Kirk Evans Blog

.NET From a Markup Perspective

Google Trends

While I am a loyalist (please, please implement searches over newsgroups!), I am secretly fascinated by some of the services published by Google. Yeah, some of the Google services are pretty much rip-offs of other services, but Google Trends is just flat-out freaking cool.

The results are interesting from an evangelist point of view.

RSS – what you are using to read this post. Atlanta shows up in the top 10? Huh… Here I have completely avoided talking about RSS over the past year.

TechEd – a yearly Microsoft conference that occurs in multiple locations worldwide. The top search location is Walldorf, Germany, which completely eclipses all other regions.

Code Camp – attributed to Microsoft Developer Evangelist Thom Robbins (if the origin of Code Camps is different than Thom, PLEASE leave a comment with correction and origination link). The interesting thing is that Thom Robbins hails from the New England area, but the top Code Camp searches are in San Diego.

Microsoft – a small software company located in Redmond, Washington. Apparently a lot of corporate ego-surfing occurs in the Seattle area, since that is the highest origin of searches for “Microsoft”.

C# – a relatively popular programming language. Top search city? Bangalore. A U.S. city occurs in the top 10, but doesn’t appear until 7th (Seattle…hmm?)

interop – getting stuff to interoperate, usually spoken in the vein of web services. Again, Bangalore shows up first here, but there are more U.S. cities represented.

Don Box – an architect at Microsoft. Again, the trend shows more people in Seattle searching for Don than anywhere else. More interestingly, more searches for Don in the U.S. than in other countries.

I think the personality data is pretty interesting. More interesting is the searches that had little or no trend data available. Jim Allchin only reports data from Seattle. Personalities like Brian Valentine, Scott Guthrie, or Anders Hejlsberg didn’t show up at all. Not sure what this indicates, other than my expectations don’t match the actual data.

Something else that was interesting: comparing search results and volumes. If you do a search comparison for java and C#, the results are decidedly skewed towards Java.  I can’t help but wonder how the results would compare if you combined the search terms for .NET languages (C#, VB.NET, C++, J#, IronPython, etc) and compared them all against the single term “Java”.  It’s kind of like searching for “English” versus a dialect.  I also wonder how much is “hidden” inside other search terms. For instance, compare ASP.NET and JSP, and you get a much different story, showing ASP.NET way over JSP in terms of volume.