tag:blogger.com,1999:blog-36118108.post3169857113109376236..comments2023-12-06T00:17:28.519-08:00Comments on Creative Chaos: What's an SDET - IIMatthewhttp://www.blogger.com/profile/05956714498778698672noreply@blogger.comBlogger5125tag:blogger.com,1999:blog-36118108.post-85747479018500681062009-03-06T19:43:00.000-08:002009-03-06T19:43:00.000-08:00Cool post, Matt! Thought-provoking like most of yo...Cool post, Matt! Thought-provoking like most of your posts are. I started as a developer although without the cred you have. Interesting to see people go both ways - my most recent fellow tester is now a Java developer on our team. But his testing ability and habit of talking constantly to customers makes him more valuable IMO.<BR/><BR/>In addition to what you said, I think curiosity, willingness to get out of comfort zone, desire for good craftsmanship, are all good qualities too.<BR/><BR/>Developer envy, well, it's true I wish I could code Java, but I have the same learning block with it as I had with Assembler. But, can understand it enough. Someday I hope to do some pairing on writing production code, just for information.Lisahttps://www.blogger.com/profile/10230090963033880060noreply@blogger.comtag:blogger.com,1999:blog-36118108.post-4550237324515928102009-03-06T10:52:00.000-08:002009-03-06T10:52:00.000-08:00Not offensive Matt - just speculative (I think I m...Not offensive Matt - just speculative (I think I may have read too much of your Bach quote into your opinion). <BR/><BR/>Our best testers don't have developer envy at all. In fact, in some cases, we have developers move from dev to test "because the challenges in test are more exciting". There are probably more going the other way, but it's usually a case of "the grass is always greener" rather than envy.<BR/><BR/>You seem to imply (whether intentional or not) that developer/testers are unable to perform empirical process and learning - or maybe you're just saying that it's important and you're not sure if it's included - it's unclear to me.<BR/><BR/>Then again, maybe it is just the Bach quote that sets me off. <BR/><BR/>A minor nit is that we have MS R&D sites in about 15 different states and at least as many countries, so it's not just Redmond, China and India. Off the top of my head, I'd say as much as 20% of MS R&D happens outside of those locations.<BR/><BR/>Finally, you make a big deal out of the hundreds of open positions. I don't know the number, but let's say it's 400. One thing to note is that there are about 10,000 testers at MS, so 300 is like having 4 open positions on a test team of 100 - not that significant, and not significant enough to imply that the positions are exceptionally difficult to fieldAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-36118108.post-35580706008029868052009-03-06T10:29:00.000-08:002009-03-06T10:29:00.000-08:00While I was disappointed by comments from hwtsam, ...While I was disappointed by comments from hwtsam, I do agree that the metaphor was a bit of a strech, and I have changed it to be the sprinter/distance runner as requested. Otherwise, I'm not quite sure where I speculated or what was so offensive.Matthewhttps://www.blogger.com/profile/05956714498778698672noreply@blogger.comtag:blogger.com,1999:blog-36118108.post-82856000716017220112009-03-06T09:39:00.000-08:002009-03-06T09:39:00.000-08:00Your initial statements are equally valid when tur...Your initial statements are equally valid when turned around - i.e. I can verify that the web page displays in numerous browsers and resolutions and that it's highly usable, but never verify that it actually calculates what a triangle is.<BR/><BR/>You are absolutely correct in saying that it's difficult to have testers who can both think "at the code level" and who can evaluate the system for stakeholders. The comparison of basketball to golf, however, is quite a bit of a stretch. It's more of a comparison of sprinters and distant runners. Not everyone can do both, but we've found about 7 or 8 thousand testers that <I>can </I> think at the code level and who are world class testers (note that we have 9500 testers at MS - details on my discrepancy will require beer).<BR/><BR/>For what we do, and for our requirements, we need people who can do both. What I've found, is that it's easier to find people who are great testers from the pool of people who already know how computers work. This is based on hundreds of interviews, and interactions with thousands of testers. There are absolutely exceptions (I'm one), and we have programs to find great testers from outside the CS field as well.<BR/><BR/>I know I'm just fanning a fire here that I have no control over - but I do find the speculation and assumptions on your part a bit out of character (perhaps you really *need* that vacation).<BR/><BR/>Starting from "Hiring developers to be testers..." you're pretty much making stuff up as far as I can tell - and I'm saying this knowing that we have worked together and respect each other. I do know that there's this view that MS is screwing up their testing because they're hiring CS folks instead of "real" testers and that's the cause of our quality problems, but believe me - the source of the quality problems is elsewhere (and outside the scope of this comment).<BR/><BR/>I suppose the fact that some people don't get it is something I should worry about. I try not to, but I do worry about the public face of testing at MS. I also realize that some folks will never get it, and I worry less about them.<BR/><BR/>From you Matt, however, I expected better insight.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-36118108.post-29537791642437871842009-03-06T09:35:00.000-08:002009-03-06T09:35:00.000-08:00I really enjoyed the post. I have been guilty of t...I really enjoyed the post. I have been guilty of trying to go to far to the programming side at times. But I love exploratory testing so it usually pulls me back to the middle. I agree the good mix on a test team is a great situation. Let the automation handle the technical details and get creative with real testing. <BR/><BR/>JeremyAnonymousnoreply@blogger.com