Since I have had the opportunity to do science under my own terms, I’ve had the chance to set up some collaborations. I wasn’t trained in this- for much of my training we never collaborated with anyone… so I didn’t have any examples of what were good collaborations… and what was to be avoided at all costs. The few collaborations I was part of during the later part of my training were kind of a mixed bag- some great, some so-so. For my own group I’m shooting for a high percentage good collaborations, because I don’t like dealing with people who don’t do their share of the work in a timely way, and who otherwise don’t live up to the obligations of the ‘good’ collaboration. I’ve already covered some collaboration badness (hold overs from earlier dealings)- so maybe it is time to talk about some collaboration goodness?
First, I like to collaborate with people who have different expertise from me. Although this can have an uncomfortable period where both sides learn the scientific language of the other side, and this was very hard for me at first, you just have to get over it. I had to remember that the other side didn’t know what the heck I was talking about either- and we sort of broke it down to baby steps until we all reached the similar levels in each other’s language, enough to communicate the important details of the experiments anyway. I’ve been fortunate enough to find several collaborators who stay up nights thinking about the same kinds of questions that I think about- but from a different angle, and having these people in my scientific circle is amazing. Aside from the high I get from great data- the scientific discussions, thought experiments (that many times turn into real-live experiments), grant and paper writing with these collaborators, is extremely rewarding.
Second, what is common to my great collaborations? I’d have to say that good COMMUNICATION, honesty, generosity, an understanding of each other’s goals, and treating each other like decent people plays a big part in this- and common interests in the experiments is obviously essential. I’m not so quick to run into a new collaboration, sometimes I even do a bit of background checking with trusted mentors who might be able to provide insight on the new guy/gal. I think carefully about my expectations from the very beginning, and I communicate these expectations with a collaborator (in writing) from the get go (and want them to do the same). This includes everything from who will do what, who will pay for what, what papers are anticipated, how authorships will be divided, how shared grant monies will be divided etc- as much as can be anticipated… and in writing. With these good collaborators, I’ve never once felt that expressing my honest opinion about something would threaten the collaboration or make an uncomfortable situation.
Starting to sound a little like a care bears tea party (that one is for you PP)… I better get out of here and call it a night!
Blue Lab Coats 
Some very sensible advice in here!
When it works, collaboration is wonderful. When it doesn’t… ouch. One issue I have, though, is about the idea of putting expectations in writing. Even with otherwise excellent collaborators, this can cause offence as it’s seen as meaning that I don;t trust them to deliver. Any tips on how to get them to think it’s their idea?? 🙂
JaneB
Usually repeating everything in an email is enough, and I usually put this in terms of ‘just so I make sure I understand correctly…this is my understanding of what we agreed my role is…what you will do… whatever…’ That’s a pretty benign way of making sure the expectations are clear to everyone… and a good way to clarify them if they are not!
email is great for putting things in writing without making it sound like you’re trying to make a legal contract.
PS: That’s why I never talk to people any more 🙂