I didn’t believe it when I first read it but I can indeed run arbitrary code on my iPhone with Jupyterlite. This is great! So finally I can analyse data using the iPad on a plane, hopefully… if I find a way to reliably start Jupyterlite from the iPad‘s local storage, load my data from there, and have a persistent python environment.
> The paper now consists of 10 Jupyter Notebooks and about 16000 lines of code with 1200 git revisions
I‘ve been there too! And about 1 year after the paper was finally published, fresh starters in my lab couldn’t run the code anymore because of an obscure Pandas error caused by a version change. Then Python 2.7 was shelved and created more problems.
In conclusion the considerable effort I put in to create Jupyter notebooks recreating every single figure from the paper was not worth it.
I created Docker Container Images for every versioned "release" of the paper that we maintain in our Gitlab Registry (a CI builds those images automatically from `.gitlab-ci.yml`, `docker-compose.yml` using `docker:dind`), so you can pull a specific Docker Image for every version of the paper that will definitely have the correct dependencies (because the Jupyter Notebooks were tested with each specific version), including Jupyter itself.
That is awesome and should be the minimum standard, but even just getting the entire infrastructure set up is something almost all scientists are not willing to do.
In my experience, one should be happy if code is version controlled in a proper way. Too often, it isn't. In ML this might be a little different but in my field at least (electrical engineering) this is not the case at all.
Yes, I know and I was pretty stubborn with my initial goals to go through with this.. (probably will also cost my employment, but I learned a lot and it was thus still worth it).
Mainly that it‘s impossible to go back a step! Sorry if my comment sounded negative — I learnt a lot just by clicking through the steps. Very easy to follow. The steps are just the right size of complexity, the code is very clear.
Thanks!
You cant go back a step at all?
Going back a step might fail in certain cases that I dont have tests for but it should work in general.
I'll keep digging.
The thing about notes is that I have gotten really used to have them available across all my devices. When I'm in a meeting I might want to jot down or cross off an item from my phone. When I'm on a train or plane I'd use my iPad. At my desk it will be my Macbook.
Texmacs notes will be available only on devices where I can install texmacs, which excludes the iPhone and iPad.
Currently, Obsidian + cloud storage ticks all boxes for me. It is free, works across all devices. It uses Markdown so it's quite readable even unrendered, and the renderer works without requiring any kind of compilation.
Even smoother is Apple Notes, but without maths support (which I rarely need in notes though).
Famous example: Ragnar Granit studied color vision in the cat for 10 years. It turned out cats are colorblind. He carried on and later was awarded the Nobel Prize for related research.
Cats aren't really colorblind, they just have reduced color perception compared to humans, but still are sensitive to many hues. In fact that's exactly what his paper in 1943 on cat vision said: https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1748-1716....
Twitter is poison for academic debate. It is already impossible to discuss a complex matter over email—most nuance gets lost. Twitter is much worse, a) because of the character limit, b) because of the crossfire that is a wild mix of relevant and irrelevant comments, plus pure trolling.
It’s got to the point where if I find myself in cordial debate with so,embody whose thesis I disagree with (in 280 character chunks at the time) I thank them for the debate.
And that’s the thing: if you go into a conversation only to ‘right’ the other person’s views and are unwilling to revise/question yours then you’re not debating, you’re preaching… and there’s way too much of that already. Furthermore some of people are very ‘polite’ (in that they let you expound your views) but I’ve noticed many just wait for an option to lunch into their monologue. That isn’t debate either. It’s socially less jarring but in a way intellectually dishonest.
It may seep in in the corporate environment. But I cannot see it in academia.
A relevant presentation will be followed by a critical discussion. If there is no discussion or only positive comments, the presentation was is very likely irrelevant, or shallow at best.
I would consider an academic environment without critical discussion as broken.
Engaging with critique is an invaluable skill that is relevant way beyond academia. They key is to never take it personal, otherwise one has lost the debate already.
Understanding the main angle of critique on the spot is hard enough, coming up with a rebuttal even harder, especially when the critique comes from a senior researcher.
Accepting points of critique that are valid as opportunities to improve the research, or at least as limitations to the generality of the findings reported is highly valued.
I always tended to think that a PhD is awarded for the ability to accept and counter critique towards work that one has spent years of their life with.
Some journals encourage the submissions of comments to published papers. After all, a little controversy attracts traffic and gives both the critique and the critiqued paper more visibility.
But not all authors like being critiqued. This is even worse when competing interests are involved, e.g. patents, or an author is employed at a research-active company that sponsored that paper.
This is particularly bad when a paper is accompanied by press releases that dramatically oversell the findings. Unfortunately many researchers seem to read press releases and confuse their content with the actual science reported in the paper.
In that case it must be possible to critique a paper and set things right, if only to have a citable record for the next time a reviewer tries to shoot down a paper or grant based on a skewed understanding of prior research.
