gjstein

I was a physicist for a time and I learned physics via numerical simulation: I would find problems I could solve by hand and code them up---solving integrals, derivatives, systems of equations all numerically and comparing the results. Only a handful of physics problems have closed-form solutions, and being able to turn an interesting problem into code and "play around with it" was enormous fun for me and helped me build intuition as well. This advice strongly depends on your mathematics background, but with some basic calculus you can already start playing around.

I have yet to find clear guidance on how to handle perishable food that belongs in my refrigerator or freezer and cannot be left out on its own for a day or two if I lack the cleaning products recommended in the article. Does anyone have any reference material on whether "isolating" such things in my fridge for a period of time might be effective?

I know "Getting Things Done" by David Allen is a book the HN crowd occasionally loves to hate, but I came across it at the right time in my life and was the impetus I needed to reorganize my life and put systems in place to ensure that I ... well ... started to really get things done. Since then, I've built up my Emacs ecosystem to support a GTD-derived workflow and I've never looked back.

Also on my list are the already-mentioned "Getting To Yes" and "Nonviolent Communication". I also really enjoyed "Good for You, Great for Me" by Lawrence Susskind, which is a slightly more real-world take on the ideals put forth in "Getting to Yes".

I also studied Physics in College and my course on Classical Mechanics was really the impetus to continue down that path for a while. Textbook was "Classical Dynamics of Particles and Systems" by Thornton and Marion.

A month ago I would have agreed with you, but since he announced his candidacy for the American Presidency I'd argue it's become less clear cut.

Hey Noam, this is some great work; I'll need to sit down and give the paper a deeper read. Also, the visualizations on this blog post are incredible.

I saw a talk on the Libratus agent a while back, and one of the most interesting takeaways was that the behavior of the bot had already started to impact the professional players, who now spontaneously bet large amounts to force other players out of a hand. Were there any behaviors your agent demonstrated that surprised you in the same way? What insights might we draw from this cooperative AI system that may have more general applicability to other planning domains?

The blog post "You Cannot Serve Two Masters: The Harms of Dual Affiliation"[1] from prominent AI researchers Ben Recht, David A. Forsyth, and Alexei Efros also highlights the challenges of co-employment by big tech companies. Well worth a read.

[1] http://www.argmin.net/2018/08/09/co-employment/

Easy enough to support this article; RSS is good for the open web and relatively trivial to support: took me maybe an hour for my personal blog, compared to the enormous effort I've put into CSS and formatting for the website proper. I'd love to get a feeling for how many folks here have a blog with RSS support and what the relative level of traffic is between RSS and other forms of traffic. Does anyone have statistics they'd care to share?

This may be application-specific, but I might worry about security if my configuration files support arbitrary closures.

Do you have a source for "overhead LIDAR flights"? I was under the impression that data was coming almost exclusively from their StreetView cars (with perhaps a few companies that provide helicopter-based data from time to time). True enough though that Google has a staggering amount of data about the earth.

The author does link this video in the article; really a cool talk!

I was a paralegal a number of years back and, due to my experience as a coder, I was able to automate a bunch of my responsibilities. Much of the job of a paralegal (and some of the responsibilities of a lawyer) are things like data entry and records keeping that are trivially automated. One of my most popular tasks was to take data from Doc X in format A and move it to Doc Y in format B. Some string parsing (with human oversight) massively accelerated my productivity.

I wrote a short blurb about my experience a little while ago: http://www.cachestocaches.com/2016/2/how-we-teach-programmin...

From the article:

> Even though List is very fast across the board no data-structure can be the fastest at everything.

This feels like a case in which everything is sufficiently well-implemented that the No Free Lunch Theorem [1] starts to play a role.

[1] https://en.wikipedia.org/wiki/No_free_lunch_theorem

The project pages are notably lacking images or interesting examples. Here is the most recent paper showing what they use this system for: [1] Toward VQA Models That Can Read [VQA == Visual Question Answering]

[1] https://arxiv.org/pdf/1904.08920.pdf

Agreed, I fail to understand the popularity of this aesthetic, but it is rather common on 'landing pages' for potentially interesting coding projects. It usually just makes the bottom half of my screen unused real estate.

Programming Languages

Love the Tufte style. One real limitation, however, is the "mobile view" in which the margin notes collapse into clickable superscript numbers. I don't think it's clear to most readers that one is supposed to click on those to expand them, meaning they aren't likely read by perhaps 50% of your readers. One thing I've seen, and adapted for my own site [1], is inserting these margin notes beneath the paragraph in mobile mode; this way, the notes do not go unread.

[1] http://cachestocaches.com

I will admit that, after a few years of writing papers on GitHub with co-authors, that I do occasionally miss the MS Word track changes. Not how I like composing my papers overall, but I don't think I'll ever be happy using Git to properly version prose.

Your last comment reminded me of this article [1] on "Google Maps's Moat", which discusses the vast resources that Google has poured into collecting data at a global scale to make Google Maps what it is.

[1] https://www.justinobeirne.com/google-maps-moat/

For those who don't know, this article is by Sebastian Ruder, who has a great blog [1]. His posts are often pretty deep dives of a particular area of interest within Natural Language Processing. A favorite of mine from last year gave a summary of the state-of-the-art in "Word Embeddings" [2], which I would recommend for anyone interested in the field.

[1] http://ruder.io/#open

[2] http://ruder.io/word-embeddings-2017/

Perhaps not quite what the OP is looking for, but I use podcasts mostly for music consumption. My favorite (for quite a while now) is The Anjunadeep Edition [1], which delivers a 1 hour continuously-mixed "deep" (often programming-friendly) electronic music every week. I'd be curious to hear what other music podcasts HN recommends.

[1] Here's a recent episode I enjoyed: https://www.youtube.com/watch?v=Chn4SZWnJbc

> “We need”, Jure said emphatically, “to step up and come up with the means to evaluate – vet – algorithms in unbiased ways. We need to be able to interpret and explain their decisions. We don’t want an optimal algorithm. We want one simple enough that an expert can look at it and say nothing crazy is happening here.

As someone who works with machine learning, I find this statement rather misleading. The researcher quoted here has a very strong bias about what it means for an algorithm to be "optimal". A system cannot both "be optimal" and also "biased" in a way that the designers don't like: such a system is indeed not optimal. The dialog in the machine learning community has increasingly been about how we might structure these systems in a way that they are unbiased; it's a shame the researcher (and the author of this article) seem to think that "simplicity" is the only option.

(In case anyone is interested, I recent wrote a blog post to this effect: http://www.cachestocaches.com/2018/7/bias-and-ai/ )

Some of the most thought-provoking work I came across when I was studying optics (laser physics) is referenced in this paper on "Optical Rogue Waves": https://arxiv.org/pdf/1410.3071.pdf

Rogue waves, made popular by their often ship-sinking behavior out at sea, are extremely rare and therefore difficult to study by waiting around for them to happen. Here, researchers inject noise into an otherwise stable laser system (recall that light is also a wave) that has some small nonlinear properties and rely on the high rate at which we can generate laser pules to observe rogue waves at a high enough volume to study their properties.

I suppose I only use Bash infrequently, but when I do it is normally in an effort to make scripts that I know will run on my various machines for data processing and the like. I am hesitant to use a framework like this (as exciting as it seems) because I fear it would mean managing more dependencies across my different development machines. What are people using Bash for that would require enough complexity that a framework like this becomes a godsend?

Edit: typos.

Ledger and the accompanying Emacs major mode are both written and maintained by Emacs maintainer John Wiegley. I use both to manage my finances. I'd love to hear about Hledger though; can someone tell me why I might switch to using this?

Love this site! Watched a couple of the videos too, and you've got some great visualizations. I know you said that you're not great at explaining what you've done with words, but having a couple of lines that express to me (especially as a non-expert) what your contribution was for each video would serve you really well: i.e. the videos look cool, but I don't know what it was that you did that was hard.

I've been pleasantly surprised to find that there are a number of music podcasts that satisfy my music needs on a weekly basis. A number of artists I like have weekly "radio" broadcasts that curate music that's caught their ear. (I listen to mostly electronic music though, so YMMV for other genres.)

Agreed. I also like to use TRAMP [1] for remote editing files in Emacs: it gets a local copy of a file you're editing so that it only requires a connection when you save/load (as opposed to constantly streaming Emacs over SSH). Not suitable for all tasks, but it means I'm rarely at the mercy of my lackluster internet connection.

[1] https://www.emacswiki.org/emacs/TrampMode

While I agree that one's Emacs setup will never be able to compete with a fully-featured IDE for any single language, I've found it impossible for me to leave Emacs because I now do everything in it.

However, with my org-mode integration:

- I have a list of all open tasks for each of my open projects, each with an associated deadline (if applicable).

- As I complete tasks, I provide a short summary of what it took to resolve the task; as such, my .org files have become a log of what I accomplished

- I also use .org to clock my time, which makes it easy to summarize a week or month at a glance, and to see if I'm spending my time where I think I should be.

- Finally, one of my favorite features of org-mode is the ability to include links to my code in each of my todo items, which means that I can immediately open files of interest (and even link to a particular version in git: http://orgmode.org/worg/org-contrib/org-git-link.html) and start editing, since I never left my editor in the first place.

To me, it's not about having all the features for a single language; what keeps me working within Emacs is the ability to have all the features I've grown accustomed to for every language, and have a system customized to my liking that I never have to leave. It's certainly not for everyone (I hope I don't come across as critical), but I'm convinced nothing will ever get me to leave.

EDIT: formatting.

Your post was a great read, and I might try doing something similar on my blog. A suggestion though: I typed in a query with an uppercase letter and it failed to find any matches. Perhaps .lower() your query if it's something you'd like to maintain? Thanks!

This. I learned to use Emacs' built-in org-mode at the same time, and they've helped me survive grad school.