Sometimes testing is about finding patterns. You have the functionality there, you have the requirements, but sometimes something is not in place, like an exception. You noticed that it does happen from time to time, it is quite important as impact on product, but you don’t have all the details to know whats going on and raise an issue or a bug. It is always I guess an option to give up and ignore it, and may seem reasonable to do in some context of risk/occurrence. However as testers we should not give up in situations like this simply because it should be our duty. I don’t say we should follow like crazy everything with low priority and loose time, but in critical situation we should try and enjoy it.

So I was thinking to select something quite bizarre, relative, maybe a little stupid and crazy. And I made my mind to find patterns in blogs, blogs that were with some relevant content to software testing and are relatively popular.

I started thinking about words count in latest blog posts. I found this tool at http://www.wordcounttool.com/ which counts words. I selected the last 7 blog posts from James Bach’s blog. ( it started to become boring to do for more posts) and this are the results(although I am not sure I pasted it right):

2323 words

423 words

572 words

554 words

627 words

732 words

320 words

The first items are also the counts for more recent posts btw. But it didn’t seem to me to be a particular pattern. I suppose it will work with selenium to count all the words in all the posts. Because the relevant post is between a title with different color and font and it ends just before the following text “This entry was posted on…” . But in a real testing environment you would probably have access to the database with all the posts and extract in a more reliable way the information from there. I guess I should test it on my own blog because I have access to my wordpress database , but I want to check also other patterns. I remain however in my mind that the blogs posts on James Bach’s blog may have an average of 500 words per post.

I want to measure and check what are the most used words in a blog post(s). First I will use another tool Google to help me find a helpful tool (and free) for that. I spend only 2-3 min and found this wonderful tool at http://www.wordcounter.com. I selected the latest post from Michal Bolton’s blog (to the date I wrote this post) http://www.developsense.com/blog/2010/06/doing-development-work-vs-doing-quality-assurance/. The tool is great and has some options like including small words or not. I tried first with include option on and here are the results:

I was expecting “a” to be on top to be honest. However I want to check the statistic for the other words maybe it will be more relevant. So here are the results:

Why is the word “extent” doing in top and why is before word “tester”? Is it random or Michael Bolton likes this word a lot?

<<At this moment I was getting hungry and put some pizza in micro. Defocusing is a normal need and should be followed for better performance>>

I looked for the word “extent” in the blog post to check the context of using it and it appears to be in a special situation. But I wanted to check for sure other posts. And the previous post broke my dreams in finding a pattern for the most used word. “Extent” was not in the top 5 and neither that the previous top words.

What next?

<<Defocusing again>>

I was thinking of checking in Cem Kaner’s blog how many sentences tend to be in a post in average by searching for “.” but I gave up quickly.

But I started to realize that I am doing wrong something. Like some bad ideas in testing for example unnecessary automation. With evolution of man people tent to loose common sense and when they do they re-learn it and get more trust in it. The same in society when growing up, a lot of good ideas, talent, is fading away because of constrains.

But going back to the “test platform”,it was interesting to try, that’s why I tried it because I satisfied in this way a curiosity.