RE: Inspections Was: RE: [AM] ANN: Mashing Deadly Myths
Paul Oldfield <[email protected]> Wed, 25 Feb 2004 04:22:35 -0500
| Newsgroups | gmane.comp.programming.modeling.agile |
|---|---|
| Message-ID | <[email protected]> |
(responding to Paul T) > OTOH, the ugliness of pair programming is that ten people > who don't know the answer to a specific problem individually, > also don't know it collectively. One of the advantages of having a team that communicates is that when somebody needs knowledge or skills they don't have, they're likely to ask. Two questions arise; what happens when they find that nobody knows? and what happens when it isn't noticed that somebody needs knowledge or skills they don't have? Experience reduces the incidence of the second problem, and pairing inexperienced people with experienced ones will help. What happens in the case of the first problem will depend very much on the team, but a team that has a culture of asking for help may look outside the team before trying to develop the knowledge and skills internally. (That was Steven Gordon's point, I believe). There aren't any 'right' answers here. However, an inspection of work that has only been seen by its creator is much more likely to discover problems than would an inspection of work that has been seen and worked on by many of a team of ten, where that team is collaborating effectively. As Scott Preece points out, a study shows that inspections after a single pairing leave 15% fewer defects in the code. One cannot be sure how many more 'visits' to that code will occur as work progresses; I tend to use a rule of thumb that each team member is likely to know 60% of the code, so maybe another 4 of the team will look at the code during subsequent development. However, we cannot expect such subsequent visits to give the code a thorough inspection. One further consideration to take into account is that when pairing, one learns not to add the defects that your paired partner knows not to add, so that over time the 15% of the study is likely to grow faster than for the single programmer. (I think that's J.B.'s point?) Paul Oldfield ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ www.aptprocess.com any opinions expressed herein are not necessarily those of Mentors of Cally or the Appropriate Process Movement ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ For more information about AM, visit the Agile Modeling Home Page at www.agilemodeling.com --^---------------------------------------------------------------- This email was sent to: [email protected] EASY UNSUBSCRIBE click here: http://topica.com/u/?bUrKDA.bWnbtk.Z2NtYS1h Or send an email to: [email protected] TOPICA - Start your own email discussion group. FREE! http://www.topica.com/partner/tag02/create/index2.html --^----------------------------------------------------------------