measuring testers

a-foton » measuring testers

a-foton » measuring testers — Tue, 22 Jul 2008 19:26:27 GMT

PingBack from http://blog.a-foton.ru/2008/07/measuring-testers/

re: measuring testers

grantholliday — Wed, 23 Jul 2008 06:39:24 GMT

Nice quote. I'm enjoying your blog so far - keep at it.

re: measuring testers

nkamkolkar — Wed, 23 Jul 2008 13:41:15 GMT

great blog jw and what a great way to think about measuring the value of testing and also testers.

I can't help but wonder tho for the thousands of testers out there doing bug bashes (and being measured in a way you are suggesting not to be), what can we change in our daily activities or thinking that can bring about this fundamental (but important) change.

Bet there is no "easy button" for this, but where do you start?

re: measuring testers

James Whittaker — Wed, 23 Jul 2008 16:24:45 GMT

Neelesh,

Leave it to YOU to be the one to point out that the two side by side posts are at odds. PEST on the virtue of finding bugs and this one the the virtue to not counting the finds! Clearly I owe you a beer for that one.

It's a connundrum, I agree, that the part we spend so much time trying to be good at is the exact part no one wants to be measured on! I think the real advice here is to get good at finding bugs BUT DON'T stop there. You need to take it to the next level and ensure that the apparatus for creating those bugs gets reprogrammed. In other words, finding bugs should lead to preventing bugs or it isn't very valuable.

But still...it is fun as hell...

re: measuring testers

ryanboucher — Thu, 24 Jul 2008 15:58:20 GMT

An interesting point but I feel that you are ignoring large slices of what a tester does.

If the tester reviews a requirements or ui spec, suggest changes and makes the specification more complete. This work making the analyst better, not the coder. Naturally this will implicitly flow through to a better initial version of software built by the coders. It’ll also flow through to the testing phases.

I also think you’re missing one half of testing which is “did we build the correct software” or, validation. We can have the ultimate code monkey cutting perfect code from now until eternity but if it’s not what the end user wants then it is useless. Naturally, if you are building COT software you probably won’t have a user until the marketing department finds one.

I agree that statistics, while interesting, shouldn’t be used to rate testers as the impact of an individual on a team is not something that can be easily or consistently defined. I disagree that the tester exists to be the support crew of the coder.

re: measuring testers

James Whittaker — Thu, 24 Jul 2008 16:31:16 GMT

Thanks for your comment Ryan. Nothing to argue about in it. But Tony's quote doesn't say 'code' it says those who design and produce code. I should have included the others too as opposed to my generalization. Good catch.

VSTS Links - 07/29/2008

Team System News — Tue, 29 Jul 2008 14:00:42 GMT

Grant Holliday on What product key do I use for TFS Proxy? and How do You test code that uses the TFS...

re: measuring testers

dking — Tue, 16 Sep 2008 20:22:35 GMT

It appears that the link to Jim Larus's page is incorrect. It is giving a 404 error at this time.

A szoftver minőségbiztosítási eszközök valós lehetőségei és korlátai

Termékinformációk fejlesztőknek — Wed, 04 Feb 2009 08:11:36 GMT

[Nacsa Sándor, 2009. január 13. – február 3.]  A minőségbiztosítás kérdésköre szinte alig ismert