Here's the weekend story, with it's sad ending, too:
"How cool would it be if I could capture visitor IPs for a certain
url and plot them on a map, like Google maps?," I thought to
myself. "Very cool," myself responded. Grab the IP from,
say, the IIS logs or whatever and do a lookup. No problem,
right? Wrong. Innumerable ips to look up, lots of DNS,
stupid IIS permissions. "How about a javascript that calls a
simple page to record the host ip and the datetime?," I said to
myself. Slightly concerned that I was going crazy, I replied
tentatively, "Sure, that sounds good."
I grabbed the raw delegation ranges text file that ARIN
has mirrored for RIPE, LACNIC, and APNIC (RIPE has AFRINIC's mirrored)
and stuck them in a db (there're only about 70,000, a lot less than the
max amount of ipv4 ips), made an IP address to IP number conversion
utility, a quick lookup utility to get the IP block of a particular IP
number and, finally, a little whois'er that queries the block
registries (not domain registrars) and grabbed city and country
name. The last little bit was manually adding geocoords (lat,lon)
to each city in the city table, but that's easy. (I'm thinking
about how to automate that bit, too.) Phew! Technical
mumbo-jumbo aside, I get the visitor's city and plot on map.

The final result looks sweet. (Petah Tiqwa, Israel?) For example, this blog's hits come from all over, see here.
This morning, as I'm perusing some news I see that Google's releasing their analytics product for free (Free Web site therapy, Reuters, 11/14/2005; Google Press Release).
Yay! I like Google. Then it hits me: I've just been
housed! Granted, they've probably had the idea for a lot longer,
more resources than just one guy and a weekend, and much, much more
data, but check out the neat picture from their main page:

It's (of course) got a map to track visitors! (Their map
is Flash, though. Odd, I thought, when they have a map tool.
Seems like this analytics suite was bought/enhanced from some company
called 'Urchin5'? May have to look into it, may not care so
much, either). It uses a simple little javascript to ping their
servers. It has pie charts. Everyone knows that pie charts are
the end-all-be all of cool. Google:1, Me: 0.
Of course, I'm not competing with their product or even into
"analytics," I'm just experimenting. I learned a whole bunch,
too. For instance, NET 1.1 (which I used instead of Java, for
kicks, or 2.0, which looks pretty neat, too) is pretty straightforward,
once you get used to its quirks. I dislike how it reformats HTML
and how it doesn't compile in the background like Eclipse.
Parsing registry data into a db's easy, lookup reconciliation's not too
bad, and generating javascript for the Google maps is a snap.
I'll keep playing around with my app, trying to make it more
efficient and automated, but I'm going to stick on the Google
Analytics' javascript on a few urls. Since they're (Google,
collectively) smarter than I am (and have a patent on address geocoding), I'll be interested to see how well my data tracks.
[Edit: here's a story about how Google's liberation of the Urchin product effects analytics firms: WebSideStory Stock Falls on Google Plans, AP, 11/14/2005.
Urchin,
also a San Deigo company, was acquired by Google in March 2005 and was,
apparently, big news in the SEO community. Urchin's
service was a $199/mo.]
[Edit: replaced the link to the Google patent on USPTO with a link to Google's patent search on the patent. Neat!]
Perusing Ian's flickr, I saw UTM coords on some of the photos. I couldn't resist. Here's a page with a Google Maps plot of the points, with links to a Microsoft Virtual Earth plot, too. And the obligatory Google Earth KML. I won't go outside, but I damn sure will fiddle with bits.
Here's a screenshot of the route in Earth: ->
In all, it was a good experience learning about UTM coordinate systems, the Maps and VE APIs, and managing waypoints/coordinates. I'll probably end up making a database to store points and some more fully automated maps js, ve js, and kml generation. Tweakery is fun.
Update: 11/03/2005
Yahoo!'s come out with a map API, so now there're three different ajax/html + 1 kml generation thingies going on. More updates when I get to a pause-point in kneading this maplication.
NASA takes periodic sat photos of the earth and every so often they get them all together and release them. Blue Marble Next Generation is the latest collection of high resolution images from 2004, one for each month with cloud cover removed, and was released yesterday. It's pretty neat to see the seasons changing all around the earth. Even more dramatic when viewed using NASA's WorldWind app with the Blue Marble add on that dynamically grabs/adds data. (Google Earth doesn't have this dataset (yet?)). If you don't have either, use the web based version done up by the NASA scientist (Reto Stöckli) who headed up the project (or, this one, if you don't mind installing ActiveX thingies).

Satellite remote sensors provide the science community with high quality datasets to better understand and monitor Earth’s environment and climate system. When used in education, these datasets allow visual exploration of the planet at moderately high spatial and temporal resolutions. However, native data file formats and file sizes are not well suited for public distribution. To help circumvent these issues, the Blue Marble Next Generation (BMNG) is a new series of 12 monthly cloud-free, global-scale images. We created the BMNG using NASA Terra MODerate resolution Imaging Spectroradiometer (MODIS) science data collected in 2004. We temporally adjusted the data using a discrete Fourier technique. This correction method removed cloud disturbances, but snow reflectance remains a significant challenge. The BMNG visualizes seasonal changes of the land surface (spring greening, snowmelt, drought, etc.) as true-color images and Normalized Difference Vegetation Index (NDVI) maps in monthly steps at 500-m spatial resolution. The images are available at no cost from NASA.
- R. St¨o;ckli, E. Vermote, N. Saleous, R. Simmon and D. Herring (2005). The Blue Marble Next Generation - A true color earth dataset including seasonal dynamics from MODIS. Published by the NASA Earth Observatory. Corresponding author: rstockli@climate.gsfc.nasa.gov
[edit: 10/15/2005 1:28AM]
Looks like the WorldWindCentral.com guys put up a tileserver / kml generator for Google Earth. Check out the post here.