I've written up my account of how I spent my time at Elisabeth Hendrickson's Open Source Test Automation Tool Love In. My goal was to have a single script using Watir to drive both Safari and Firefox. With a bit more work I'd like to have the same script driving Celerity. I like the idea of a continuous integration setup where the Celerity tests run first for fast feedback and then the tests run in the browser to ensure compatibility.

CITCON Amsterdam is October 3rd & 4th and the final day to register is tomorrow, September 19th. We currently have 124 people registered from across Europe, ready to talk all about Continuous Integration, Testing and related topics. Hope to see you there!

In this blog entry I talk about how simple changes can produce dramatically different results. Developer testing is one of the practices, one of those little things, that add up to excellence:

"… there is no secret; there is only the doing of all those little things, each one done correctly, time and again, until excellence in every detail becomes a firmly ingrained habit, an ordinary part of one’s everyday life."

Read more.

Crap4j 1.1.6 is out. The new version features historical trends of CRAP metrics, and comparison by similarly tagged projects. Exciting!

CITCON Europe 2007 starts today in Brussels, Belgium. We're in a nice central location, very close the Brussels North train station. As an open space event everyone has the opportunity to propose a topic for discussion. The one I'm most interested in is talking about Crap4J and other ideas for metrics to spot bad code.

My normal response to a NoClassDefFoundError is to ask google. This generally gives me a thread I can follow and eventually solve the problem, but it isn't very efficient. It looks like a better solution would be the What Jar? website... if only there were more jars in the index.

Have some extra time on your hands? How about uploading a bunch of the jars to save me some time in the future? :)

Kevin says all the cool kids are doing it...

Kevin pushed a new version of JUnit Factory yesterday. The coolest new feature is that you can now execute your hand-written tests remotely on our server and get a coverage report out of it, such as this dashboard report for CruiseControl.

To make this work safely for us we run the tests under a security manager and restrict what the tests can do, so some tests will fail. But if your tests are clean according to Michael Feathers' Set of Unit Testing Rules you'll be just fine.

As Kevin posted on the Agitar Developer News page, three of us will be at Agile 2007 this week. Bob is offering his tutorial "To Catch A Bug, Think Like a Bug" and I'll be leading a discovery session with Alistair Cockburn on "Creating Change One 'Tic-Tac' at a Time". The rest of the time you'll find me at the Agitar booth, so if you'd like to talk unit testing, continuous integration, or just hang out, stop on by and say hi.

Julio doesn't like my blog -- it's too impersonal, not enough of me here for a friend. He's right of course, but I think Rands might have given me a solution in Twitter.

The problem is the tyranny of the pending entry. I typically have a queue of dev/testing/ci topics I want to blog and I find it hard to let more personal items jump the queue. In theory I could ignore that I still haven't shared the advice in On Writing Well that sounds like refactoring and dash off a note that Leanne has started he own softgoods workroom... but only in theory. My mind just doesn't seem to work like that.

Today though I read Rands description of Twitter as a yard sale for casual information and was interested. Then I read "I'd prefer to be drinking with y'all, but I'll take what I can get" and knew I had to sign-up.

If you're thinking it's been too long since we've had a beer together maybe you should sign up too?

"…AgitarOne is an easy-to-use workgroup product that greatly facilitates the use of unit tests and helps sites get as much benefit as possible from this activity. The result is shorter QA and debugging cycles and much better predictability of the software process. For many sites with large, important Java projects, this solution is attractive and compelling. All such sites are likely to derive value that far exceeds the cost."

We're always looking for bright engineers to add to our already amazing engineering team - but we're specifically looking to fill two positions ASAP - Senior Software Engineers for our core engine development and quality engineering teams.

Those of you familiar with our product would know that we solve some very tough problems ... most of our products work on problems which are defined NP-hard and then solve them to a reasonable and usable level.

So, if you're looking for a challenge and want to be surrounded by equally smart and bright individuals, if you dream algorithms, and if you have a passion to write and test software, then read the posting and apply for either the Senior Software Engineer or the Senior Quality Lead Engineer; other positions are also listed at http://www.agitar.com/company/careers.

If you know someone else that could be a better fit, I would appreciate your forwarding this post to them ... thanks for your help!

I like how Alistair is bluring the lines between Wiki and Blog on his site. Martin does something similar, but his Bliki is read only. Alistar, otoh, is using MediaWiki to run his full site, including his blog, and encouraging people to post their thoughts on the discussion page for the related blog.

I like the informal collegial feel and I'll be interested to see if a community does in fact develop.

This picture was just about the last quiet moment before they opened the exhibition hall doors. Shortly after that the horde decended and at the booth we've been just shockingly busy.

I'm encouraged by the number of people who are interested in unit testing and want to hear our pitch. We've been handing out JUnitFactory business cards like candy -- the ability to generate dashboards is a killer new feature -- and the printed copies of The Way of Testivus are popular as well. (I'm still waiting to see someone put on our karma > dogma t-shirts. Maybe tomorrow..)

I've been mostly busy occupied describing AgitarOne to people but in between I've managed to catch up a bit with Paul Hammant (who is pleased I finally have the include.projects working so he can Branch by Abstraction) and Andy Glover (who is threatening to join us in Belgium). Andy introduced me to prolific traveling speaker Neal Ford, who must enjoy being on the road far more than I do. I got within handshaking distance of ASM author humble committer Eugene Kuleshov but was too busy talking with someone in the booth to do more than exchange greetings.

Looking foward to my next break and the opportunity for some longer conversations...

Reading John's post reminded me that I should let people know I'll be at JavaOne this week on booth duty for Agitar. My favorite parts of these events are all the side conversations so if you want to talk about CruiseControl, CITCON, developer-testers/tester-developers, OIF, how to run an open spaces conference, JUF, adopting developer testing, introducing change in general (I'm looking for material for my 'Tic-Tac' session with Alistair Cockburn at Agile 2007), or The Way of Testivus (PDF) stop by the booth or send me an email and we can set up some time to talk. (Following Bret's session I'm really interested in hearing from people who have had good experiences using RSpec on a Java project.) See you there...

Jared posted a cogent argument for peer code reviews that is worth reading if only for the killer line

With enough code, all eyes are shallow.

Some developers are easily test-infected - they take to unit testing like a duck to water. Others need some time and encouragement, but eventually "get it". A third group appears to have immunity to test infection. I invent a test-gene model to categorize these groups and look at its implications for the future of developer/unit testing.

Yesterday I rashly predicted it would take me about 28-ish hours to get home from Bangalore, but now I find myself writing this from a hotel in London. It seems the relatively good luck I've had in my traveling all came to an end on this trip and over the week I've had to deal with (1) my luggage not arriving with me in Bangalore, (2) my Chennai-Bengalore flight being delayed by four hours, (3) a two-hour sit on the tarmac in Bangalore waiting for the fog to lift, leading to (4) missing my connecting flight in London. Reflecting on these minor setbacks over my bland English breakfast (I miss those idli with sambar already!) I decided there was a lesson in here on one of my favorite development/process topics, which is feedback.

It is 3:55 AM and I need to head to the airport in about 30 minutes but before my lovely 28-ish hour journey back home I wanted to jot down a repeating theme from the STeP-In Conference that just concluded here in Bangalore...

Brian Marick is a funny guy

Those in the Agile world all know of resistance to Agile from those middle managers who see it as a threat to their power to command and control. Telling such a person that her sabotage endangers the company's ROI is like an abbot standing in the path of Christian raiders and threatening them with loss of their immortal souls: sometimes it works, but nowhere near often enough. And it never works with the worshippers of Odin.

Listened to Uncle Bob give his Clean Code talk today and took some notes. Here are the bits I thought were worth sharing.

I'm in Boston tonight, in town for Software Development Best Practices. I'll be giving a talk on Wednesday, along with Alistair Cockburn, on "Creating Change One Tic-Tac At a Time". The idea for this talk grew out of a conversation that Alistair and I had back at the Jolt Awards in March and it incorporates information and ideas from a wide range of sources. For my part I'm drawing on The Moral Animal for the importance of status, Taking Charge of ADHD for the idea of a token economy, Flow for insights into what people consider rewarding, Influence for some "weapons of persuasion", not to mention a host of others I can't name, plus my own experiences leading development teams and our experiences at Agitar helping our customers adopt developer testing. These last two categories are probably the most important, because that is where I've come to believe that cultural change is simply the most difficult task anyone can undertake, and changing the practices of a development team in any significant way require a change of culture. Alistiar has a similar view on the importantance and difficulty of cultural change and brings his own diverse and illustrious experiences to bear on the discussion. If you're at SDBP hope to see you there, but if you're not I'd be interested in your thoughts on the topic (j t f at a g i t a r dot c o m).

Scott Ambler calls it Green Shift, but it sounds more like Bull Shift to me. (via Jason Yip)

Nine gathered tonight at the Thoughtworks San Francisco office tonight, 4 of whom attended Agile 2006 and 5 of us did not. Things I remember from the conversation, in no order of importance:

• Agile has crossed the chasm, "everyone doing it" was the word on the street
• Success stories even in very large teams/companies
• Wanted more discussion on how to change culture
• Coding Dojo session was excellent and would make for a good future BayXP meeting
• Data no longer a four letter word! (Refactoring Databases)
• Surprisingly valuable discussion by some Microsofties of their TDD Pair Programming Game (complete with phase transition diagram)
• The Open Spaces part was allocated a horrible space -- think low ceilings and harsh lighting -- and this combined w/the relative neglect compared to previous years led to a feeling of low energy...
• ...except the Getting Things Done session got high marks
• Crushing Fear Under the Iron Heel of Action got high marks with Ron and Chet described as living cartoons (in a good way)

I know there was more but I didn't take notes, so that's all you get, unless some kind soul adds more info in the comments.

James Shore is involved in what seems like an great opportunity for computer science students at Portland State University to learn some real world sklls. Along with Dr. Andrew Black, PSU Professor of Computer Science, they are putting on the course Extreme Programming: Principles & Practices. If you're up in Portland, Oregon it looks worth checking out...

It's like what Whitehead said about notation: "By relieving the brain of all unnecessary work, a good notation sets it free to concentrate on more advanced problems, and in effect increases the mental power of the race."

Outstanding post, Brian. I always wondered what the little star was for in GMail. Now I know. Your post has a little gold one next to it.

It doesn't matter if you call it Refrigerator Code or Girl Code or Beautiful Code or simply Clean Code, the meme is out there that is isn't enough to say the code works and then leave it at that. Even if you find yourself unmoved by Kathy's aesthetic arguments you should weigh Uncle Bob's assertion that "keeping your code clean is not just cost effective; it’s a matter of professional survival."

He's actually practicing two of the best practices I know of: regular one-on-ones and code reviews. In my own management career I found it very hard to stick w/the weekly one-on-one schedule even though I believe it is hugely important. And code reviews are probably the best sofware development practice that (virtually) nobody does. This guy is obviously working very hard to put theory into practice and for that he has my respect.

The Pragmatic Programmers recommend learning a new language every year. Not because you need to know a whole bunch of languages, but because other languages use idioms that you might not think of using in your everyday language.

Tight feedback loops provide excellent learning environments. Bacteria learn to beat an antibiotic in a decade or two. But it will be quite a while before dogs evolve their way around the doorknob.

I would like to introduce myself as a new Agitator. I am Bob Evans and I am working on some of the next product features for the Agitator.