Ted Graham on .NET

As my current client is early in the architecture phase, I’ve been reviewing the MS Data Access Application Block. I did a brief experiment, and I’m having problems retrieving output parameters from stored procedures. I created a simple stored proc (with CodeSmith):

Roy (a fellow blogger) wrote, "Interesting! Can you write some of your experiences about working XP style? There aren't that many of us practicing it... :("

I haven't been doing pure XP. My last two projects have been done with a partner and we used just enough methodology to get the project done predictably. Of the 12 tenets of XP, we used OnsiteCustomer, PairProgramming, TestDrivenDevelopment, ContinousIntegration, RefactorMercilessly, SimpleDesign, CollectiveCodeOwnership, CodingConventions and SustainablePace. We didn't do PlanningGame, SystemMetaphor or SmallReleases. We didn't have an XP coach and didn't involve our customer in feature prioritization as much as we should have.

A negative observer might have called our process HackAndSlash. For two experienced developers who work together comfortably, pair programming and lots of nUnit tests keep things in good shape. The Wiki I linked above is a great place to learn more!

It seems like most software shops write their own bug and issue tracking solutions. As a consultant, I see lots of projects and a surprising number use a homegrown solution. I just got out of training on my current client's system and it has so many gotcha and special behaviors that the demonstrator made several mistakes. I've always wanted to use FogBUGZ but I've never had the chance. Do you have a favorite system?

From initial discussions of the technical architecture I think we are planning on handwriting all the stored procedures and the data objects that wrapper them. I'd much rather use LLBLGen or another data layer tool to automate this, but it may not be my decision. I'd love to hear anyone's experience using data layer tools. Ideally I'd like round-trip generation, but could consider generating the initial objects and maintaining them by hand. I read the Great Data Generation Layer Debate with interest, but it was a discussion of different tool styles instead of first hand reports on using the tools. Regardless of how we build the data layer, I'd like to use nUnit to test it. I've had great experiences with nUnit and many of the consultants agree, but the MCS architect I discussed it with thinks writing the tests will slow us down.