What Pair-Programing is Not
People often ask how can I justify two people when one will do. Will this not, just double my cost? To answer this question I think it is important to discuss what pair-programing is not.
How much of your time do you spend writing code? If you answer 100% than you are lying! Between checking your emails, going to meetings, talking to people, mentoring others, finding bugs and thinking about fixing them, coding is something we do surprisingly little of. I only do coding about 10% of time, but i am probably at an extrema, so lets say you code 25% of time. Even when you are coding, you spend a lot of time testing what you have just coded, and realizing what you have just done does not work. Your end product may be 100 lines of code, but it may take you weeks to get there.
Out of 25% of coding how much do I pair? Not enough, but lets say 25% of the time. Overall you are only talking about little over 10% of my overall time I am pairing which you can say that I only cost 10% more (if you can apply accounting logic to coding). But what does this extra 10% of “cost” buy me?
At most places I have been at when a new developer shows up, they hand him a computer and say, “OK, get up-to speed and build this feature X.” Then most places do not expect anything out of the developer for few months while he “gets-up-to-speed.” Few months? I expect my developers to check in a feature the first week! But you can’t just demand it you have to approach it differently.
Let me tell you about Arthur, the last developer which joined us for a month, and how productive he was. Arthur showed up on Monday at 9am. I sat down with him behind his computer and said, “Let’s get you up-to-speed.” I made sure that he was driving, and I was talking and we started installing all of the software which he needed to develop code. The compilers, the database the IDE, checking out the code from the repository and running it. The key here is that Arthur was doing all of the work and I was just telling him what to do. By noon, he was up and running, thanks to a very intensive hand-holding from me. We went to lunch and talked about the application, not in general terms but in specific terms which Arthur could translate to what he did few minutes ego.
After lunch, I said let’s go and fix a bug. I picked a simple bug and for the most part again i was doing most of the talking but made sure that Arthur had the keyboard and understood what, and why he was doing things. By 3 pm we had a fix. We ran all of the tests, and submitted the code. We watched the build to make sure we did not break anything.
This is pairing, and it is exhausting! By 3 pm I was beat and went home, but Arthur stayed to play some more with the code. Turns out he fixed another bug on his own before he went home. Not bad for first day. Now I could have fixed the bugs myself in 1 hour flat both of them, but then Arthur would not learn anything. I sacrificed my productivity to make Arthur productive in a single day. If I did not it would take Arthur weeks before he would figure out how to set everything up how things worked and enough courage to fix a bug. Yet that is exactly what most companies do. Think about the confidence Arthur had on day one working with us. He was up and ready and he fixed two bugs on day one. WOW!
More importantly is what we did not do. We did not get distracted by checking email. How can you, you have a single monitor, and that means that Arthur would read my email. Arthur did not waste time looking for documentation which is poor or does not exist, he was spoon fed the knowledge in the most efficient way, and he did it all by himself, i was just talking, he was typing. It is amazing how much more focused you are when you have two people working on something.
Next day, we paired some more, but this time we started to build a new feature. Again I was taking Arthur through the implementation making sure that he did all of the work. By the end of the week Arthur has implemented several features and the amount of pairing we did has dropped of significantly, but instead Arthur started pairing with other engineers which had the know how which he needed to get the features done.
At the simplest level pairing is helping people one-on-one, not through emails, let me show you behind a keyboard, instead of telling you in abstract. The amount of pairing and keeping each other on track depends on many things, but you will find that more you pair the more you will learn from each other and the better the code becomes.
But Arthur was a new engineer coming up-to speed, what about existing engineers? Well let me tell you story of Pepper the senior tech lead on a project. Pepper came to me and said, Misko, I don’t like the way we deal with X in our app, what could we do to make it better. Now here is a case where Pepper has all of the know how of his app, and I know next to nothing. So we sat behind the code and I started writing what I would imagine an ideal app would look like, knowing very little about the code. Pepper had to explain why my view was wrong or too simplistic and in the process we would learn from each other and slowly converged on an ideal design, which was neither his nor my idea wholly but a collaboration of our ideas. We started refactoring, sometimes I would drive and sometimes he would. We would take turns painting ourselves in the corner and then backing out. Many times Pepper would work without me, as the refactoring was tedious, but straight forward and two of us were not needed. All together it took us 2 weeks to finish this refactoring. Pepper did the lions share of the work, but we spent about 3 full days behind a single computer together. That is what is pairing.
If you would ask Pepper, he would tell you that pairing helped him focus. He would say that without the pair he would not have had many of the ideas which we ended up having. Most importantly there are now two people which really understand the details of the design. When I or Pepper will pair with someone else, we will share these new ideas with the new developers. As team pairs with each other ideas and knowledge multiply and flow making sure that the code-base remains consistent no matter which place you look. It is as if a single person has written it. This is a great thing which helps with maintenance.
Pairing does not mean that two people always do everything together and checking each others work, that would be boring and a waste of time. Pairing means that people bonce ideas from each other and have discussions, not in abstract, but behind a keyboard, keep each other focused and on track. Result of which is code which is better than the sum of its parts. Pairing is about sharing the knowledge with your team mates so that everyone is on the same page. We will go further if we make sure that all of us pull in the same direction, something which is rare in the industry at large.