Getting email from cruisecontrol turned out to be a little more work than planned. Turns out sendmail has been eliminated in favor of Postfix on 10.4 and maybe 10.3, a mail server I know nothing about. After a short run through this article I was up and running.

For now I’m just running it in the background just running it in the background with root seems reasonable enough:

sudo /Applications/cruisecontrol-2.2.1/main/bin/cruisecontrol.sh &

That way it runs most of the time and I can kill it if I need do by issuing a

sudo kill #

command. Since I’m really just running it so I can help my development team understand it instead of relying on a single expert it should work well enough.

Came across an interesting idea in a directory structure for the examples in a Spring tutorial at Zabada Technologies. I’ve never seen it before, but it does solve a common problem I’ve had with mixing jars that were only needed for testing or things like the servlet-api.jar which would already be in any container.

The solution is simple. The main top level

directory just contains all the jars that are actually needed to run the app. Another top level

is just used for dependent jars like the servlet-api.jar and the junit.jar that would only be used for building.

Since I still haven’t looked into AppFuse and it’s directory structure I don’t know whether I’ll adopt this as a default for the team yet, but it is a nice simple solution. I just have to run it by our development practices committee, and convince them of the usefulness.

After another frustrating session with the example code in Spring: A Developer’s Notebook was the last straw. As a manager I don’t really code for a living anymore. On good days I get to do some coding at work, but often the day can be eaten up with meetings and taking care of problems so my programmers can work relatively interruption free.

My goal is to always stay technical, but what that means is spending a lot of my own time after hours coding, reviewing tools and frameworks, and keeping up with things like this blog. Add in that I have a family with two young daughters and I don’t have time for much else. So when some example code goes wrong it’s doubly frustrating.

So I’m off to try some online Spring tutorials at Zabada Technology.

After initially holding some development architecture decisions close to the vest with myself and the other web development manager, we decided to open up the process a bit to the senior developers on our teams. It’s an experiment of course so we’ll see how it goes.

The great benefit of opening it up is that you get a lot more perspectives and ideas. Just having the meeting opened up some surprises. It turns out after discussing what our default persistence layer framework and mentioning that we’re currently using iBatis by default but we’re looking at Hibernate or even EJB 3.0 that one of our senior developers have already put Hibernate into a module of a working product. After a discussion of source control options between CVS, Subversion, or Rational ClearCase we decided to stick with CVS for now and start exploring Subversion. Of course we also learned that at least one of the developer’s had already done some initial research and played with the Eclipse plugin.

The disadvantage is instead of coming to a consencus between one or two people, know we have to achieve concensus among six or more. Technical types can get pretty religious about things, but I think we have a pretty reasonable bunch so we’ll just have to see. We’ll be meeting every two weeks or so.

Doug Kaye of IT Conversations has proposed the idea of expanding IT Conversations into capturing audio from presentations of all major IT conferences. Then the content might be released as a bundle for $50.00 or so as a new revenue stream for conferences. Since I can only afford the budget and time of perhaps one conference per year I think this is a great idea. I’ve been able to attend Web 99, Web 2001, and SD 2005 West over the past years and I’ve found these conferences really recharge my batteries. In fact even after attending SD 2005 about 4 months ago and pushing myself through it with a nasty case of shingles I got so many new ideas I’m still trying to figure out some ways to implement them.

The only problem I see with the model is the conference organizers have to agree. It might appear to them that if people can listen to the presentations later for a small fee, they might skip attending altogether. The other problem is finding enough volunteers to do the recordings, editing, and uploads. I’d volunteer here for any conference I attend once the details get worked out.