Greetings Geekdom. My apologies for the long absence and my apology to the gentlemen who I was conversing with regarding is Automated Testing "testing"
First to you: I got your response email. I will post it up here in the next day or two with my thoughts on your questions.
Second to the geeksphere:
My absence can only be blamed on two things: hectic holidays and being extremely busy at work. All I've wanted to do since Thanksgiving was come home and veg out. And if I wasn't veg'ing out, I was out X-mas shopping. :)
So anyway...things have finally settled down a bit.
On the professional front, my new gig extended a FTE position to me and I started with them as an FTE on the 29th of December. And man can we say they keep me busy...OY...I've created more test cases in the three months I've been there than I did in an entire year at my previous gig.
Now granted a lot of these cases are not reusable action created scripts. For the most part, they all have test case specific code in them but they work and I'm saving a manual tester 8 manhours worth of repetive work every week. Given the cost of my time, we'll recoup the time spent in three months of weekly exection. How's that for justifying one's existence. BOO-YAH!
Additionally I'm now coordinating automation development by several offshore resources we have through an offshore firm in Mumbai.
Now in gerneral, I am not a huge fan of outsourcing. I'd rather hire americans and have the work done here. But so far I've been really impressed with these guys. They can think on their feet and adapt to changes in priorities with the agility of the best onshore resources I've worked with. Not to say I still think we aught to hire american..but at least I'm getting more stuff done than I could by myself.
And finally I do have a QTP/.NET Web related question for the geekdom.
One of my applications under teset is a .NET Web app. (Not sure which framework but I suspect it is 2.0.) Now when this application is initialized it starts out as a regular website rendered in a browser.
Once the user logs in to the system, the app renders a popup for the actual app with no "Go" entry fields, no "processing" e or anything etc.
As a result of the supression of the browser's native navigation clues etc, the users of the system complained that they couldn't tell when the app was doing something. So the dev added a "Processing" function that appears when the app has to go off and do something. It disables 90% of all of the objects on the page and in some cases if you do an if exists it will fail to find the object as entered into the OR. I call this the "grey screen of ennui" because it is ANNOYING as all get out.
The principle problem with this feature is that the display of the feature does not put the browser into a "non ready" state. Therefore "Sync" is useless.
My question is, can someone suggest another way of dealing with it? Right now I have a function that utilizes a for next loop running If Exists over and over again. However, I'm finding that it is not 100% reliable. So I need a better way....
Any suggestions the geekdom might have aside from requesting an application enhancement to make the grey screen of ennui put the browser into a non-ready state would be most appreciated.
More to come on the "Is Automated testing testing" debate.