Working with better programmers than me > Working solo > Working with others
I guess it is a common preference
1: http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
For me, it's about experiences and perspectives. When I work with other programmers who have unique approaches to problems that help me work more efficiently, I enjoy working with them.
I also enjoy working with less experienced programmers who are willing to learn and practice new, more efficient methods.
I really dislike working with programmers who think they are awesome, but are really clueless. I try to help them, but their pride gets in the way. It's funny when they try to school me on stuff. I just don't have time for that. I feel bad for their clients.
But sometimes I run across people who know little tricks or constructs and I'm like, Wow, that's awesome, teach me more!
And i think it's great to be able to dictate code out loud. And it's a lot easier on your wrists.
Even though he's not universally popular around here, I'm going to admit to liking Seth Godin's "Let's call it the talent department" argument: http://sethgodin.typepad.com/seths_blog/2008/02/marketing-hr...
At my current work, I am the only programmer in an office of designers and sales. We (I) make CMS and other custom systems for clients. We really want to move to a generalized platform instead of one-off custom systems, but due to a number of issues, we can't.
One of those main issues is that there is no one like a programming buddy to "talk shop" with. I feel I am an OK coder, but everyone gets those times when they think "Should I implement this with A, or use B? Does this need to be extensible, and if so, to what degree?"
These are questions that usually cannot be answered by someone who is not deeply involved in the project, and someone who knows what the resource costs are for implementing features, as well as the cost for needing to implement those features down the line.
I can of course work on a team when the project is beyond the scope of a single developer. I prefer to work alone, since that entails exactly zero coordination overhead.
It's easy to get into the trap of thinking it's all about code, but there are lots of ways to add people. Sometimes, coders don't want to add people, because it's one more person asking them to code something. Coders often don't know how to push back and say, "This is what I'm working on, this is what I need you to do."
Every answer to an IT problem is "It Depends." In this case, the question of "Does adding more people make the project go faster or better?" is "It depends."
This question can't be answered with a general absolute.
That sounds awful. That's three people who have to coordinate in order to display a new data element, and a fourth if the new data element has any kind of interactive functionality.
It's much better to break the project up into functionally independent pieces, and have developers who can work with all of the technologies each piece touches. You're much more likely to lower coordination overhead that way.
As far as the actual programming, yes, I'd prefer to do it alone. I love being able to make major decisions (think: changing an internal API) without having to consult with someone.
We all have headphones that we wear as a "do not disturb" signal. I feel most productive on days when I can dive in and wear my headphones for most of the day.
I think I'd like working by myself if given the chance, but I've always worked for fairly traditional companies.
I have no need whatsoever for a "coding buddy." I'm considerably more comfortable working on a project solo. Even with a good team, I hate the coordination overhead.
I can, of course, and good team skills are important. I'm not saying that. But the coordination costs of software development increase super-linearly, so I tend to prefer using the minimum number of people that can possibly do a job. The break-even point is around 4-5 - after that, adding more people makes things slower, unless you can split the project into autonomous chunks.