Posted by Jeff April 19, 2007
I’m getting google hits for the phrase “eclipse automatically run junit tests”… but that’s to Nosy, which is for python code.
For java/junit code, use the ct-eclipse plugin. It runs your tests, in the IDE, whenever your code changes. At the current time it only supports Eclipse 3.1 (!):
CT-Eclipse (Continuous Testing for Eclipse) builds on the automated developer support in Eclipse to make it even easier to keep your Java code well-tested, if you have a JUnit test suite. With CT-Eclipse enabled, as you edit your code, Eclipse runs your tests quietly in the background, and notifies you if any of them fail or cause errors. It is most useful in situations where you would already have a test suite while you are changing code: when performing maintenance, refactoring, or using test-first development.
It’s all about quick feedback loops.
continuoustesting, eclipse, java, junit, unittesting
Posted by Jeff April 19, 2007
Alt-Enter: Send msg
g i Goto inbox
c Compose
/ Search
# delete
r Reply
a Reply All
f Forward
The whole list is here. It’s a shame that ‘#’ is difficult to type, requiring the Shift key. Presumably using the ‘d’ key was too dangerous.
gmail
Posted by Jeff April 17, 2007
I was reading an interview with the comedian Demitri Martin, and it turns out he’s a deep thinker–
At some point I created a point system, like breaking my life down into categories, and then in each category trying to achieve certain things in a week’s time. Every Sunday night I would tally up what I had achieved, for a total possible of 35 points. It was mind, body, career, personal management/relationship contribution. It was pretty funny. It was really ambitious in retrospect. The stuff I set out to do each week was pretty much impossible. I kept track of it for 27 weeks. I had a binder in which I actually was consistent for half a year. Every week I’d carry it around with me. I never got 35 points. I never even got close. Years later I found it, and I was like, “Oh my god, this is crazy.” 4 points was my lowest week, and I think 24 was my highest. When I made the system I figured I’d be topping out in the 30s, and once I’m close to my maximum, I’ll just bump up each category, I’ll just make the goals a little harder. And then that way I can develop a balanced set with the different things that I’m trying to learn how to do. I averaged 11 points out of my own system. So I failed kind of miserably. But the cool thing is that ever since then I haven’t really ever been bored. I haven’t watched TV since then, and I just never really feel like there isn’t something to do. That changed my perspective. So it’s like, draw a picture if you’re sitting somewhere, or write something down, or write a palindrome. It’s just about all the different opportunities in one moment. It changed my perspective on time and creating things.
This made me laugh because it’s exactly the kind of system I imagine building. Turns out someone else has built it - it’s called joesgoals. JoesGoals sports a stunningly intuitive and friendly user interface, which you can play with on the site:
This is friendly data collection — like a CRF in a clinical trial. What can you do with the data? Determine cause and effect (causal analysis)? One person found a correlation between cycling and migraine relief:
Bradley has been using Joe’s Goals to track his migraine attacks (-1 point), the times he has to take medication (-1 point), his cycling activity (+1 point), and when he wakes up in the morning (+1 point). Printing out the 30 day report he took it too his neurologist and the results surprised both of them. After reviewing the results the neurologist looked up and said “Why would you ever stop cycling?”
Like most geeks, I have an interest in self-experimentation, but not the discipline to gather data reliably. Seth Roberts discovered the Shangri La diet via self-experimentation. The spirit of curiosity and observation persists in his readers - see this post on the effects of DHA omega-3 oil. His seminal paper Self-experimentation as a source of new ideas describes his 12 years of self-experimentation as a basis for idea generation. It’s basically R&D, finding unexpected correlations.
JoesGoals is myware and a way of turning mundane tasks into a game. The simple data model is blessing and curse. It lacks numeric variables. This means you can’t record interesting variables such as weight, mood or sleep. Hopefully it’ll move in that direction in the future. The creator has a real knack for good UI. With a simple CSV download and good analysis software, users might discover other unexpected and helpful correlations.
causal analysis, GTD, joesgoals, selfexperimentation, sethroberts, shangrila
