I dropped a few hints last week about what we (the test team) has been doing recently. And one other hint was my lack of entries for a week in mid-November - I was on vacation. Vacation typically means long travel times, and what better way is there to fill the hours crammed onto an airplane than writing powertoys?
Yes, we have some powertoys in development that I hope to release soon. Here's the list, and the name of the creator:
- A template manager. Yes! Finally! And it's being developed by Jeff Cardon, so you know it is high quality. Apply templates to currently existing notebooks.
- Table summation (me). Add columns of numbers in tables! Earth shaking!
- A privatizer (removes names from notebooks) and cleaner addin. Helps fix random focus behavior in just published notebooks. (Gary Neitzke). Yes to privacy!
- A Journal to OneNote Exporter (or importer, depending on your point of view) (Lin Wang). Don't stay with the Windows journaling tool!
- Create Task Requests from Meeting notes (me) - More task integration with Outlook!
- An audio fine tuner (Jeff) - make OneNote into a Karaoke machine!
- An image rotator (Gary) - I'm turning backflips for this one!
They are all mostly ready to go and have been through most of the planned testing. Soon, some of the team is going to get together and have a "bug bash" on them. Since most of our testing is very regimented (click button 1, type "foo", click button2, verify the behavior) we employ "bashes" now and then just to hit the application with unplanned testing. As an example, you may open the dialog for the image rotator, then close it 100 times to ensure its actually being released. It's not very directed, tends to result in a scattershot of bugs from all over the place (hmm - what happens if you try to create a task request from ON's meeting note pages exactly during the time change from daylight savings to standard time?) and is generally fun. Testers get to stretch their creativity and ignore the planning and administration side of testing for a time to just dive in to answer "What if I do this?" over and over again.
There are some obvious downsides to this approach that include results of a broad pattern bugs of random priority, and at the end of the bash, you really do not have a good idea of where in the planned testing effort you are. And it's virtually impossible to know what the real state of the application you are testing actually is since there is no record of testing completed.
But it's useful, fun and generally a good time. I'll keep you posted on the results. And hopefully, these will all be done in time for anyone who may have some gift giving needs before the end of the month...
Questions, comments, concerns and criticisms always welcome,