The only article you must read today: Mike Dobson's analysis on the tools and tactical mix Google might use to update its maps after walking away from Tele Atlas.

If you aren't flattened after this, drill it deeper with "Google And Map Updating" Part I and Part II. Beware - this a long read and well worth every line.

Mike Dobson hits the nail. Check the comments as Mike Moore (with Navteq Developers Network on his day job) is commenting and rising a dispute that's long overdue.

Argleton is a place that doesn't exist - except on Google Maps.
The story is developing since about a week, brought up by The Telegraph:

Google and the company that supplies its mapping data are unable to explain the presence of the phantom town and are investigating.

Tantalisingly, “Argle” echoes the word “Google”, while the phantom town’s name is also an anagram of “Not Real G”, and “Not Large”.

One theory is that Argleton could have been deliberately added, as a trap to catch companies that violate the map's copyright, The Telegraph states. If that was true, statements beyond would contradict - as a mapmaker you'd better know where bogus data is. Google is clueless and Tele Atlas said it would now wipe the non-existent town from the map, quote:

"Mistakes like this are not common, and I really can't explain why these anomalies get into our database."

As the story from The Telegraph is from Oct31 it'll be intersting to see when Argleton will actually be "wiped". One more week?

... says Robert Thomson, managing editor of the Wall Street Journal, on WSJ.com’s plans to move into the local space.
Oh yeah, and they’re going to experiment with micro-payments, too. (quote via Cory Bergman's Lost Remote, Poynter.org has an excellent transcript.)

"Google devalues everything it touches. Google is great for Google but it's terrible for content providers. (...) Google doesn't distinguish between the quality of the content around which it serves up ads, it is concerned with quantity rather than quality."

Mr. Thomson has been editor of News Corp.'s Times of London, was brought in as publisher to the WSJ by Mr. Murdoch after the completion of the $5.16 billion acquisition of Dow Jones in December 2006.

Apparently, Thomson is expressing more than just his personal opinion: "Aside from the six years working together on the London paper, the two men have much in common, though Mr. Thomson is 47 and Mr. Murdoch 77. Both are Australian natives, have Chinese wives and are raising young children. They have vacationed together and even share the same birthday, March 11." (see source at WSJ)

If "going local" is Mr. Murdochs' new mantra - how would the imperium outmaneuver Google Maps' global dominance in mapping the local space?

We had the story earlier in June last year:

Google Mapping Nairobi inhouse ... what's next?

Ed Parsons has a story based on news from NetworkWorld.
Googles' recently opened Nairobi office suspectedly didn't find its own location on Google Maps - and kick-started a new Nairobi city map that didn't exist before. (...)

In June 2008, our open question was: "Will Google follow the OpenStreetMap or is this a one-timer?"

Yesterday, the Mapping News Blog reports (via Slashgeo, kudos!):

"Google are now allowing you to download the vector data (Kenya only so far) in either Keyhole Markup Language (KML) or in ESRI Shapefile format.

Google seems to now return data from Google Map Maker contributors to the public. It's only those datasets that can be released because bindings with Tele Atlas permit for only this. Anything else effectively would kick TA out of business. As the idea behind Map Maker was to engage users to map areas of the World where the duopol is limited or non existent has proven to work regionally and in Nairobi.

Anyway, only non-commercial-non-profit entities seem to be legally entitled to use the data. The fundamental issue of "Who owns the data?" isn't solved - just shifted.

Visit the original site as it has a lot of more information.

update: Mikel Maron at Brain Off has a deeper analysis and insight: "First Thoughts on Google MapMaker Download"

... at Google Maps. We really understand your "ongoing efforts to partner with public sector organizations to add their map content to Google Maps and Google Earth" because indeed and "by definition, organizations like local governments are the most authoritative source of geospatial data for their jurisdiction". Coincidentally this is, what we do: use the best data available, merge it to make even better maps.

So your and Google's goal is the "sharing of geospatial data".
Does it encompass the mutual sharing of data, of geodata at large? Do you, Google, share data with us?

As we encounter it at United Maps - and despite the fact that geodata is already paid for by taxpayers' money - the public sector has a delicate pricing and proposition on selling geodata to the private sector. Put short: it's shamelessly expensive to buy that data all around in Europe.

You, dear Google, you may afford.

An interesting point though:

If you run the numbers for just the U.S. where there are many federal agencies with geospatial data, 50 state governments, some 3,000 counties and over 30,000 cities and towns, you quickly get an idea of the volume of relationships you'd have to develop and manage to add data from all governments (...)

Compare this to Europe:

The picture is somewhat different:

These are 33 federal surveyors agencies instead of one U.S. TIGER. In the U.S., TIGER is free. In ol' Europe, data costs money, and the sheer number of cities and towns quickly more than doubles up.

Frank Taylor recycles some of his good posts like this "About Google Earth Imagery", spiced up with further detail (that I don't detect easily). The baseline in Franks original article was "how many people initially think Google Earth will show imagery in real-time. Or, that surely it will only be a day old." 

Anyway, the post made me think about ways to check the real original date of satellite and aerial imagery. I'm sure there was a pointer to at least one of the larger mapping services but I am unable to recall. Any help appreciated!

Fueled by recent debates about the precision of our own data based on primary datasets from Navteq or Tele Atlas, I almost accidentally found a slick and straightforward site made by Ryan Jonasson that simultaneously geocodes on two invisible side-by-side windows displaying Google Maps/TA and Microsoft Virtual Earth/NVT:

Try it on your own: it's a park in Paris, France, search for "Ruisseau de Gravelle". Nothing wrong on the first glance.

If you'd zoom in, you quickly determine a deviation of a whopping 157m between the satellite imagery and the hybrid street layer:

Is this a random occurence or is it a systematic dispatch?

We thought, the quality thresholds and precision expectation would be much better than 157m - at least this is, what we prepare for.

The otherwise well-informed GPS Business News, run by Ludovic Privat is one of the rare industry services that we happily pay for. Besides Ludovic being something like a Rolodex on legs, his broad industry experience fuels insightful analysis. Most of the time at least.

In his last piece and from our decent point of view, Ludovic gets it substantially wrong by titling "Google: building its own map data, step by step - Integration of user generated data into Google Maps gets started in 16 countries".

Why?

The indication Ludovic cites is that "Tele Atlas has been offering (Google) a detailed coverage of Morocco for over 6 months; however, Google is not using it."

Well - nobody knows about the data delivery deal those two have, neither the financials or other requirements are publicly known. Maybe Tele Atlas' data isn't sufficient from Google's perspective or parts of the Tele Atlas map world are excluded from the deal or whatever legal or technical complication may be around. You know how things may become difficult if two elephants meet.

In fact this is rather isn't a new information nor does it make a point.

We had it for Nairobi, we discussed the rights Map Maker contributers at Google Maps may retain (none) and we shed light on how Tele Atlas would potentially benfit from use generated Map Maker additions besides their own user driven change requests. Finally, O'Reilys infamous Brady Forest provided valuable context giving a clear interpretation and put it all together, much better than we could.

So where's the beef?

There still is a sincere dinstinction between gathering GPS coordinates and serious maps.

Besides all the bells and whistles of wikiesque approaches and OpenStreetMap, there always will be enough headroom for professional mapmakers, comprehensive products and reasonable coverage to substantially distinguish from crowdsourced maps. We encounter the difference each and every day by interacting with datasets from both Navteq and Tele Atlas - and OpenStreetMap and official cadastre sets. There a hilarious omissions on any side and I'll post the best-of in the coming weeks until we launch our hyperlocal map of Germany.

Driving around in streetview cars doesn't generate better maps - it just generates more pixel data that cries for integration. Raster imagery doesn't really make better maps, it only gets you better looking maps. Or, as Dr. Joe Berry puts it "raster is faster but vector is more corrrrector" 

Driving around on streets per se generates maps for vehicles (and I won't elaborate on this for now).

Even if The Register was right last year posting "Britain to fall within a year", then the UK should be completely street viewed in autumn 2008:

(...) these boys are at times starting their engines at 7am, and covering up to 160 miles in a single day. Now, the UK's roads extend around 233,000 miles. If we assume Google has ten black ops vehicles working five days a week (call it 260 working days a year), covering an average of 100 miles a day, it will take the search monolith less than 12 months to complete its dark task.

This calculation works if the weather is OK. Even in the UK it rains, sometimes. Anyone taking a bet?

Now approximate that calculation to the rest of Europe: you do the math. It's pointless even if you'd want to map only highly populated regions.

Ludovic moves on with "Google is surely harnessing enough data to create navigable maps of these areas." - and again: Google yet isn't in the vehicle navigation business (as Tele Atlas and Navteq are), Google is in the search/find, the people navigation and finally in the eyeball business. Logically, street view imagery is a likely extension of Google's acclaimed mission "to organize the world's information", it is a waypoint, not the ultimate goal.

Besides all that, Google may be "in danger of losing its focus", according to California Assemblyman Joel Anderson who is proposing a new "security" measure that will require Google - and other providers of similar map data - to blur images of schools, churches, hospitals and government buildings.

Anderson's proposed bill would prohibit operators such as Google Earth from making aerial or satellite images of schools, places of worship, medical facilities or government buildings available to the public, unless those images have been blurred. (...) would also ban (...) from providing street view photographs or imagery of those buildings and facilities

Ludovics conclusion might be pointing to the wild:

"Nowadays the question is not IF Google intends to develop its own maps, but better WHEN they will be ready for prime time. "

Collecting unstructured user additions and change requests is one thing that Google seems to handle perfectly - integrating mountains of data points into fully navigable and structured vector databases is a completely different beast and yet unsolved.

Finally, the question is: if user generated maps are to become better than anything else - why doesn't Google just support Cloudmade and its APIs and replace Tele Atlas?

In its meteoric history, Google never produced data, it smartly gathered and aggregated data.

Guess why.

"(...) After admitting that she uses Google Maps, [new Yahoo CEO Carol] Bartz said Yahoo Maps isn’t as good as it should be because 'we haven’t paid any attention' to it.

CFO Blake Jorgensen, who resigned last week but is still on board, added that an online mapping service is very expensive to maintain, and that he doesn’t foresee Yahoo pouring significant additional investments into Yahoo Maps in the near future."

Ooops.

Strategy, politics or just a PR lapse? 

Google's recently released Latitude causes sort of a stir out there in the blogosphere.

Seems to work on both connected mobile handsets and desktops and it basically allows you to share your location with a set group of friends. Allow/Disallow: as simple or as complex as can be.

Brady Forrest comments that Latitude isn't tied to an application and only needs your GMail/GTalk contacts to work on. A slap towards the 200-pound social networks Facebook and MySpace: who if not them should have come out with a locate-a-friend-de-luxe version?

Besides Yahoo's announcement launching OneConnect almost exactly a year ago, CNET's Caroline McCarthy is quoted saying ...

“There’s a reason why no mobile socialnetworking company has broken out yet.
They haven’t found themselves — on a map, that is.”

How true this is! The maps haven't changed much since then - it's still "maps made for cars", and frankly — still the good ol' Navteq or Tele Atlas vehicle stuff, some house footprints manually added offshore.

Would you locate yourself on a bare bones street map or go for something much more sociable?

<END shameless selfpromo>