Eddy’s Sofa And The Nightmare Of A Single Global Places Register

“Eddies,” said Ford, “in the space-time continuum.” “Ah,” nodded Arthur, “is he? Is he?” 
“What?” said Ford. “Er, who,” said Arthur, “is Eddy, then, exactly, then?”

Why,” he said, “is there a sofa in that field?”
“I told you!” shouted Ford, leaping to his feet. “Eddies in the space-time continuum!”
“And this is his sofa, is it?” asked Arthur, struggling to his feet and, he hoped, though not very optimistically, to his senses.

Jump onto Eddy’s sofa for a moment and fast forward to a possible 2015. 

After the location wars of 2010, the problems of mutually incompatible geographic identifiers have been solved with the formation of the Global Places Register. Founded by a fledgling startup on the outskirts of Bangalore, the GPR offered an open and free way for individuals and corporations to add their town, their business, their POI. All places added became part of the Global Places Translator, allowing Yahoo’s WOEIDs to be transformed into OpenStreetMap Ways, into long/lat centroids, into GeoNames ids or even, for the nostalgic, Eastings and Northings.

With this thorny problem solved, location issues became a thing of the past, no one used the phrase ubiquitous location any more and location really was a key context. 

Until 2014 when, inspired by the move from CDDB to Gracenote, Global Places Pty Ltd promptly ringfenced the entire database under a commercial license and added a mandatory application id to all their APIs. An application id that cost a lot of money in licensing.

Of course, the fact that the GPR was now the sole global source of places doesn’t mean that it’s complete or authoritative. Being controlled by a single corporation, the GPR is an easy and obvious target for hacking and litigation. A dedicated team spends all their time removing places and POIs from the register that have offensive or religious connotations, either as a result of user generated contributions or malicious hacking.

Sadly, the cities of London in Ontario and Ohio are no longer found in the GPR; they’ve had to rename themselves after a successful trademark and copyright case brought by London in the United Kingdom, which is now the sole place in the world allowed to call itself by that name. This does make geodisambiguation a bit easier though.

Likewise, you won’t find any newsagent or newsstand POIs in the GPR; they were all removed following a successful DMCA takedown by the League of Concerned Conservative Fundamentalist Parents, who didn’t want their precious offspring to be able to locate purveyors of potentially offensive adult material on their GPR powered LBS apps on their Android mobile internet tablets.

The same goes for any business with the word jolly in it; they were all removed and forced to rebrand following the success of Jolly Jet trademarking the word jolly and then pursuing an aggressive litigation program against any business unfortunate enough to have that word as part of their name.

This is all just a geographical bad dream. It’s not real. Wake up. Back to 2010 now, safe and secure in the knowledge that this could never happen …

Yet there is a growing clarion call for an open global database of places, POIs and business listings that will allow all of the disparate geographic identifier systems to be rationalised and used interchangeably. It’s the problem of what I’ve started calling GeoBabel but an actual global database of places isn’t the solution to this problem.

Tyler Bell recently commented in an O’Reilly Radar post that an actual (open) database of places is the preferred (long to mid term) outcome. But experience has taught users of geographic data sets that there is no one single authoritative data source for their needs, except in the most facile of use cases. Most users instead aggregate data sets, using them to obtain at least partial coverage for their needs; this despite it being forbidden in most commercial data vendor’s licenses. In a similar vein, existing users will already have made a commitment, both financial and technical to tagging and using geographical references of a certain sort, be it long/lat or a place based id.

For this reason alone, the most likely outcome will be what Tyler calls an “open and accessible Local Listings Crosswalk, essentially an API that take your URI and translates it to my URI so we can ensure we’re speaking about the same (place or business)“. While this is Tyler’s bet for a short term solution, if built, this will allow the data set creators to focus on what they do best, creating the data for their market segment or niche, without the inevitable intellectual property concerns that contributing data to such a project would raise.

Despite the success of OpenStreetMap, solely crowdsourcing the URI and id data is probably not the answer, though OSM would be a not unreasonable starting point for such a venture. Instead we need something that fulfils the role of this crosswalk, translating URIs and identifiers in a common, open manner.

Let’s call it the Geosetta project, the Geographical Rosetta Stone …

There can be no single owner; there can be and should be a shared set of contributors, from the crowd, from the corporate world and all points in between. But let’s not get bogged down in a long winded debate over data formats, endpoint URIs or the relative merits of JSON vs. REST.

It needs to be open, it needs to be discoverable and as Flickr’s Kellan Elliot-McCrea correctly points out, you “(need to be able to) get out, via the API, every single piece of information you put into the system“.

The location marketplace is moving at a phenomenal speed and the location industry needs to keep pace. If we don’t then a single, commercial solution may just rise to the fore (Steve Lee of Google was recently quoted as saying “(we’re) interested in solving the problem but it’s not easy“) and maybe the Global Places Register will be one step closer to reality.

Random footnote: The people I asked to read this post before I posted it seemed to think that Global Places Pty Ltd was a thinly veiled reference to Google. It’s not. It’s a thinly veiled reference to what seems to scare the heck out of the industry at the moment and that’s a single, dominant player who holds all the cards and who you have no choice but to deal with. At the moment, this is a strange and bizarre amalgam of FourSquare, Twitter, Google and Facebook but this will undoubtedly change as the future unfolds.

Photo credits: Hell-G and skëne on Flickr, under a Creative Commons license

One thought on “Eddy’s Sofa And The Nightmare Of A Single Global Places Register

Comments are closed.