Edmund Schweppe wrote: > I've been using Goldratt's definition, as given on > pg. 139 of my copy of _The Goal_: > > > "A bottleneck", Jonah continues, "is any resource > whose capacity is equal to or _less than_ the demand > placed upon it."
But in an XP team working at full speed, the additional demand is not placed upon any individual, it is placed upon the team. So the team may /be/ a bottleneck, but the team does not /have/ any bottlenecks within itself. > Obviously, this isn't the same as the definition you > gave below: > >> A bottleneck is a required step, or passage, which >> is performed at a slower rate than the average work >> capacity. I was not attempting a definition: merely contributing to a conversation, expressing a facet of the idea that the term "bottleneck" conveys to me. > I'm pretty sure that I don't agree with this > definition, by the way; a very quick process can > still be a bottleneck if the load on it is too > high. Consider hitting a Slashdotted web site over a > dialup line. Even though the web server's routers and > Internet connections are far more powerful than the > pokey little modem, there's a good chance that the > bottleneck will be the web site's connection, *not* > the modem. Yes, of course. It seems to me that my description of bottleneck fits that, if we consider /actual/ capacity, in real-life scenarios, not /theoretical/ capacity in idyllic circumstances. >> If everyone in the team can, at any time, add >> finished features to the software, there are no >> bottlenecks. The velocity is still limited by >> people's work capacity. > > There may not be any bottlenecks (by Goldratt's > definition) in the *development* parts of the team, > but there is going to be a bottleneck > *some*where. Probably on the Customer side of the > Whole Team. It seems to me that part of the Customer role is /placing demand/. So, by Goldratt's definition, that would not be a bottleneck, would it? It could even be argued that /all/ of the Customer role is placing demand. Regardless, I think that some of the Customer's work can (beneficially) be shared by the rest of the team (including developers), and also that the Customer role can (beneficially) be carried out by more business people, in such a way that the Customer capacity balance the development capacity. > Let's suppose that the team has demonstrated that it > can sustain a velocity of 20 points per week. Then > the Customer side finds that it can only generate 10 > points worth of stories per week - due to big > arguments over corporate strategy, for example. So > the Whole Team slows down to 10 points per week for > several weeks. Now, if the big corporate strategy > arguments are resolved, and the Customer can get back > to writing 20 points worth of stories per week, > wouldn't you kind of expect that the developer part > of the Whole Team could still sustain a 20 point per > week velocity? Yes, I would expect that. I don't think it demonstrates that an XP team has bottlenecks, however. I think it demonstrates that it can be difficult to provide a steady stream of definitive stories to an XP team. One way to deal with that is to provide a steady stream of "as good as we know today" stories to the team, and use feedback to nourish the big arguments over corporate strategy. Regards, Dominic Williams http://www.dominicwilliams.net ---- To Post a message, send it to: [EMAIL PROTECTED] To Unsubscribe, send a blank message to: [EMAIL PROTECTED] ad-free courtesy of objectmentor.com Yahoo! Groups Links <*> To visit your group on the web, go to: http://groups.yahoo.com/group/extremeprogramming/ <*> To unsubscribe from this group, send an email to: [EMAIL PROTECTED] <*> Your use of Yahoo! Groups is subject to: http://docs.yahoo.com/info/terms/
