Back in November of last year, we released Urban Turtle 3.6 which feature major performance improvements. To achieve this, we used a variety of code profiling tools, including Eqatec Profiler. I really liked the straight-forward interface of the product making it simple to compare profiling reports. A few months later, the Eqatec team emailed me to notify Profiler users about a great offer. Put simply, they offered a free Profiler license if I simply requested a trial license for Eqatec Analytics, a tool used to monitor application usage. I didn’t think much of it at first, but have to admit I find it hard to ignore free stuff. Plus, I was really interested in an automated solution to give us actual usage feedback from Urban Turtle users.
I started the trial and was immediately impressed at how easy it was to plug it in the application code base. A few method calls here and there and I was already looking at a web dashboard showing me what features were being used (on my development machine, mind you). I was particularly interested in the feature timings data. With a few more calls, I was quickly tracking how long some features were taking. I was imagining how useful this would be if it were data from actual users. The advantages to using such a tool are just obvious. You quickly know what versions of your application are being used (you all update to the latest version as soon as it is published, don’t you?), and better yet, what features are actually the most popular. This would be so useful when planning out the next releases. It is crazy to think we haven’t had that in the product since the very beginning.
I really enjoyed my trial and we are looking at how to seamlessly embed this in a future version of the application. All usage data would be anonymous, of course, and we are trying to come up with an opt-in/opt-out scenario. We do not want to force anyone to send us this data, and we understand that a server running a TFS instance may not have access to the internet. However, I am now convinced that having such data even from a minority of users would be of tremendous value to us. And I believe that at least a few of us will agree with me and will be more than willing to help us deliver a better turtle, release after release.
