... No, it wasn't OK. I say this not because it wasn't OK for your project (it was probably OK for your project--and did you follow the double negative...
Hi Lisa, can't you make a story that says: 'Make automated acceptance testing available to testers'? (maybe a simplistic answer??? As a developer I would be...
775
STEURS Stefan
stefan.steurs@...
Jan 2, 2002 10:11 am
... I was off for a while, but I'm back. ... A quality factor or attribute is not the same as a story and it is not the same as a test. Robustness would apply...
776
STEURS Stefan
stefan.steurs@...
Jan 2, 2002 10:42 am
... Two questions: 1. Is it worthwhile to reduce the test set by 20% or not? 2. Can you really make a 120 truelly random tests? Randomness is very hard to...
777
STEURS Stefan
stefan.steurs@...
Jan 2, 2002 11:06 am
... To me there are two basic objectives in testing: 1. make sure that the software does the right thing 2. make sure that the software does it right If it...
778
STEURS Stefan
stefan.steurs@...
Jan 2, 2002 11:52 am
... I don't get this. If the developers are running the tests then why should anyone else re-run them? If I look at our Unit level testing (which is a poor...
779
STEURS Stefan
stefan.steurs@...
Jan 2, 2002 12:59 pm
... One problem I have with this is that risks are presented as being external only. There are internal risks that have to be dealt with as well. In "Make X...
780
STEURS Stefan
stefan.steurs@...
Jan 2, 2002 2:26 pm
... I think there is a fundamental mistake here. Creating a thousand truely random tests is probably even more difficult than making a thousand ...
Michael Bolton wrote: hard-coding is a pernicious practice, whose practitioners should be ... coding in your ... Bad. I don't know about hard coding in the...
... testing ... That's a good idea, unfortunately in this case, we are on the last iteration, and aren't picking new stories. But I think in the future I...
... can ... It will hurt them later when they have to regression test, but they don't think about that now. Since we are doing the un-XP thing of having...
... should ... In our case, working with an external customer, the developers' integration machine does not look like production. We also have to test on a...
... care ... don't ... can ... We had a similar problem on a distributed project (some XP practices). We added an entry on the story that the tests ran in the...
786
Michael Bolton
mb@...
Jan 2, 2002 7:27 pm
... Well, it is a pretty tough task. People do see the hurt of bad practices a lot more quickly and definitively than they see the joy of good ones. However,...
Happy New Year everyone. A test tester (strong mutation testing tool) for Python code and PyUnit tests has been released, called Pester. To quote the Jester...
788
agile-testing@yahoogr...
Jan 3, 2002 5:43 pm
Hello, This email message is a notification to let you know that a file has been uploaded to the Files area of the agile-testing group. File :...
789
agile-testing@yahoogr...
Jan 3, 2002 5:45 pm
Hello, This email message is a notification to let you know that a file has been uploaded to the Files area of the agile-testing group. File :...
... You've already received some good responses to this post in my absence. I wanted to point out that in my particular example the volume of code under test...
... Are you saying these manual steps were the acceptance test, or were they codified in another test? Maybe I'm missing something, but it seems that the ...
... I agree for all those reasons and also because it implies that lowering the quality level may yield a reduction in cost. There's also the sociological...
... Can you clarify what kind of values are being hard-coded? For example: assertEquals(10, doubleMe(5)); or assertEquals("c:\myspace\config",...
794
Petersen, Erik
PETERSEE@...
Jan 8, 2002 12:29 am
I am 100% with Ron on this one. The "1000 test have passed scenario" can occur when you move onto a development project that has not yet been released. Tests...
... See my previous post. I'm guilty as charged when it comes to testing assertions that aren't system dependent. Production code, however, is almost never...
796
STEURS Stefan
stefan.steurs@...
Jan 8, 2002 6:34 am
I think hard coding values is bad in any code. Test code should be of the same quality as production code. Test code may be equaly difficult to maintain as...
797
STEURS Stefan
stefan.steurs@...
Jan 8, 2002 6:42 am
An a posteriori analysis of test objectives can only lead to the conclusion that you would have been better of specifying your test objectives upfront and...
I'd like to offer a little respectful counterpoint to the following post. Around Tuesday, January 08, 2002, 1:42:48 AM, STEURS Stefan ... The thing is, after...
... Here's an example assertion in a programmer test: assertEquals(10, doubleMe(5)); What, if anything, is "bad" about this? Indeed, given that tests are a...
800
STEURS Stefan
stefan.steurs@...
Jan 8, 2002 3:05 pm
Why not say const MIDDLE_OF_RANGE=5; const EXP_DOUBLE_MIDDLE_OF_RANGE=10; If I want to make permutations of the test, I decide to replace the constant values...
At 04:42 AM 1/2/02, STEURS Stefan wrote: [...] ... That's the point of the Hamlet and Taylor paper. IF you have a test design technique that lets you zero in...
... I'm OK with that, so long as you keep talking to Bob. What does he think about how you're spending his money? Beyond that, I've learned to be suspicious of...
