and cheap for databases to index.

I like this approach. Permanent unique identifiers are built in from the start; Text files are the source (here they are on GitHub: - other representations can be generated from these (e.g. geo-aware database) as required (e.g. for performance).

Every WOF record contains a property called wof:concordances which is a dictionary of simple key/value pointers to the identifiers of the same in other databases.

"wof:id": 101736545,
"wof:concordances": {
    "fct:id": "03c06bce-8f76-11e1-848f-cfd5bf3ef515",
    "gn:id": "6077243",
    "gp:id": "3534"

The concordances provide identifiers of representations of the same object within other data sets.

I also like the way that they provide explicit mechanisms for dealing with challenges that come up regularly in geo-data. For example,

Superseded / Superseded by
One of the large and difficult philosophical questions that working with geography raises is this: How do you decide when something has simply been updated versus when has something fundamentally changed?

Good to recognize this difference explicitly.

Hierarchies in Who’s On First are represented as a list of dictionaries, where each item is a dictionary containing a full hierarchy.

The crucial point here is that hierarchies are represented as a list of dictionaries. This means that different perspectives about where a place fits into the hierarchy can be allowed - the ambiguity (or difference of opinion) does not have to be resolved before the data can be used.

(Thomas Kalka) #6

@mattw, thanks for bringing this up again.
After reading more thoroughly more interesting design patterns are to mention:

  • “quality” measures inside data
  • named alternative values (“name:language_preference”, different coordinates for different purposes)
  • data first (start project by publishing data without apis)

Ah, I did not know, that github displays geojson nicely.

Proposal for converging Semeoz work and Transformap wiki of maps
Humanitarian aid and International development
(Jon Richter) #7

Since you wrote this it is repercussioning in my head.
Thinking in the most simple solution to a problem is also called The Rule of Least Power.