Paper: PyTeCK: a Python-based automatic testing package for chemical kinetic models

[kyleniemeyer]

Still (very much) finishing this paper, but wanted to get current progress submitted. Will have final draft committed tonight 👍

[kyleniemeyer]

Alright, this is the actual complete draft... apologies for getting it in late.

[kyleniemeyer]

Hmm, noticed that on the PDF site it seems like my paper isn't building... strange, it worked fine on my local machine

[sbenthall]

Noted. @sarostru any idea what may be causing the build failure?

[kyleniemeyer]

@bryanwweber suggested the build error might be due to the accent in "Zsély" in the LaTeX log file, although I don't know why that particular name (and not the others with accents) would cause a problem.

[sarostru]

I don't know exactly, I'll be able to look into it tonight. Sounds like maybe a missing font package or something.

[katyhuff]

@kyleniemeyer Yeah... that's a bit weird. Your PR builds fine on my laptop. I'll review from that pdf. Maybe docutils needs to updated on the server?

[kyleniemeyer]

Thanks! I've got some fixes I can push tomorrow, if you haven't started your review.

On Jun 24, 2016, at 1:07 AM, Katy Huff notifications@github.com wrote:

@kyleniemeyer Yeah... that's a bit weird. Your PR builds fine on my laptop. I'll review from that pdf. Maybe docutils needs to updated on the server?

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub, or mute the thread.

[katyhuff]

Sounds good. I can certainly wait to read it on a plane that will take off in about 28 hours. So, I'll git pull before I take off and review anything that's in the repo at that point!

[bryanwweber]

@katyhuff @sarostru I was unable to build it on my MacBook Pro with MacTeX 2016 and everything-updated conda Python. The error is here: https://github.com/scipy-conference/scipy_proceedings/blob/2016/publisher/build_paper.py#L146 when Python tries to decode the TeX log file, it fails, and if you look at the log file is has \xe9 (or something like that) in a line describing an overfull hbox warning in the references, without the finishing byte. I'm not sure why TeX (I assume its TeX's fault since its in the log file) is not properly encoding the é that @kyleniemeyer mentioned.

[kyleniemeyer]

OK, @katyhuff I'm more comfortable with the state of the paper now 😁.

Also def not sure why it wouldn't build, still no problems locally.

[sarostru]

@bryanwweber thanks for investigating that was as far as I got as well, I wasn't able to resolve the problem beyond tracking it to a Tex problem.

At this point in the process we could replace the é with an e to get it building and then resolve the problem later.

[katyhuff]

The options supported for ignition-type are not clear to me. That is, would "peak detection" be an allowable option? Perhaps I'm misunderstanding. Or, perhaps a little extra explanation of this model parameter is warranted.

[kyleniemeyer]

This is discussed in a bit more detail on page 5; is that explanation more clear? I have added "discussed in more detail later" here.

[kyleniemeyer]

Thanks @eteq and @katyhuff (and again to @thewtex)! I'll be responding to line comments that don't require any discussion with 🎉 to indicate that I've addressed them.

[eteq]

Looking good, @kyleniemeyer - thanks!

Did you give some thought to using Zenodo or otherwise pointing to a specific (reproducible by the reader) version of the code?

[kyleniemeyer]

Thanks @eteq! At this point I believe I've responded to all of your line comments. As for the things you mentioned in your longer review:

• I added the spelling out of PyTeCK with the first mention of the name in text via cca6c9f
• I do plan to submit to Zenodo for a DOI—I've mainly been waiting to get to the point of an official release, that will correspond to the final version of this paper. (I'm also one of the people helping coordinate efforts to standardize software citation practices so I'm fairly aware of the importance of citing a specific version 😛.)
• I also plan to submit to PyPI once the pre-release version is DOI'ed. For now, I've added some install instructions to the PyTeCK README, for using either python setup.py install or pip install .
• I absolutely plan to write some Sphinx-based docs, and host them on Read the Docs. I hope to get this completed soon after this paper is finished.

I'm pretty sure I'm happy with the software for a pre-release version, so I should get a DOI and submitted to PyPI (with associated paper updates) shortly.

[kyleniemeyer]

Hi @katyhuff—I believe I responded to your line comments, most of which weren't too tricky to resolve (I think). I do still plan on adding a short example at the end.

Did my changes (or responses) clarify the confusion about the ignition type and where molar production rates come from?

[eteq]

Awesome, thanks @kyleniemeyer (both for those things and the software citation article ;) I'm fully satisfied by this, so this looks good to go to me!

[katyhuff]

Thanks! Looks great to me as well.

[kyleniemeyer]

Thank you! The only things left before I consider it "done" are to add the appropriate citation with DOI for v0.1 of PyTeCK, and I also want to add a nice usage example 😄.

[sarostru]

Maybe I don't need to post here, but for consistency :)

Due to the complication of the proceedings deadline lining up with everyone's travel schedule to attend the conference we've decided to allow authors to spend an additional 2 days, that is until July 12th to complete any major changes to the submission. This may impact your enjoyment of the conference, so we hope not too many of you need to take the time. For reviewers, we ask that you read over this final document before the conference is over and post your decision here in the comments on or before July 17th.

Thanks and have a great Scipy!
Your proceedings co-chairs Scott Rostrup (@sarostru) and Sebastian Benthall (@sbenthall)

[kyleniemeyer]

Still working on the usage example—I'm trying to include something real rather than a toy problem. Should be ready tomorrow or Monday!

[kyleniemeyer]

OK, I've pushed what I believe is the final draft of the paper. I've added a new usage example section (with figure!), as well as a DOI for version 0.1.0 of the software.

[katyhuff]

🎉

[thewtex]

👍 Well done!