I spent about three hours today pairing, and it made my brain really really tired.

I’m spending Mondays in the lab, with the ATAT team (Automation Tools and Technologies – we love our Star Wars references!). They’re a really strong technical team – some deep sysadmin command line stuff going on that I don’t really understand, which means that as a pair, I’m basically a rubber ducky. Today I was able to actually contribute, since we were doing documentation in the wiki, preparing for the Black Thursday/Black Friday traffic and sales peak.

I love wikis. I love the idea of them – that they’re a collaborative knowledgebase – and I love the actuality, at least in the case of Confluence, the wiki we use. Turning tacit knowledge into explicit knowledge makes me happy, and thinking about information as something that can be refactored and tuned makes me even happier.

If you’re wondering what “pairing” is, it’s doing knowledge work (usually programming, which is where the term comes from) with another person. There are a number of ways to do it, but for my money, the most effective is with a single shared screen, two keyboards and two mice. That’s what we had today, and it was great. Kevin S (I’m the third Kevin on the team, as it happens) and I fell into a rhythm of trading off keyboard and mouse, catching one another’s typos and engaging in a continuous dialog about what we were doing and the most effective way to do it. I’ve had this experience before when pairing, and it’s great. Time zips by, and good work gets done. It took us about an hour and a half to get the first page done, then we sped up to the point where we did the last two in under 10 minutes each.

But it’s tiring. It’s tiring because when you pair, you are in state of focus for a long time. Since there is another person there to take point, you’re not driving all the time, but you are mentally engaged. When your attention might flag, the fact that your pair partner is still paying attention helps to keep you in the zone. When I am soloing, I often find myself taking breaks from the primary task to catch up on email or check other distraction sources (Twitter, I’m looking at you). But when I’m pairing, I can’t do that. I actually found myself turning to my laptop to check email out of habit, and then realizing that I couldn’t do that, because I was pairing.

So constant focus is great because of the high quality of work, but it’s tiring.

Another reason my brain was tired is that pairing exercises multiple brain functions; not only was I solving problems, I was talking about the things I was doing and why I was doing them, and listening to my pair talk about what he was doing and why. And because we were trying to document, we were spending time swapping between:

• the wiki page where we were writing
• terminal windows where we were trying out commands
• AWS console windows to grab server IDs
• Google search results to look up wacky results
• other wiki pages where we were cribbing previous text

All of that contributes to a constructed mental state, which consumes mental energy.

And then, to make matters worse, what we were doing was interesting, so when Kevin took a brief break to refill his water, I kept working and finished the page we had been on. This is sometimes a breach of pair etiquette, depending on the team. In this case, I didn’t do any real work, just tightened up some formatting, fixed a typo and saved the page, and then showed it to him when he got back. There went the only chance I would have had to take a break…

So after three continuous hours of that, my brain felt like tapioca pudding.

It felt great.