10 year reproducibility challenge: reproduction of J. Chem. Phys. 112, 4417 (2000)

[mcbaneg]

article.pdf

Original article: Gottfried and McBane, J. Chem. Phys. 112, 4417 (2000)

PDF URL: https://github.com/ReScience/submissions/files/3882888/article.pdf
Metadata URL: https://github.com/ReScience/submissions/files/3888185/metadata.zip
Code URL: https://github.com/mcbaneg/tenyears-submission (also: http://faculty.gvsu.edu/mcbaneg/virial6-resciencec.zip)

Scientific domain: Chemical Physics
Programming language: Fortran
Suggested editor:

[khinsen]

@mcbaneg Thanks for your submission! Before we start reviewing, could you please check the Metadata URL?

Note that I have edited your submission slightly. Our "PDF URL" is supposed to point to the reproduction article, not the original article.

Update: I'd better say that I tried to edit your submisson. GitHub lets me do it but what it shows is still the original text. Could you please update the URLs yourself?

[mcbaneg]

Okay, I have modifed the PDF link to point to the .pdf of the draft that I attached to the posting (that is, uploaded to GitHub.)
I was unable to do the same with the metadata.yaml file because when I try to attach it I get a message "We don't support that file type, Try again with [list of file extensions it accepts, including .zip]" So, I made a .zip file out of metadata.yaml and uploaded that; I hope that works.

The original metadata.yaml is present at the site I originally listed, but either the web server doesn't automatically start a download when a link to a .yaml file is followed, or the browser doesn't recognize that as what it should do.

If there's a way to upload the metadata.yaml file to the GitHub submission, I'd be happy to do it to save the user the hassle of unzipping.

-G.

[khinsen]

Good question. I have never used the GitHub file upload feature for anything but images. Most ReScience authors create a GitHub repository for their submission, which contains all their code plus the article source code and the metadata. That makes it easy to put links to any file anywhere, and also allows reviewers and editors to propose modifications (which, as an editor, I do for the metadata in the end of the publication process). Another advantage is that GitHub repositories are automatically archived by Software Heritage, so they are safe in case of GitHub shutting down, being hackes, changing their rules, etc.

Anyway, the ZIP you uploaded for the metadata works well for me!

[khinsen]

@pdebuyl Would you willing to review this paper?

Note that it is a special case: the first submission to the Ten Years Challenge, written in the absence of any specific instructions from ReScience. With the consentment of the author, we plan to use this article and the review to prepare more detailed guidelines for other authors.

Also, since we expect to have many submissions to this special issue, we will try a revised reviewing procedure as well, with a single reviewer.

[pdebuyl]

Hi @khinsen I will review. I'll take notes along the way :-)

[pdebuyl]

@mcbaneg Are you familiar with git? At some point, I think that we will upload the old code and likely modifications to the old code that were needed to get it running. I can assist if necessary, by either uploading myself your contributions to github or by providing specific instructions to do so yourself.

[mcbaneg]

Pierre, I do know git, though I have not used GitHub much. I’ll set up a repository. Do you have a BLAS library handy to use, or would it help for me to put the BLAS reference sources there for you?

…

-George. George C. McBane (mcbaneg@gvsu.edu<mailto:mcbaneg@gvsu.edu>) Professor of Chemistry Assistant Dean for Research, Facilities, and Analytics College of Liberal Arts and Sciences Grand Valley State University B-4-229 Mackinac Hall, (616) 331-2506 Chemistry site http://www.gvsu.edu/chem/ CLAS site http://www.gvsu.edu/clas/ individual site http://faculty.gvsu.edu/mcbaneg/ From: Pierre de Buyl <notifications@github.com> Sent: Wednesday, November 27, 2019 4:32 AM To: ReScience/submissions <submissions@noreply.github.com> Cc: George McBane <mcbaneg@gvsu.edu>; Mention <mention@noreply.github.com> Subject: Re: [ReScience/submissions] 10 year reproducibility challenge: reproduction of J. Chem. Phys. 112, 4417 (2000) (#11) @mcbaneg<https://github.com/mcbaneg> Are you familiar with git? At some point, I think that we will upload the old code and likely modifications to the old code that were needed to get it running. I can assist if necessary, by either uploading myself your contributions to github or by providing specific instructions to do so yourself. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#11?email_source=notifications&email_token=ANZTG5WSXZQ6GHSBMKYXLR3QVY477A5CNFSM4JQ32KW2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEFI4ATI#issuecomment-559005773>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANZTG5Q5QDXYAC3YKTWX4PDQVY477ANCNFSM4JQ32KWQ>.

[mcbaneg]

Pierre, There’s now a GitHub repository, https://github.com/mcbaneg/tenyears-submission, where you are listed as a collaborator. Only the master branch exists at the moment. The code, sample input and output files, and article source information are all there. Best regards, George. George C. McBane (mcbaneg@gvsu.edu<mailto:mcbaneg@gvsu.edu>) Professor of Chemistry Assistant Dean for Research, Facilities, and Analytics College of Liberal Arts and Sciences Grand Valley State University B-4-229 Mackinac Hall, (616) 331-2506 Chemistry site http://www.gvsu.edu/chem/ CLAS site http://www.gvsu.edu/clas/ individual site http://faculty.gvsu.edu/mcbaneg/ From: Pierre de Buyl <notifications@github.com> Sent: Wednesday, November 27, 2019 4:32 AM To: ReScience/submissions <submissions@noreply.github.com> Cc: George McBane <mcbaneg@gvsu.edu>; Mention <mention@noreply.github.com> Subject: Re: [ReScience/submissions] 10 year reproducibility challenge: reproduction of J. Chem. Phys. 112, 4417 (2000) (#11) @mcbaneg<https://github.com/mcbaneg> Are you familiar with git? At some point, I think that we will upload the old code and likely modifications to the old code that were needed to get it running. I can assist if necessary, by either uploading myself your contributions to github or by providing specific instructions to do so yourself. — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub<#11?email_source=notifications&email_token=ANZTG5WSXZQ6GHSBMKYXLR3QVY477A5CNFSM4JQ32KW2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEFI4ATI#issuecomment-559005773>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ANZTG5Q5QDXYAC3YKTWX4PDQVY477ANCNFSM4JQ32KWQ>.

[pdebuyl]

Thank you @mcbaneg for the update.

I will follow mostly the regular review process that you can read at https://rescience.github.io/write/ and https://rescience.github.io/edit/ , adapting of course to the specifics of the challenge.

I provide a first check list here (the first four items come from ReScience's guidelines).

• The actual replication of the research. The reviewer must evaluate the authors’ claims about a successful replication, applying the standards of the field.
• Reproducibility of the replication. The reviewers must be able to run the proposed implementation on their computer, and obtain the same results as the authors with the limits of the state of of the art of the field.
• Clarity of the code and the instructions for running it. Uncommented or obfuscated code is as bad as no code at all.
• Clarity and completeness of the accompanying article, in which the authors should clearly state why they think they have replicated the paper (same figures, same graphics, same behavior, etc.) and explain any obstacles they have encountered during the replication work.
• Past source code: availability and licensing
• Modifications needed to reproduce the results.
• Update the source code: availability and licensing.

I'll provide a first report this week so that we can move forward with the review.

Regarding BLAS, I installed it via the Debian package system and added -lblas to the gfortran link line :-)