The fact that papers are written for the audience of critical reviewers hinders adoption: possible users are not aware of academic mores and will be discouraged by described limitations.
With adoption in this context, I mean that people outside of your direct group of colleagues and acquaintances will use your algorithm/tool/method. I pose that the way papers are written currently is not optimal to convince people to use it. Many aspects are irrelevant to people who simply want to use your idea. like related and future work and even references to some extent. But more important is the way the benefits are presented: lots of limitations are usually named as well as threats to the validity of the evaluation.
For someone simply looking for a good tool to do X or a method for Y, this might be off putting. Especially if you compare this to how products are normally being marketed. And not just in your local supermarket, this also holds for software:
“absolutely the best thing that happened to my inbox in recent months!” (Boomerang)
“Simple, fast and powerful media player.” (VLC)
“Google Chrome runs websites and applications with lightning speed.” (Chrome)
“Ubuntu is the world’s favorite free operating system, with more than 20 million people preferring it to commercial alternatives” (Ubuntu)
That makes me actually want to use and try it! And these are all (sort of) free software systems, so money making is not the incentive here.
The fact that we need to phrase our tools in a certain way in paper will not automatically attract users and I think this is something we need to be more aware of. So make a nice website for your tool or method, write a blog post on it and get it on Reddit.
Simply put: think about how to “sell” your research, even if you’re not selling it.
———————–
Note: Boomerang for Gmail happens to be the best thing that can happen to an inbox, but that is irrelevant here.
Perhaps a small detail, but I think it would already help a lot if there were some standard way to refer to your tool’s website from the paper. The way it currently works is, if someone wants to find your tool, he/she has to browse through the entire paper, in the hope of finding a link hidden away in a footnote.
While I don’t think a paper is the right place to actively “sell” your research to potential users, it shouldn’t discourage adoption either by making the tool hard to find. Instead of linking to the tool’s website in a footnote, it should be directly on the front page; perhaps it should even be included in the paper’s BibTeX. Anyone who may be interested in the tool from a user perspective can then easily find the website, which would be a suitable place to convince someone to give your tool a whirl.
“perhaps it should even be included in the paper’s BibTeX”
This is an excellent and very practical suggestion! You should adopt it as one of your propositions 😉
I would add than to put the tool on a ‘safe’ place, like GitHub or SourceForge, rather than on a personal or university website and it would be even more perfect.