Last week Alceste mentioned, in a comment about using personas for testing, that the sample set we have on www.microsoft.com/evidence may be self selected and skewed toward the educational market. That's a good point. If we had thousands of case studies, I would not be as worried. Since we are still a bit short of 100 I wanted to point out two new studies and show how they will affect our testing efforts.
First, the robotics company Autonomous Solutions has really embraced ON. Check out the report at
http://www.microsoft.com/casestudies/casestudy.aspx?casestudyid=4000001729. I'll not quote any of it here, since it is rather lengthy and reads like a sales pitch. Honest, we did not make this up! One way this will alter our testing configuration plan is their use of VPN. In the "Solution" section, Mr. Torrie points out his use of VPN as a method to ensure everyone can get to the notebooks they use wherever they are. We (the test team) will take a look at VPN usage among OneNote users and reinforce our usage of that as a configuration for our own shared notebooks.
Looking down near the end of the article shows some nice words about OneNote which hide some good test cases. Mr. Torrie mentions he frequently drags and drops tabs (sections, to us) around to lay out his notebooks in a more logical manner. We already have test cases for dragging and dropping sections from within OneNote, so it's nice to get validation that our cases mirror real world usage. We can expand on the simple "drag/drop" once case to make it more iterative - automate moving sections and move them thousands of times.
There is also a design component to his comments which we will look at from a PM point of view. The question I had when reading this was "Why does he feel like he has to move his sections around?" Obviously, he may want to make his data appear alphabetically, old projects may get archived and new ones take their place, he may rename sections as new data is gathered or any other such legitimate reason may be why. He also may feel constrained by the UI - there is not enough room in a notebook to show all the sections at once on his computer and he has to make section groups to ensure there is enough horizontal space to show all his sections. These statements drive the tester in me up a wall - I want more details to understand exactly why this happens in every case.
Overall, though, this case study focused on OneNote and gives ideas for new and reassurances for existing cases.
The Coast Guard study at http://www.microsoft.com/casestudies/casestudy.aspx?casestudyid=4000001646 is similar to our educational studies I've already mentioned. In the middle of the article is a comment that the instructors "transferred documents, charts, diagrams, video clips, and photographs into the Office OneNote 2007 notebook." We have a repository of real world notebooks we use for automation, and we can now double check to see that we have some samples with all of these different data types in a single notebook (I checked, and we do). This is similar to the ODE notebook I created using the MIT Open Courseware class on differential equations. The only thing I did not do was embed the video files - I played them externally on my Zune. Again, it's good to get reassurance that our data set is valid.
Go ahead and check them out - the robotics article is incredibly detailed and oriented to OneNote. And the company website it very interesting, especially if you like robots!
Questions, concerns, comments and criticism always welcome,