Closed JHartzer closed 1 year ago
Phenomenal comments @JHartzer ! Thank you! By following your points, I have worked on the statement of need and the conclusion. Let me know what you think 😃
Excellent @mrsonandrade! My apologies on the delay. Thank you for doing this work and I will go ahead and close this issue.
I think, like mentioned in the other review, there needs to be additional work on the statement of need. In its current form, I don't believe the need is clear or it aligns with the solution that is being presented. The cited paper by Nedjah and Silva doesn't appear to call for this kind of work, but rather a standard controller software that would be used on-machine with a standard hardware platform. Since this work focuses on the congregation of swarming behaviors that can then be combined into swarm algorithms, I don't think it solves the need cited in this paper. However, I do think there is importance in this work. Specifically calling out the need to quickly prototype and develop new algorithms in python and leveraging existing datasets would more closely align with the solution that is being proposed. This is mentioned later in the paper as well, but I think it would strengthen the submission and this section to add a similar note here.
I think the conclusion could be expanded as well. Both sentences are a bit clunky, start with adverbs, and run on a bit. It might flow better to have a more academic paper conclusion "presented in this work is pyswarming, a package for... This package provides behaviors for swarming robotics... Written in python, it is more flexible and allows access to datasets... Therefore... etc."
I think improving these two sections would greatly improve the quality of the paper, as the other sections are quite strong.
Nitpicks: