i guess it was bound to happen
well, the build is being tested and we had four "bugs" get reported by
the clients today. i think two of them were essentially changes to the
functional specifications so i guess that doesn't really count. but the
fact that they found an actually "showstopper" as the guy who found the
major bug put it kind of floored me. but it was bound to happen
sometime. full coverage is not possible. also the bughunter was the
main developer on the project for three years, i have only been on it
for two months so i am not too crushed. that feature was actually part
of build 1.087 which never went into production, so he only found it
when they got build 1.088 for testing.
because of that, and elfriede dustin, i am going to do some research into planguage as the concept sounds very interesting. it was written up by tom gilb
in order to define a clear language to be used in specs. i would love
to be able to remove ambiguity from the specs, and as i am writing
specs as well as testing i can actually do that quite easily. i have
just opened the planguage glossary. 602 pages!!! that will have to be
logged into my GTD system tomorrow.
back to effective software testing.
learn from it
10:10 PM, March 30, 2006
.. Posted by philk10
As Jim says in his latest blog, don't beat yourself up and expect to be Joe Stud Tester and catch everything
Just learn from the ones you didn't catch as to why you didn't catch them - sounds like you're doing that
The next release you'll be better prepared
and the one after that
Joe Stud Tester
08:19 AM, March 31, 2006
.. Posted by metalbaby
But my contract says i must become Joe Stud Tester or else...
Next time I am not going to let the senior developer rush me when I want to do regression testing. I said at the weekly meeting that I wanted two days to do regression testing, AND got approval from the project manager on this. I finished testing on new features at 2pm on Tuesday. The developer told me that as he was taking Wednesday off work so it had to be done on Tuesday. Next time my foot will be coming down and it will be done on Thursday damn it!
but cheers Phil.This is a process and I am getting better all the time so it is ok.
chris
Edited by metalbaby on March 30, 2006 at 03:01 PM
Putting a foot down
02:39 AM, April 6, 2006
.. Posted by jimhazen
Chris,
Yes, you are right to put a foot down (on the developers neck if needed) on this issue. The product is not done until it is done and everyone (including the developer) needs to be present and available while the product is being completed. This means taking the day off before release is extremely 'Verbotten'. They should only, ONLY, be allowed to do this if they have someone to cover their stuff (that person had better know it intimately).
In as far as other things in your post... Yeah, those are what I call 'Left Field' errors (meaning you didn't see them coming), and you will need to learn from them. As the saying goes, $#!^ happens.
You're on the right course getting your PM to buy into doing things this way. I'm sure they don't want to get an ear full again from the client or the people above them.
Jim
taking the day off
11:03 AM, April 6, 2006
.. Posted by metalbaby
hey jim,
it is kind of okay for the devs to take the day off them because they client usually takes a few days before they start doing UAT. my issue was more that i couldn't do regression testing because he wanted to take the day off.
but thanks for the support. i don't want to fight with the devs, i need to make sure they understand we are all on the same team. but the problem dev is french belgian. nuff said? maybe.
chris
{ Last Page } { Page 10 of 12 } { Next Page }
|
About Me
Links
Categories
Articles Books
Recent Entries
Preparation for a new job Testing Tools phil started me on a screed... thanks mate. my delicious account articles and effective software testing
Friends
neillmccarthy philk10 jimhazen thepthial PeteNairn EklecticTester
|