FooCamp was a ton of fun – nothing beats spending a weekend in beautiful weather (outside at a conference? it can happen) talking with lots of smart, interesting, and funny people about a complete range of issues (humanitarian aid in Africa, to tracking your vehicle mileage in GoogleEarth).

The primary take-away I gained insight on was how much human is behind the machine/network for a great solution. There is a lot of research and effort spent designing better algorithms, faster processors, and automation. While we may be able to achieve better results and understanding through mass computation and filtering, it’s the human ability to pattern match and understand complex, arbitrary concepts that make for the best tool.

So the question remains, where is the boundary between the human-machine operation? How much effort do you let the machine perform, and how best can you allow human-bias to influence the results at a sufficiently high level that the entire thing scales. Are users tagging sites enough, or should their be a concierge type service that users can submit questions and get semi-realtime responses or aggregated similar historical responses (similar to an *answers site)?

Searches that are filtered by our social networks provides a very good solution. I’ve specified by preferences by the friends I’ve chosen. However, this is a very passive solution and doesn’t account for the fact that I may like Bob and his taste in music, but I wouldn’t trust his recommendation with food. It seems like there is some solution between tagging, IRC/IM, forums, and traditional search.

Speaking of search

Another very exciting thing was pimping the new OpenSearch-Geo. It’s interesting how few developers who are usually “in the know”, know that OpenSearch is a very much alive, and easy to implement, standard that can greatly enhance their service.

I think we’ll soon seen a number of popular sites get OpenSearch capabilities and also some better browser support.

Now that the weekend is over, I get to enjoy a relaxing vacation for a week in Hawaii before heading off to the UK for more conferencing.

Mickaël ‘Korbinus’ Graf has released an improved geo:truc. If you haven’t tried it yet, geo:truc is a great and simple service for generating the GeoData markup in currently 8 different formats: Machine tags, geo Microformat, GeoJSON, GeoRSS (GML), GeoRSS (simple), HTML, KML, and GeoRSS-W3C.

To use it you can just click on the map, or enter a location in the form field. Then click on the link for the format you want to see. Read more about it on Mickaël’s blog. One of the neatest improvements, and very useful for integrating into another application, is the webservice:

• http://geotruc.net/gt.php?q=Detroit,%20Michigan&format=georss
• http://geotruc.net/gt.php?q=Detroit,%20Michigan&format=w3cgeo
• http://geotruc.net/gt.php?q=Detroit,%20Michigan&format=machine
• http://geotruc.net/gt.php?q=Detroit,%20Michigan&format=geo
• http://geotruc.net/gt.php?q=Detroit,%20Michigan&format=geojson
• http://geotruc.net/gt.php?q=Detroit,%20Michigan&format=gml
• http://geotruc.net/gt.php?q=Detroit,%20Michigan&format=html
• http://geotruc.net/gt.php?q=Detroit,%20Michigan&format=kml

What is really great about geo:truc is that it provides a very simple and easy to use tool for users to get somewhat complex information. I believe his primary purpose was to provide scientists a mechanism for geocoding their specimen identification experiments. These are non-GIS users who want to store geographic information. Now, they can do it and include it on their websites/documents and share with the world.

But why geo:truc? Mickaël’s reasoning behind the name explains it very well:

“Truc” is a french word meaning an undefined thing, so I joked by calling it “geo:truc”.

Why is it that most major universities maps are still relegated to the old scanned paper copy? Check out the beauty that is University of Michigan’s Campus map. It’s amazing that a major university still uses GIF images for their campus maps.

Now, they know what mapping is, check out their cool transit services. There is even a 3D Atlas of Ann Arbor in Google Earth.

It’s not hard to make this into a more usable dynamic map. I used the MetaCarta’s Map Rectifier to take the campus map image, rectify it using several control points (intersections and circles work great for this) and created a slippy campus map.

You can play with the actual Campus Map here on MapSomething. The next steps would be searchable campus directory, click on buildings to get info on rooms, open times, phone numbers, a way to upload your schedule and have it plan out your route, etc.

Maybe I should send this on over to the School of Information or the Community Information Corps

I was gone for 5 days to the Ontario Curling Association’s Colts Provincial Playdowns, the top-tier competition after playing down against 130 other curling teams. We held our own, but the competition was very stiff.

It was a tough time to be away, a lot of amazing news came out. First and foremost is that Google adds support for GeoRSS. This is exciting news because it demonstrates the maturity and interest in the syndication of geographic content in blogs, CMS’s, sites, and news.

This will also add a little bit of more difficulty moving forward in GeoRSS. Now that a major company has added support, and assumedly a lot more developers will add support now as well, then the specification has to be much more cognizant of future changes, users, and upgrades. Before, the specification was really guided by the majority of developers using the standard itself. If some spec was changed, we all went out and updated our libraries. Now, however, we really need to denote versions, and how users can update their tools to accomodate both the new version and backwards compatibility.

On top of that exciting news, Google also open-sourced part of the GoogleMaps library. See the
gmaps-utility-library-dev FAQ. Currently this is limited to the GMarkerManager, but demonstrates their interest in opening the library up for interesting projects, ideas, and hacks.

The geowanking community has taken notice of the Twitter service. If you’re not aware, Twitter is a simple service to share what you’re doing. Twitter just lets you post a line of text, and you can do it via SMS, IM, their web interface, or any number of libraries that are popping up.

There are already two twitter-map mashups: TwitterMap, and GeoTwitter. GeoTwitter is a simple “red-dot” map, which parses a twitterer’s home location and plots that using Googlemaps XML.

TwitterMap is much more full featured. It actually polls the current public profile feed looking for “lat: lng:” specification of your current position, and then puts that on a full-frame map, with different colors depending on how long since you’ve updated. The problem is that Twitter has gotten very popular, and updates happen frequently. TwitterMap only polls the public list every so often, so there’s a 1 in 20 chance your post will actually be grabbed.

The developer, Patrick Kollitsch is hard at work on an updated version that should fix this issue and have more features to boot.

I definitely like the idea of showing a fading history of when the person last updated, or to see a trail of their travels. Also, seeing the user’s community of friends with lines linking them – especially if mixed into the history, so see as they come together and apart, would be very cool.

More to come on possible geocoding & picoformat ideas around Twitter, and like services (dodgeball, plazes, et al.)

Update (3/28/07): another TwitterMap has shown up that offers a very simple and effective interface. It also offers TwitterVision, which is a semi-realtime updating of twits as the show up and their location.