This week’s highlighted chatter is about the never-ending quest for better ways to access and use other people’s software. I don’t think there’s anything new here, but it may be a nice reminder for developers that others want to use the things you are developing–make it easier for them to do so.
This seems to be a reoccurring theme: I read a cool new bioinformatics paper that develops some method for doing exactly what I want to try out on my data. I try to find the code so I can apply the method to my data. Some times the code is not available so I have to contact the author. Other times, the code is available but so poorly documented that I have to contact the author and ask for clarification. Most frequently, the code is available, reasonably documented, but takes some strange input format that I’m not sure how to massage my data into and I spend a lot of time just getting everything in the right format.
What are some of your tips, suggestions, or recommendations for developing more user friendly bioinformatics software? There must be industry standards that we can learn and borrow from.
The ensuing discussion was valuable. Good ideas, good techniques. Have a look.