A recent Sprint goal on an intranet project went something like this:

To migrate most of the intranet content and train at least one content author from each division.

On this Sprint there were probably 8 or 9 actual product backlog items, but the main theme was apparent–moving content. The word most was used because given that we didn’t know the entire scope of the content we thought 50% plus was probably reasonable. As it turned out we were able to migrate 80% of the content within the Sprint. Still the idea is to set a goal and then leave yourself a bit of wiggle room. You really can’t predict how the next 30 days will go at the start of a Sprint.

Many product backlog items we signed up for were left out of the Sprint goal, but the theme that we had to move the content was clear. If you try to cram in too many backlog items into a goal statement you just end up with a long paragraph that no one can remember. It’s much easier to get behind a simple theme for the Sprint.

The theme idea also applies well to parties. Our oldest has had a theme birthday party each year from rubber ducks to a Hawaiian hula party. A Sprint that runs like a theme party complete with a pithy goal statement is off to a good start.

On vacation in San Diego I ended up at one of my favorite locations, an Apple Store. Turns out with two 5 and a half year olds the Apple store is a mini entertainment center.

Apple’s Photobooth, one of the free little applications you get with all the new Macs, proved a bigger hit than I anticipated. My daughter and her cousin took turns trying out all the possible effects from a Fisheye lens effect to the ‘Light Tunnel’. I had to pull them away to cries of:

“No, Squidy needs to be in the next picture.”

“Please, just one more picture.”

It’s a cute little app that even fakes a flash effect by flashing an all white screen before taking the picture. You wouldn’t expect Photobooth to convince someone to buy a Mac, but subtly it adds to the whole experience you have in the store:

• It entertains two wired 5 year olds for 30 minutes easy.
• It’s so simple there’s no confusion on how the interface works.
• You’re having fun which isn’t often isn’t the case with your typical Dell box at work.
• It gives the shopper a reason to stay with a display machine in the store and walk away remembering how much they enjoyed the cute little photo program on those Macs.

So even if you never use Photobooth at home this little app has helped improve your whole Apple store experience. Clever.

I had quite a few one-on-ones with developers today and almost all of them started with:

“Wow, I didn’t realize you took out the desk.”

Last Thursday my desk partition was removed and replaced with a smallish circular table for meetings and to have more of a team room atmosphere. My normal office these days is a cubicle out on the floor with everyone else. It’s a bigger layout, but I think I still surprised some of my team. In an effort to seek improvements however small, I’m constantly adjusting things.

This time it was an office desk versus a table, tomorrow it may be adjusting our code review process or adding an impediment chart to a Scrum project. I think it’s the scientist in me that loves the experimentation and testing of a hypothesis. If you try out a lot of ideas some of them turn out to be really useful, and who wants to do the same thing all the time.

Confluence is available for a free license for non-profits who can’t afford to pay. Our local java users group (SACJUG) fit the bill for this, though no one’s ever been quite organized enough to officially register as a non-profit.

At the last SACJUG meeting we debated over what wiki/content management system to use and eventually settled on Confluence. The feeling was as a java users group we needed to use a Java based tool. I volunteered to see if they’d grant a license since my company is already a Confluence customer. About a week later I got a nice email from Atlassian granting SACJUG the right to use Confluence for site.

Kudos to Atlassian for the program for non-profits. And if you’ve tried several wikis and they never caught on you might try out Confluence. It’s really taken off in our organization and the pricing is really reasonable.

I’m a tools guy. I love to try out new tools. I get excited by the most boring of build tools like ant, maven, or rake or even code review tools like Jupiter or Crucible. In the book Applied Software Project Management they introduce the concept of building support for change through tools:

Another way to approach this situation is to pitch the changes you want to implement as technical tools, rather than as core software engineering concepts. Most people who have been in the field for a long time are used to routinely applying new technical tools that have never been tested in the organization. This is also a good way to gain concensus among the programming team for your ideas.

— pg 217 Applied Software Project Management

If a tool can help with adoption of a new process I think it can make sense, but it’s a dangerous path. Relying on tools for everything can force you into things like Model Driven Architecture. And sneaking practices in with tools is a might bit disingenuous and less than fully transparent. I prefer to use tools to support practices I’m introducing.

Some tools make the practice possible:

• Cruisecontrol for Continuous Integration.
• JUnit for unit testing.

Some make merely support the practice and make it easier:

• Clover to help with TDD.
• Crucible to help with code reviews. (especially if you’re not pair programming)
• Checkstyle to help with code reviews.

Tools can support multiple practices much as XP practices support one another. Pair programming reduces the need for written documentation, enforces real time code reviews, and leads to collective ownership of the code.

Finally, never be afraid to ditch a tool that really isn’t working. I tried using XPlanner for managing a Scrum project, but it really just took more time than it was worth so I dropped it after a single Sprint. If you’re fighting the tool or caught up in tool complexity you’re better off stepping back and dropping it.