Computational Biology Resources Lack Persistence and Usability

1 post / 0 new
sichan
sichan's picture
Computational Biology Resources Lack Persistence and Usability

A very interesting PLoS Computational Biology article on how open access bioinformatics software is (not) supported after publication. I think we've all experienced situations where we needed to use some bioinformatics software and encountered issues. When you contact the author, you'll usually get one of these six outcomes (ranked from worst case to the best case scenario):

1. The authors of the paper do not respond. Authors rarely respond after the first request, but some do not respond no matter how many times we contact them.

2. Authors of the software engage in a dialog and promise to provide the software. In the end, however, they never provide the code due to lack of time, resources, and (we suspect) lack of incentive.

3. The student or postdoctoral fellow who wrote the software has since left the group, and the author, who has started new projects, has no time or interest in providing the software. Sometimes, it is also revealed that the software requires subjective manual post-processing, and it could thus be argued that the published results are irreproducible.

4. The software is eventually provided, but it cannot be made to run, and the authors are not eager to help.

5. Authors say they need to work further on the software, and eventually (over months or years) we receive it, it works, and if we have problems authors continue to collaborate with us.

6. The authors provide their software immediately, assist us with our local installation, and even improve usability of their software upon our suggestions.

Read the complete article here