cjauvin

I see this journalist (Cade Metz) everywhere these days it seems: this article, a NYT article about the Google ethical AI debate, and the article about Slate Star Codex..

His previous book (On Intelligence) is very stimulating and contains many ideas that have interesting connections with Deep Learning. I'm looking forward to read this new one.

I understand your feeling and I would tend to agree for things in other domains.. but for a programming language, I find that if:

(1) It doesn't make the language less efficient

(2) The new feature is coherent with the rest of the language

(3) It doesn't make learning the language harder

(4) It brings joy in my life (as f-strings do)

Then.. it's ok :-)

I remember reading Avogadro Corp.[1] a couple of years ago and being impressed at how its AI (NLP in particular) felt quite in touch with the real stuff. I think it was self-published but I'm not sure.

[1] https://www.amazon.com/Avogadro-Corp-Singularity-Closer-Appe...

More context about the paper: https://www.scottaaronson.com/blog/?p=5253

The article discusses it in sections related to Emacs' attitude toward modernization and progressiveness (see the citation from Ahmed Khanzada).

I am working with Heroku these days and I find that the CLI experience they offer is really great and seamless.

For an alternative take on a related topic, this is really a fantastically well-written and practical (free) book: http://craftinginterpreters.com

Some other packages I can't live without: expand-region, git-link, pass, treemacs, vterm.

I manage all my projects using a constellation of flexible (and hand-picked) tools it offers: magit for versioning, Org for documentation, note taking and time tracking, blacken, isort and prettier wrappers for my code formatting, ivy, counsel and dumb-jump for searching stuff and moving around, and I could go on and on. I understand that Emacs doesn't feel as "modern" as PyCharm or VS Code in a sense, but I don't think the "it's not an IDE" argument holds. It can be (if you want and take the time for it) a very powerful IDE, customized to exactly your needs. It can also be a lot of other things in addition. It's a difficult to classify tool, in a category of its own.

I'm using it on Focal and it's fine. Can you link to that bug? Maybe something related to hardware acceleration? I had trouble with that in the past year.

I own a dishwasher but I never use it.. I happen to share Bill Gates' opinion about the virtues of manual dishwashing: https://www.cnbc.com/2016/09/30/why-bill-gates-loves-doing-t...

Anyone would have a recommendation for a good choice of wireless earbuds (with a mic) for a Pixel phone?

I would recommend a white noise machine for that, I haven't been able to live without mine for many years now: https://www.amazon.com/Marpac-Classic-White-Noise-Machine/dp...

I too maintain an old and complex ExtJS 4.1 app and I never dared upgrading the framework. I carry a big zip around and it will be that until the end of its life or when I finally decide to rewrite it using React. For such an old framework it’s surprisingly bug free though.

A couple of years ago I was also a great fan of this paradigm where you try to convince yourself that you understand a math concept by coding/simulating it (a procedural, rather than conceptual understanding, if you will). Here for instance I studied the so-called "Secretary Problem", using the tools you mention:

http://cjauvin.blogspot.com/2012/12/find-true-love-on-dating...

From a very high level perspective, I think what is somewhat surprising here is the fact that given that the Game of Life rules are quite simple, it should follow that a powerful generalization mechanism (which is what deep neural networks are often described to be) should be able to easily derive them.

One related thing that I began to do a while ago is to always have a `<project>.org` file going along any of my projects, which is a free-form Org-Mode journal of all the things and ideas related to that project, including of course esoteric commands for certain things (which sometimes can be hard to find in my bash history). With Org-Mode it's easy to fluidly change the structure of this journal, which makes it a very powerful tool.

I'm genuinely a bit surprised by that, that was always my high-level understanding of what the essence of neural networks was (at least feedforward vanilla ones), would you care to elaborate?

Ah I'm glad to finally understand why Emacs 27 starts so slowly: I installed it as a snap (while my Emacs 26 was from a PPA), and I spent a great deal of time trying to find what was the culprit in my init.el config, to no avail.

> So embeddings are bits of binary code that are associated with the words and word-snippets that the computer ‘reads’. These embeddings never change.

I'm not entirely sure but I think this definition of embedding is wrong: first they are not binary, they are floats (as the other parameters/weights) and they do change, as the error backpropagates through them. They are simply "swappable parameters", explicitly corresponding to each word, thus it's possible to detach and reuse them for other purposes after training, which is not necessarily easy (or meaningful) with any other weight matrix in a model.

I am in Canada and it seems to be available.

I don't know much about macros and it revived my interest for this topic, thanks. I wonder if it could be used to embed SQL statements directly in Python code? (not sure it would be a good idea, but it would make for an interesting experiment maybe).

Wow thank you! So I just learned the proper name of this "puzzle", which my father also showed me a long time ago (and for which I used to fill pages and pages of trials even though I knew that it had no solution), and which I tried to render as an "applet toy", a couple of years ago on my blog (see my other comment below).

I was fascinated by a "puzzle version" of this problem as a child, and a couple of years ago I wrote some little Processing applets to explore the idea of "gamifying" it:

https://cjauvin.blogspot.com/2012/02/implicit-bridges.html

https://cjauvin.blogspot.com/2012/02/eulerian-hoax.html

I'm also using this excellent tool, along with a nice Emacs integration for it (https://github.com/NicolasPetton/pass) and I really like it for the control and flexibility it gives, especially in the context of a developer's environment.

The React tutorial is one of the best of the kind. You can go from zero to a reasonable understanding very efficiently.

Typical Emacs user experience: sees a list of packages some other user likes, picks one with a name he finds intriguing (in my case eyebrowse), checks it, finds it cool, adopts it (possibly for a long time).

I find that "Clever Hans" is a great metaphor to describe the gulf that exists between current state of the art NLP and human intelligence.

I had a K6-233 which was always on the verge of overheating (or maybe I was obsessing a bit too much about monitoring it). I envied my friend's K6-200, which was more stable it seemed.