Editor's Notes

1. Select any vertex and use it as a vertex for the n-2 triangles that are produced when this vertex is connected to each of the other vertexes.

2. Calculate the coordinates of the centroid for each of the triangles produced using this formula:
XC = (X(1) + X(2) + X(3))/3
YC = (Y(1) + Y(2) + Y(3))/3
3. Calculate the area of each triangle produced using this formula:
[(X(2) - (X(1)) * (Y(3) - Y(1)] - [(X(3) - X(1)) * (Y(2) - Y(1))]/2
(areas may negative! ... for reasons I don't want to go into)

4. Calculate the area of the whole polygon .
5. Calculate the fraction of the total area that each triangle contributes.
6. Calculate the product of each ordinate from #2 and the result of #5.
7. Sum the results from #6.

For some reason, I thought a simple approximation was the averages of the x and y coords around the polygon... hmm...

I've seen several homemade and open source GIS demos that rely on SVG output for vector display in a web browser. Firefox and Opera support SVG natively, but until now, Internet Explorer users have relied on a plugin from Adobe.

At this point it's old news, but Adobe is discontinuing support for their plugin which isn't really supported on Vista, either... so there's no SVG support in IE 7. From blogs, it looks like the product team is trying to figure out if they're going to include it in IE 8.

I signed up for the deCarta developer program today to try out their AJAX and Web services APIs. Looking around, it appears that their technology may be behind some of the bigger names in Web-based Location Based Services. Pretty cool.

The API looks nice and straightforward, though it may be a little light. It's pretty new from the look of it. I like the map styling capability. I could imagine that some RPG/movie/console gaming folks would go nuts with that. Create a look-and-feel that fit Harry Potter, Eragon, etc.

I would say that they may have given too much control for the developer over the infobox. It's not a standard that automatically resizes like Google's box. The customization ability is probably great, but leaves some potentially ugly design and content sizing up to the developer.

So far, I think it's a nice stab in the right direction.

Today, Directions published an article of mine on GIS and Accessibility for blind and visually impaired users. It's an off-mainstream topic, but given the ageing workforce issues that the computer industry at large is encountering, it's relevant.

Directions reports this week that Google Maps and Google Mobile Maps now have traffic data layers for some major cities. There is much speculation over how they are doing it... I imagine they are getting some webservice traffic data from these major metro areas, but I could conceive of them using news/accident feeds or other tools to approximate the conditions, as well.

What will they do next?

Try this link for San Francisco traffic:

http://maps.google.com/?z=10&ll=37.779399,-122.419281&layer=t

Here's their blog post on the topic:

http://googleblog.blogspot.com/2007/02/stuck-in-traffic.html