Link Details

Throw in jsp and servlet and struts (not necessarily 2) into the graph and you get a tie with asp.net. Certainly Java web frameworks land is segmented, so on the positive side there are plenty alternatives to choose from, but than again neither one turned to be a silver bullet.

Yardena, it would be really hard to accurately get a read on those values... I just tried "struts-2" to get "Struts" but not "Struts 2" and the line was hidden among the other Java technologies, not even near .NET

Also if you add "jsp" or "servlet" you are going to be picking up all other search hits like "jsp editor" "servlet deployment" and other "noise" that might not accurately represent the technologies being deployed.

That's the trick with Google Trends, it takes a while to figure out what your results actually mean.

Thumbs down. I am really not clear what these graphs mean. Just indulge me for a moment and type in india and china as two words into google trends. In real life China is bigger than India on almost every major metric reflecting size - yet google trends would have you believe it the other way.

(Humour) : Theres another way to settle this - just pair up each sets of frameworks on googlefight.com

Dhananjay,
So... you voted the story down because you weren't sure about how Google Trends worked, but there wasn't necessarily anything wrong with the article?

Then I hope this helps:
http://www.google.com/intl/en/trends/about.html#1

Grails?

rkalla,

Seems to me my euphemism backfired. Here's what I meant - you were trying to correlate google trends output with size or popularity in usage. That correlation is not nearly as well established as would be required for a writeup of this nature. Perhaps job postings might have been a better indicator. People may search more for something because it has more hype or its documentation sucks. Similarly more documentation content may lead to a higher news reference.

I gave you an example of India and China just to show a different, but otherwise easily understood situation where it was difficult to correlate the output of google trends with real life metrics. I am sure there can be many more such examples. The most you can perhaps infer from google trends is what is being searched for more or what is referred to more frequently on web indexed text - not what is larger in terms of usage. I did not vote it down because I did not realise how google trends work - I voted it down because the statistical footing on which the article was based just seemed too fragile.

Here's an example of using a slightly different statistical basis - enter the text on the following line on the search box of this url : http://www.indeed.com/jobtrends :
"struts 2", "java server faces", "adobe flex"

Even though this does not portray the actual popularity - it does search based on job postings which makes the correlation of whatever hypothesis you might draw just a little bit stronger.

I'm also voting down. This has been discussed countless times, google trends stats must be taken with a grain of salt. Your analysis includes JSF, which a little googling will show that includes Joint Strike Fighter more than Java Server Faces. Same thing applies to struts.

"So... you voted the story down because you weren't sure about how Google Trends worked, but there wasn't necessarily anything wrong with the article?"

Apparently, you don't understand how it works either. Dhananjay is right here, job trends are harder to misinterpret.

Dhananjay & icoloma,
Thanks for the clarification on Google Trends; if I do another analysis of this in the future I will certainly include the job postings (and maybe some other trends?) to get a more accurate read.