6 Matching Annotations
  1. Apr 2022
    1. Many of the examples are specifically meant for first-time GitHub users and can likely be adopted with just several hours of practice (e.g., storing data, creating virtual notebooks, making code citable)

      Here, or elsewhere, may be good to point out that while there is a startup cost to using GitHub (learning the skills listed), it is also a way for students to develop skills in VCS/GitHub which can help them to be successful in academia and/or industry.

    2. could link this back to the “Project Management” section or even “Collaborative (code) editing”.It actually seems to me like this entire section could potentially be combined with “Collaborative (code) editing”

      I agree, I think that the asynchronous working is better as a part of the Collaborative code editing than as its own section

    3. these code owners can change through time allowing for the transition of research software from one cohort of researchers to the next

      HW:another point worth mentioning here is how GitHub allows for archiving package versions that were used when the code was developed. E.g. if users include an enviornment.yml and/or an environment.txt file, it ensures that as packages continue to develop with new syntax, code will still run as it did when it was developed.

    4. .

      Beyond just using GitHub for peer-review of manuscripts, it may also be worth mentioning GitHub as a way of conducting peer code reviews within a team (although this may be better suited to "collaborative (code) editing"?. E.g. if I request to merge a branch of my code in, I will request a review from a colleague before incorporating it into the main branch. This by the design aids team members in improving their code by looking at code from other collegues. While many researchers do not set aside time to review code of people in their group, it can be a good way of structuring time.

    5. ong-term data archiving service funded by CERN

      My understanding is that Zendo also allows for storage of larger files? Documentation mentions up to 50 GB per dataset

    6. A collection of files (e.g. a directory) tracked by git. Commonly shortened to “repo”

      It could be helpful to mention here that repositories can be public or private since it is referenced later under "fork", and in a few application sections. Something like "Repositories are managed by an owner and can be listed as "public" where the repository will be visible to all github users or "private" where the repository is visible only to added team members."