In early 2006 a patriotic Palestinian uploaded a collection of placemarks to the Google Earth Community pinpointing old Arab villages that were abandoned or forcibly cleared in the upheavals that followed the establishment of Israel in 1948. The post generated much discussion and many downloads, which meant that the attached placemarks were soon visible in Google’s “Best of Google Earth Community” layer, generated automatically from an algorithm for interestingness and visible by default on Google Earth.

As a result of this promotion, patriotic Israelis discovered that one of the several hundred placemarks, which puts the alleged destroyed village of Arab Ghawarina on top of the modern Israeli town of Kiryat Yam, was in all likelihood misplaced. Not amused by having this erroneous placemark appear above Kiryat Yam, the town in question filed a legal complaint in February 2008. Media reports from the time indicated that the Palestinian author of the layer, Thameen Darby, was open to correcting the location of Arab Ghawarina, but for some reason this never happened, possibly because the entire thread was soon locked down by the forum’s moderators.

The result was a campaign by some patriotic Israeli groups to get the offending placemark removed from Google Earth. On June 2008 The Jerusalem Center for Public Affairs accused Google of anti-Israel bias for not removing user-generated content from a default layer that the center felt was slanderous:

Generally, Google allows all kinds of organizations or individuals to create overlays with their own information on its map. These overlays are only available to those who specifically request them, but they are not automatically incorporated into the core map of Google Earth that every user entering its website can see. Disturbingly, Google has incorporated the Palestinians’ overlays and their accompanying narrative into its core maps of Israel. As Google maintains editorial control over its core layer, it has responsibility for its content, which it clearly has not adequately exercised.

The main problem with the above accusation of anti-Israel bias is that content for the “Best of Google Earth Community” layer was indeed “automatically incorporated”, based on an algorithm for interestingness without further human input by Google, and that there were in fact many, many other examples where user-generated content visible by default would favor the narrative of one group over another, and be promoted to default visibility simply based on the interest it created on the forum. An O’Reilly blog uncritically repeats the report’s accusations of bias as true.

In July 2008 the Zionist Organization of America (ZOA) wrote a letter to Google CEO Eric Schmidt and in a press release “condemned” Google for knowingly permitting Google Earth to “become a vehicle for promoting false and demonizing political propaganda about Jews and Israel.” Part of the letter reads:

Google has permitted anti-Israel propaganda to be elevated on Google Earth so that it automatically appears when Google Earth users seek geographical information about Israel; other content contributed by outsiders has not been treated the same way. It is shameful to see that Google has chosen sides and is knowingly and deliberately promoting anti-Israel falsehoods.

Again, the record shows otherwise. All other content was treated the same way, and so the accusation that Google chose sides does not stand up to scrutiny.

What is true, however, is that many people assumed that being part of the “Best of Google Earth Community” layer somehow amounted to an endorsement by Google. This was not the case, though Google never made that case particularly clear. Such a lack of clarity amid a growing sense that Google Earth has become the de facto geographic gold standard allowed accusations of bias to gain currency.

Fast-forward to the present day. Log on to Google Earth as of a few weeks ago, and you will no longer find a “Best of Google Earth Community” layer. Zoom into Kiryat Yam and you will no longer find the offending placemark turned on by default. Instead, you need to go to Gallery > Google Earth Community and turn on that layer manually. Then, indeed, the placemark for Arab Ghawarina becomes visible again.

Patriotic Israelis are claiming credit for the change:

Google is no longer in the business of delegitimizing Israel because we made our voices heard. Such efforts need to continue, calling into question every bit of misinformation and outright falsehoods that are published about Israel.

Interestingly, a Jerusalem Post article from Aug 31 explicitly calls the new default “Places” layer a more robust replacement for the now-defunct default “Best of Google Earth Community” layer. Nobody at Google is quoted on the record in the article as this being the intention, though I suspect it may have been sold as such to the reporter off the record. The article also contains other information that I haven’t been able to find on the record elsewhere:

Key to the new layer are special algorithms that corroborate information received through one source with the other sources. According to a company statement, this will make “it easier for users to learn about a given place through photos, videos, and annotations contributed by users around the world.”

But it will also allow Google Earth to automatically corroborate any information received from users before displaying it on the default layer. Only information appearing in more than a single source will be displayed in this layer.

On Google Lat Long Blog, Google made no mention of such algorithms when announcing the layer a few weeks ago. I’m a bit sceptical as to their usefulness, if they exist — the Panoramio photo and Wikipedia components of the Places layer have community guidelines that discourage tendentious speech in any case, but YouTube and Google Earth Community components are more accepting of such speech. I’m not sure how a popular but tendentious georeferenced YouTube video is supposed to be filtered out by an algorithm. Time will tell if the Places layer is geobomb-able, though patriotic Israelis can at least be certain that Thameen Darby’s layer of disappeared Palestinian villages is not in any danger of popping up: the places he references no longer exist, and hence won’t be getting any Places placemarks.

Overall, I’m happy with the change. I love the Google Earth Community layer — it is my favorite because of its generous and raw overabundance of both objective and subjective information, no matter where I am on the globe or how much I’ve zoomed in. It’s best that Google makes clear it is user-generated. I see the new Places layer as user-friendly alternative to the full-blown sensory overload that can greet new users turning on all the layers.

I’m less happy with the accusations of anti-Israel bias in Google Earth, based on the default presence of the Arab Ghawarina placemark, when clearly these were unfounded. The worst that can be said about Google in this episode is that it didn’t clearly label user-generated content as such. It would be great if those groups who made the accusations would now retract them, but part of me suspects that they weren’t made in good faith on mistaken assumptions about the inner workings of Google Earth. I may be wrong.

Is this case now closed? Don’t count on it. There is a part in ZOA’s letter to Eric Schmidt that has the potential to cause many more sparks to fly:

The problem of Google Earth is immediately apparent when users “fly to” the satellite map of Israel and seek geographic information about the state. As one news report noted, users “will get much more than a geography lesson.” Google Earth users are immediately greeted with two rectangles prominently displayed on the map with the words, “Every Human Has Rights.” When users click on these rectangles, they are treated to political propaganda about the suffering of Palestinian Arabs allegedly as the result of the so-called Israeli “occupation.” Users will reasonably, but wrongly, conclude that because this information appears by default on the Google Earth map, it is accurate and reliable, when in fact, it is nothing more than politicized propaganda that conveys falsehoods to Google Earth users about Jews and Israel.

Those Every Human Has Rights placemarks are still there by default, as part of the Preview layer. One is about a grieving Israeli mother whose daughter was killed by a Palestinian suicide bomber, and the other is about a man teaching Palestinian children non-violent ways of dealing with their predicament; but both placemarks also implicitly criticize the character of the Israeli presence in the West Bank, a perspective shared by many who are not anti-Israel or anti-semitic, like myself.

What’s interesting is that this layer, produced under the aegis of The Elders and Realizing Rights, was made in collaboration with Google Outreach and its contents vetted by all three organizations. If patriotic Israelis want to pursue the matter of Google’s perceived anti-Israel bias further, this is likely to be the most promising direction for them. (Full disclosure — I was responsible for the layout and design of the popups of the Every Human Has Rights layer.)

The 6th International Symposium on Digital Earth (ISDE6) has been announced, with a call for papers. ISDE6 will take place in Beijing in September 2009. This symposium’s topic: “Digital Earth in Action”. From an email:

ISDE6 aims to provide an international forum for world-class scientists, engineers and educators engaged in the fields of digital earth, earth observation, geo-informatics and relevant applications to review the progress of Digital Earth during the last decade and discuss the achievements of Digital Earth and the recent developments.

I’m not sure what the point is of holding such a symposium in a country whose government censors satellite imagery and maps with indiscriminate glee because it thinks its own citizens are too stupid to make up their own minds from uncensored information. It’s a bit like holding the World Summit on the Information Society in censorious Tunisia in 2005 — inappropriate, on the face of it, or at the very least ironic. There is always a chance that something like ISDE6 helps breach the great Chinese firewall, and I hope it does, but that certainly wasn’t the case in Tunisia.

My suspicion is that China’s leaders still see earth observation and geo-informatics as elite academic projects, while much of the online world has moved on to regard it as a mass-participatory exercise in unfettered mapping and mashing. The one big story in the field of GIS these past 3 years has been the democratization of access to geospatial data of the kind previosuly only available to governments or affordable to corporations. That lesson is still lost on those in power in China. I hope somebody says as much, bluntly, at a presentation at ISDE6.

So somebody wrote a press release yesterday about Google Earth Magnetic Cows, but it sounded so silly that I decided to go for celebratory end-of-panorama-trip cocktails instead of dirtying my blog with it. Imagine my surprise to see it hit the interwebs with a vengeance today, but the greatest surprise of all was that only one blog so far, The Earth is Square, has reported this “news” skeptically. [I just noticed Google Earth Blog is also skeptical:-)]

The bottom line: It’s entirely possible that cows have a magnetic sense, but we’re not going to find out about it by looking at Google Earth. What possible fallacies might there be? I came up with these in, oh, a few seconds:

• What cow? Have you ever seen a cow on Google Earth? Are you sure? Most of Google Earth’s imagery is not in high enough resolution to let you discern cow from car from tent from bear. And you certainly can’t tell which end is up, so to speak. In any case, your sample population of successfully found cows is bound to be miniscule compared to the total population.
• Now cow? The imagery in Google Earth is exclusively taken on cloudless days — or we wouldn’t have had the imagery. Are we sure the cows weren’t using the sun to align themselves, say to maximize solar exposure in winter and minimize it in summer? As the best satellite photos are taken when shadows are at a minimum, i.e. noon, that alone would lead to aligned cows.
• Stupid cow: Cows are walking stomachs. If they hit a fence while grazing, they will move along it. At most, this study tells us that fences have a slight preponderance to point North-South — and fences are even harder to see on Google Earth than cows.

So what happened? Either this story is an elaborate hoax, or it isn’t and these scientists are just not very good, or the story has been reported inaccurately. It’s a cow toss as to which of these explanations is the right one, but I suspect hoax, in part because of the study’s claim that it could discern a predispostition in cows for magnetic north over true north — a difference that is miniscule in most places on Earth. It doesn’t help that you have to pay the US National Academy of Sciences $10 to read the paper — that’s the price of a cocktail here in Stockholm, one which I’m about to go buy.

• Ogle Sweden update: Click on icons on the auto-updating map below to see the panoramas I’ve been taking this past week:

  
  View Larger Map

• Jailbird’s eye view: I’m pretty sure that prison life in Norway is, compared to the rest of the world, very nice. But should you ever feel the urge to escape Bjørgvin prison in Bergen, you can now consult a gigantic satellite view of the grounds on the floor of the prison building. The images are copied from Google Earth, and are painted onto the floor as part of an art project by local students. Here is the Flickr set that alerted me to it. I think it looks quite excellent:

  

• Surface in the wild: Microsoft’s Surface multitouch screen/GUI is now on display in several Sheraton lobbies in assorted cities in the US, and it features Virtual Earth (though not, apparently, 3D). Here is the Seattle PI’s video report. Virtual Earth makes its appearance half-way through:

  I want.

• Google Earth API update: From a few days ago: Google Geo Developers Blog: New Earth API resources: Complete API Reference, Sample Code, Release Notes, FAQ. Mac support is still forthcoming.
• Where is the path? Where is the Ordnance Survey?? Mapperz points us in the direction of Where is the path, a UK site that correlates locations on Ordnance Survey’s web maps with, among other things Google Earth for browsers. It looks fabulous on the video demo:

  Unfortunately, I didn’t get to try it myself, because:

  

  Way to go, Ordnance Survey. Your free product is unusable, by design. Just as predicted.

CatholicGauze is keeping track of the Russia-Georgia war, updating his post on Google Earth Community with battle locations as the war progresses. He also adds the route of the oil pipeline and the boundaries of the various breakaway regions.

Extremely high resolution version

Ogle Earth is changing to Ogle Sweden — vicarously, for the next two weeks, at least. I’m road tripping around Sweden, on assignment for the Swedish Institute, taking 360-degree georeferenced panoramas and blogging the process over on Sweden.se. The blog’s format is still a work in progress, but do check it out — it will be officially launched in the next day or two.

Currently I’m in Umeå, and heading for well North of the Polar Circle, before driving back down to the Southern coast and back to Stockholm via the Baltic islands. There is much to see and do — Sweden is not short of sweeping vistas, and not small either — so I haven’t settled into a routine yet that gives me much time for Ogle Earth; we’ll see how that goes. Perhaps I’ll just cross post from Sweden.se.

Extremely high resolution version

In addition to posting the panoramas to the blog, I’m also mapping the trip. I was planning to create some KML manually for each day, upload it to a server and into an instance of Google Maps, but now Barry Hunter has just made my life a lot easier, releasing goKML.net for Google MyMaps just as I was wishing for something like it. Now, I can just create a MyMap for each day and use the KML it generates for that blog entry’s map, but also collect all these MyMaps into one big aggregate trip map — with no duplication of effort!

Oh, and as Barry noticed, Google MyMaps now produces GeoRSS as well, in addition to KML.

(PS the panoramas still need to get buttons. Meanwhile, click and drag to look around, press Shift to zoom in, Command (Mac) or Control (PC) to zoom out.)

• Daily Arctic ice animation: Ross Swick of the US National Snow and Ice Data Center writes:
  

  The near real-time sea ice animation has made it’s way through our review process and been released on our virtual globes home page. It’s much improved as a result of everyone’s feedback. The released version includes both sea ice concentrations and sea ice extent. We’re also generating 3 files per day – 30, 60, and 90 day animations – so we can accommodate users with memory and bandwidth limitations. As before we’re using network links so the updates load automatically into GE at startup.

  The sea ice minimum is about six weeks away at this point and we can already see an above average amount of open water in the Arctic. As usual this is imagery, not data, so standard caveats apply.

  It’s definitely worth checking out the 90-day version if you can and then, when all the images have downloaded, dragging the time navigation handle quickly to and fro. The result is quite beautiful.

• Developing countries heart Google Earth: Jonathan Thompson types “Google Earth” into the Insights for Search tool, and discovers that the highest regional interest for the search term is… in Sudan, where US export law forbids the download of Google Earth application altogether (and Google is complying). Jonathan speculates that perhaps its relative popularity is due to its unavailability there, or due to the publicity surrounding the application’s default Darfur Layer.

  

  If you look at the other countries on the top-10 list, it’s remarkable that all are developing countries, seven are predominantly muslim, and six are run by authoritarian regimes (Myanmar being another case where US laws prohibit the downloading of Google Earth). What’s heartening is that this list is pretty much the same as the one where widespread use of Google Earth would do the most public good.

  Residents of countries where Google Earth is forbidden can of course take heart in the fact that the same imagery dataset is available via the web, without US-mandated restrictions, on Google Maps. And soon, very very soon, Google Earth will be in the browser for all — though it will be interesting to see if such a browser plugin is subject to the same export laws as the standalone version.

• FastSatfinder: This Google Earth browser plugin app lets you find the direction to a long list of geostationary satellites from any location on Earth (if it’s in your line of sight, duh). I can’t test it on my Mac yet, but no doubt it works in a similar fashion to the previously reviewed Dishpointer. FastSatfinder also comes as a $30 Windows standalone version, with extra features that I suspect are of use mainly to the dedicated hobbyist. (Via Boxier’s Weblog)
• Street View, post-disaster empathy tool: In the latest update to Street View, all of New Orleans is visible, and three years on from Katrina chunks of it are still looking miserable.

  
  View Larger Map

  Are you empathising yet? Want to help? That’s the idea behind Street View in New Orleans, according to Louisana Lieutenant Governor Mitch Landrieu writing on the Official Google Blog.

  If it works, why not take it to the extreme — I wonder what Street View of a Darfur refugee camp in Chad might do? (Via James Fee)

• Driving directions with the Google Earth API: Google Maps Mania flags a post on the Google Geo Developers Blog: “Simulating Driving Directions with the Earth API and Maps’ Directions API” (direct link to app) and makes a demo video.