6 ms·
Guess they've never heard of Fred Brooks...?
by jey 16y ago
Guess they've never heard of Fred Brooks...?
- nailer 16y agoWho's that? Google turns up nothing.
- ovi256 16y agoNobody did. At my ex-employer, I did an informal survey, with a simple reformulation of the original question: "Would adding developpers to a late project help or not ?" Everybody answered yes, even some people that DID read The Mythical Man Month.
- DannoHung 16y agoOf course, the question is loaded. You're assuming that they will be adding developers who have no familiarity with the project. Project resources have lifecycles, especially for larger ones. If you have, say 200 developers on a project at the start and as the project reaches maturity, you reduce that amount to 50, those 150 other developers don't magically forget everything overnight. So, if six months down the road, you need more dev resources because you're falling behind a little, adding some of those devs back will probably help speed things up. Notice that the company that added the devs in the story has 400+ heads already. What they could be doing is hiring new guys to take on some of the non-exchange related work they were doing and shifting some project-experienced devs back on. Of course, that's the best possible scenario. What they're probably gonna do is just fuck things up.
- arethuza 16y agoNice people don't use the word "resources" in that context :-)
- DannoHung 16y agoCritically thinking people don't harp on word choice while ignoring the argument.
- arethuza 16y agoActually I accepted your argument (and upvoted) - just the use of "resources" like that is something I hate... (Note the smiley on my comment).
- protomyth 16y agoyep, never ignore the actual argument based on word choice, but word choice might give you a rather large clue on the mindset of the people making the argument.
- pmjordan 16y agoThe article also alludes to "open source developers" - this could be interpreted such that they're bringing on developers who already work on specific OSS projects to do specific integration/debugging/optimisation work. But who knows, the whole thing is pretty damn vague.
- jefffoster 16y agoWhen you're a development manager and someone offers you some extra developers what'd you say? If you say no to more developers, then it's very easy for your decision to be blamed for the project failings. At some point, the decision is probably more about politics than increasing the likelihood of the project succeeding.
- quicksilver03 16y agoIf politics are so important, that is if the system wants to be gamed, then game it: accept the extra developers then have them do something totally irrelevant to the project completion.
- JunkDNA 16y agoI am always amazed at the number of very senior IT people who have never heard of him. I read his book as a very junior employee and felt like I was in some kind of secret club compared to everyone above me. People can be forgiven for going down technical dead ends or any of a multitude of bad decisions that are only obvious in hindsight. There is zero excuse for repeating mistakes that are well-documented and avoidable. The funny thing is, it's not like his book is all theory. Anyone who has spent 5 or more years near most big corporate IT shops can readily identify most of the things he mentions in their own organization.
- fendale 16y agoThe Mythical Man Month, Peopleware, The Pragmatic Programmer and all the classic Joel On Software Articles - they taught me so much. Unfortunately, just like you I have encountered very few people in my day job who have heard of these books. I am not sure if that is reflective of a big company, or IT folk in general?
- chrisaycock 16y ago> There is zero excuse for repeating mistakes that are well-documented and avoidable. I once showed the CAP Theorem to a former employer. A few of the guys there were still convinced they could have a fully distributed system that would be consistent and available. I left soon after.
- jdp23 16y agothat was also my first reaction when i saw the headline. i first read Mythical Man Month when i was i grad school, and didn't really get it. i mean, i understood everything he was saying, but it didn't resonate with me. a decade later i was working at Digital Equipment Corporation and had much better understanding. and after doing another startup, i wound up at Microsoft. then i _really_ got it.