tag:blogger.com,1999:blog-8951904624959546499.post925664312570640211..comments2009-03-23T18:27:49.685-04:00Eric Jacobsonhttp://www.blogger.com/profile/08216361684596485033noreply@blogger.comBlogger6125tag:blogger.com,1999:blog-8951904624959546499.post-10111414641307415132009-03-23T14:19:00.000-04:002009-03-23T14:19:00.000-04:00Hear, hear. I felt like I'd grown as a tester the day I decided to just log the bugs instead of spending two hours justifying myself for creating the bug. Nice blog :)Marlenahttp://marlenacompton.com/noreply@blogger.comtag:blogger.com,1999:blog-8951904624959546499.post-37254031348649368072009-03-19T13:00:00.000-04:002009-03-19T13:00:00.000-04:00Anonymous,<BR/><BR/>Good points! However, the devs shouldn't be asking testers if devs should fix with j, k, or l. <BR/><BR/>Some testers will answer that question and some devs will continue coming to testers for solutions. That could be bad.<BR/><BR/>Most solutions are business questions. And even if they're not, testers are not designers and should not be making such important decisions. Right?Eric Jacobsonhttp://www.blogger.com/profile/08216361684596485033noreply@blogger.comtag:blogger.com,1999:blog-8951904624959546499.post-3690723498972903492009-03-19T12:49:00.000-04:002009-03-19T12:49:00.000-04:00Philk,<BR/><BR/>Yes, "system should not throw unhandled error" seems like better expected results than "?". It would raise fewer eye brows.<BR/><BR/>But if we just use the inverse of the actual results, how much value are we adding? Hmmmm.Eric Jacobsonhttp://www.blogger.com/profile/08216361684596485033noreply@blogger.comtag:blogger.com,1999:blog-8951904624959546499.post-24133259473131921092009-03-18T19:39:00.000-04:002009-03-18T19:39:00.000-04:00there's likely to be less communication overhead though if the expected results are there.<BR/><BR/>expected results: ?<BR/>actual results: x<BR/><BR/><I>dev: oh, actual results are x (blah blah blah). i could do j, k or possibly l. i better clarify.</I><BR/>(and dev bugs you right in the middle of your testing)<BR/><BR/>OR<BR/><BR/><I>dev: i'll just apply n. it'll probably be good enough.</I><BR/>(and once aut is back in your hands)<BR/><I>tester: curses! this approach isn't enough (or he should've done m instead)!</I><BR/><BR/>i suppose another factor is how smart the dev you're working with is ;)Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-8951904624959546499.post-88167699503819479632009-03-18T17:45:00.000-04:002009-03-18T17:45:00.000-04:00Damn straight. Run tests find bugs.Mark Crowtherhttp://www.blogger.com/profile/11540303029361604504noreply@blogger.comtag:blogger.com,1999:blog-8951904624959546499.post-11114768782376263882009-03-18T15:00:00.000-04:002009-03-18T15:00:00.000-04:00Phew, I thought you were having a go at my blog when I saw the title..<BR/><BR/>isn't one of the expected results "system should not throw an unhandled exception/crash" ?Philkhttp://www.blogger.com/profile/00281118161548464012noreply@blogger.com