XtC27022001

Another good evening chaps. Well done. It was nice to meet Ken and experience a less manic XpGuru.

KenAuer is teaching in London 28ii-2iii, and will be at the meeting. I propose to run it as I did the Kent Beck night i.e. have a mildly-structured Q&A session, perhaps with some small-group breakouts. Other proposals welcome of course. I expect we will arrive about 6:30pm, eat, start about 7:30pm - see how it goes. BruceAnderson

Points:

• the crowd usually firms up around 7:30, as some people have to travel.
• the kitchen closes at 8:00, and they tend to run out of food there.
• I suggest we defer any structured activity until after the plates have been cleared.

• KenAuer
• BruceAnderson
• SteveF
• StuartBarker at long last! + 4 From the Vizzavi Team (Equant: BeatrizGonzalez?, MattRyan? and JonVaughan?, Deloitte: MartinSpedding?). VincentMassol may also be along, if he's in London.
• DuncanMcGregor, also GarethSB, plus PeterBunyan? and RichardCare?. Maybe JeremyGreen? as well.
• RachelDavies
• KeithB
• IvanM
• TimM
• JohnClayton - slots himself into the middle here...
• DaveKirby
• AdamGilmore
• PaulSimmons
• ChrisCottee
• and that makes 21!
• ...any room for a little one (SallyMoss)?? [so - who are you sending?]
• AdrianHowarth
• JamesGamble
• SvenH Looks like it'll be a cosy one tonight...

On a project which requires FDA sign-off for the testing he has got an automated acceptance testing framework going. It was easy to persuade the customer to do this because they had to do testing anyway and automating it saved time. They either get the customer to write the test or write it themselves and show it to the customer. They are testing GUIs by driving the event model rather than recording mouse clicks and have a simple language to specify the test. They don't have customer on-site all the time but get round this by using a wiki visible to the customer and have frequent emails saying what what their assumptions are. Plus asking customer how they would test a story up-front.

-unreliable recollections by ChrisCottee who left early and thus missed out stuff

• Ken's serious long-term commitment to the studio method and to an apprenticeship system
• some customers aren't worth working with i.e. just arent' suitable for XP (or whatever your approach is) and it's better to look elsewhere for work
• some people reported success with sneaking XP into an organisation in a small team, succeeding with their project, and then telling how
• discussion on generating code and schemas, sparked I think by Ken's saying he had a test spec language that generated tests
• a crowded room of interested people
• some kind of resolution of the "XP doesn't recognise architecture" or "XP magically creates an architecture even if you don't know what that is" discussion, namely that you do need to recognise which stories will drive the architecture and make sure they get done early, despite what the users might say

BTW Ken's course is going well and I expect he will be asked back