Xtc20030225

Previous:Xtc20030218 next:Xtc20030304

Attendees:

• MikeNew I'll be there!
• SteveFreeman Back on the ground for a change
• AdewaleOshineye - who really will make it this time
• GrahamWright
• IvanMoore
• AnotherShaunSmith
• PeterBrown
• PaulHammant
• MarcusBaker (I am praying to the god of trains).
• BrettMcHargue (I was there and wondering if I had accidentally walked into the TDD group ;)
• BillStott - a newbie ! ([email protected]). Many thanks for your warm welcome guys.

Apologies:

• RachelD

Feedback by BillStott:

• Can anyone think of a good shared metaphor to guide the future development of the group? I thought of "chatroom" but that doesn't really give much direction and doesn't differentiate us from an online chatroom. Does the group have higher ambitions than that?
• The idea of passing the sauce bottle before speaking (like in LordOfTheFlies) worked well for a while. Has anyone any other ideas for improving communication at the meeting? What about creating a list of topics to discuss in the next meeting's wiki page? Does anyone think there is value in appointing a 'scribe' to take notes and generate a feedback section for the meeting's wiki page?
• Some form of induction for newbies (like me) was discussed. Perhaps we could introduce a "buddie" system so that newcomers could be paired with an "old hand" for an hour's 1-to-1 chat sometime before the meeting. Is anyone willing to 'buddie' me?
• There was a lively discussion about the pros and cons of various configuration management tools. Does anyone think it would be worthwhile to produce a review of such tools? I'm just thinking of basis stuff like cost, availability, environment, main features, advantages and disadvantages.
• The question of whether or not it was possible to completely test a software product was raised. The room was divided between people who thought that eventually you could acheive this goal through comprehensive structural (unit) testing and the people who thought that issues of deploying products into an unspecified environment would alway preclude this goal from being reached. Would some future discussion on the nature of testing be useful?
• The issue of code-coverage not equating to testing was discussed. It was said that running a test suite that generated 100% coverage and reported no failures did not necessarily mean that testing was complete. The argument was that people could write tests that had few or no asserts and so greatly reduce the value of such testing. For example: I could get 100% coverage and no failures on a bond yield calculator that gave completely wrong results. Tests need to be written that not only exercise each line of code but check its resulting actions. We didn't get onto the subject of whether you should only test the public interface of a component.
• On the subject of communication, it was argued that we needed to produce more documentation. It was suggested that white-papers and articles needed to be published to get the XP message across. Anyone interested in collaborating on this?

The culture of the group is to keep things lightweight. We usually don't get even a little bit formal unless there's a perceived problem, and the common view has been that if you have a problem then you should fix it. This keeps us flexible but is a bit rough on newbies without the confidence to intervene. The role of the XtC is more like a support group than a structured incubator, and we've had mutiple successful events, papers, and concepts come out of our boozy discussions. Personally, I'm wary of imposing structure on our regular meetings, but in the end that's the decision of the people who actually turn up and we're on our 2nd, or even 3rd, generation of members now. --SteveF

"I agree on keeping things lightweight. We did have a bit of a problem last Tuesday because the conversation kept on drying-up and people kept on wandering in and out of the room. I left at 8:30 because the meeting appeared to be over, but just as I was leaving more people arrived (perhaps they were waiting for me to leave, grin). I'm proposing that we create a list of topics for discussion, perhaps as a set of index cards, so that people can wander in and out to take part in discussions that interest them. The index cards could be transcribed to the wiki so that people who were unable to make the meeting can still have their say - I'm prepared to do this work, but it's pointless unless other people think it would be worthwhile. So what do you think? Is it worth a try? I need a bit of encouragement!

I'm very impressed by the activities of the group, by the way - it just took a bit more time than I would have liked to get to the meat of the matters. I didn't feel in the slightest bit intimidated and hope you don't mind a few suggestions from a newbie. BillStott"

Take a look at Xtc20030107 which was one of our more structured meetings. That structure effectively emerged from a confluence of interests. If you're interested in proposing a structure then may I suggest taking a look at BookOfTheMonth and XtcProject for some related thoughts. And don't forget to take into account the NatureOfXtc

• Attendance is very fluid. Some people turn up once a month others once a quarter.
• Some people have been turning up and doing XP since the beginning of time whilst others are still curious about this extreme thingy.
• Some people are currently DoingIt whilst others are trying to incrementally apply XP in more or less hostile environments.

Thanks for the feedback, yes Xtc20030107 is the sort of thing I was thinking about. Can we try it for next week? BTW - I like the idea of BookOfTheMonth and have made a contribution. BillStott