Watch a Ph.D. thesis dissertation being written in (somewhat) realtime
theoriginaltakeaway.comForget the dissertation. Nobody really reads those (really, not even my committee).
What we want to watch in real-time is the defense. Especially the spanking tunnel at the end.
I was expecting to see the ten-year process involving teaching assistantship, employment, marriage, and children, all illumninated by the light of a candle at 2 am in a basement office next to the crib.
Whilst not as slick as the JS version, I did much the same thing with mine: https://vimeo.com/76969179 when bored one day. I couldn't find any tools to create exactly what I wanted, so knocked up https://gist.github.com/archgrove/6766129, https://github.com/archgrove/PDFProofSheet, and a bunch of shell scripts to generate a folder of images. Fed this lot into iMovie, and voila.
That's really neat! I'm not sure if this is intentional because your work is private, but I would love it if I could click on the images or SHAs to go actually see the thing.
At the moment I'm a little hesitant to make it public. Partly because there's a large chunk of unpublished research in there, but also because some of the early drafts are a little embarrassing.
Perhaps when the thesis is deposited (and hence public) I could add that in for "historical" interest.
Definitely don't make it public. I know people who have had their thesis topics scooped.
Yeah, totally understandable.
Yep, I instinctively clicked on one of the pages to see it bigger.
Related, I loved watching pg writing an essay live : https://code.stypi.com/hacks/13sentences
I wrote something similar while I was writing my diploma thesis, albeit not that visual. For me it was handy to see whether I was making any progress, so I regularly updated a CSV file with the current number of pages, lines, words, figures, code lines, and code comments. The idea was that on every day I did something I could at least look at the data and see how much (also the script automatically adjusted page, line and word count for the appendix which I didn't want to count).
Excellent! People writing important papers really should be using Git. When documents are extremely complex, previously removed portions can have pertinence later.
Seconded -- at least for anyone using LaTeX or another git-friendly format. I used git for my masters dissertation recently, and it was very useful to be able to checkout old versions and compare them with its then-current state.
This doesn't work particularly well for fields not using LaTeX as their document creation tool.
Why is that?
I would think the GP means that git works much better if you are creating a document in a non-binary format (of which LaTeX is one example.)
This is really cool stuff. Reminds me of a stripped down version of notch's live coding sessions for Ludum Dare.
Any reason you chose LaTeX as your content dev language and publishing over something a little "easier" like DocBook XML and Publican (which uses LaTeX as an intermediate build language to create PDFs)? Is it required, what you're used to, or purity of notation?
Partly it's what I'm used to, but I've yet to find anything "easier" for the large amount of mathematics I typeset. There is a lot of notation and very little graphics or difficult formatting.
I think LaTeX remains the go-to language for this kind of stuff. Certainly, every STEM academic I've worked with has used it.
There's nothing more suspicious than a math paper written in something besides TeX.
It's probably because the author needs to describe complex mathematics and DocBook is unable to describe mathematics[1]. I can't speak much about Publican however.
'content dev'. yuck.
Why am I obsessing over the arrow not quite lining up on the points of time? Until the fourth tick, then following some pattern..
While you mention that you are hesitant to release the content of the dissertation prematurely, could you instead release the build tools you used to make your "flatplans"?
It would be a huge help to those of us about to write our dissertations :-)
The code is horrible, but it's here: https://github.com/hds/repo-flatplan-animation
You'll find the JS bits in the repository for the website which is there too.
You could do it in Authorea (Git+LaTeX) and actually show us what you are writing too!
That's an excellent solution for big problem in my country (not at PhD level though) namely plagiarism. Make students to use such system, turn autocommit on and you can see their progress and all deviations (big chunks of text appearing from nowhere etc.).
once they learn about it nothing will change.