mixedmath

This is perhaps only tangentially related, but I used to live next to castle ruins in the UK. Although they were ruins, the grounds still belonged to the crown. There was a thriving rabbit population there, and it was forbidden to hunt on royal land without permission.

So my town had a royally approved rabbit hunter --- who was also a butcher. He was the best butcher in the town.

Unrelated, but we also had a separate fruiterer, who procured fruits. I have never seen someone called a fruiterer anywhere else.

Don't ask to ask.

https://dontasktoask.com/

I routinely work with people using screen readers and for whom plain text email is perfect, while HTML email is a minefield of unreadability. It's worse that different email providers use different, inconsistent, and variable HTML formats.

The more I work with people using accessibility tools, the more I realize that web accessibility is an afterthought (if thought about at all).

That's pleasant. At one point it was approximately one half-screen away in lag (at 125 wpm or so). But this gave an additional impression of zooming.

The top image is an excellent choice of image. Perhaps I've become too used to seeing a random visually appealing SVG at the top of articles --- but the top image here shows two laptops forming an obviously shared desktop.

How do you like working from a WeWork? Do you find that there is any community there, or is it just a nice place that isn't your home to work?

When you read "instance", think "email provider". Does it really matter if you use gmail or yahoo or protonmail or anything else? In some ways yes, but in many ways no. You can read emails from any other email provider. And if you don't like one, you can move to another.

In the opening paragraphs, this post indicates that this other post of David Vujuc [1] falls victim to common misconceptions about what a REPL is. I'm not sure what misconceptions this post is referring to; perhaps I also have these misconceptions. But I also don't think this post clarifies that. Could anyone here make it more explicit?

[1]: https://davidvujic.blogspot.com/2022/08/joyful-python-with-r...

If they got rid of outlook, then I'd have to find a new email address to give away to every signin place that wants my email and I suspect will email me promotional garbage (or sell my email).

I guess this means that I'm reducing the signal they get for what "good" email is, as approximately 100 percent of the emails I get there are promotional or spam.

This seems to hold pretty well. I've been using vim for approximately 15 years now too and my current setup and preferences are surprisingly similar to those in this post.

There is a certain organizational aspect here that I think is important to emphasize: the given paradigm is that vim is a text editor forming one part of the design environment. Assembling this environment inside a tmux session is particularly easy and powerful. And plugins can add some quality of life.

In addition, I also like the vim-slime plugin. All this does is facilitate sending lines/regions/blocks to a REPL (or in principle, any tmux pane). It's similar to the tmux send-keys function described in the OP.

A different paradigm of vim use is growing in popularity, which is the transform vim into a full IDE on its own.

I love man pages. But it took me many years before I realized how easy it is to write man pages for tools that I write (and further, I write in markdown and pandoc happily translates them to troff for me).

A few years ago I learned that one can also add a new section to man. I've added the `eg` section, which I fill with examples that I think will be useful to future me. For example, typing `man eg awk` opens up my man page of examples for awk, which is more pointed.

This website began as an online copy of Abramowitz and Stegun's book [1]. It's not hard to find high quality downloads of that book, e.g. [2].

[1]: https://en.wikipedia.org/wiki/Abramowitz_and_Stegun [2]: https://www.cs.bham.ac.uk/~aps/research/projects/as/

I wish that google would enable quote search functionality for youtube too. That is, if I put quotes around a youtube search, then it will only return results containing that in the title or description, and not show related videos or anything like that.

What does "djb" mean here?

Several years ago I built a chess engine in C, based on various ideas gleaned from the chessprogramming wiki [1]. I haven't looked at in a while, but that wiki was an extraordinarily useful resource.

I didn't write a better stockfish or anything, but I was surprised that it didn't take too long to write a chess engine that could routinely beat me at chess.

I'll also note that chess protocals are very standardized, and it was straightforward to attach the engine to one of the many chess GUIs out there (I used Chess Arena and scid). This was a great project and I highly recommend it.

[1]: https://www.chessprogramming.org/Main_Page

This is really wonderful. And I like how simple and clean the pages look, even through markdown+github. Cheers.

I write a lot of LaTeX and articles. I sort of do this, except that I also really want reasonable line lengths. Each sentence starts on its own line, but I also hard wrap at 80 characters. I have a completely different set of editing tidbits for writing equations.

I guess. I've used the same xmpp integration for almost 15 years now. I remember thinking it was going to stop working at some point in the past, but it kept on chugging. I think maybe this time it will really die.

Too bad!

I have a few requirements for an email client. My primary email is through the same company I use for webhosting (Opalstack in my case). I'd prefer if other companies didn't read my email, so I don't use gmail/outlook/similar via IMAP/POP or forwarding or anything like that.

I also only send plaintext email. This sounds like it should be easier than sending HTML, but many mobile email clients only send HTML or "rich emails". This includes the gmail! (This used to be possible in earlier versions of google's phone email apps. Maybe it was Inbox? I don't remember now. But being unable to send plaintext email from gmail on mobile is what caused me to move away from gmail).

Now I use mutt on my own computer, K9 on mobile, and if absolutely necessary Roundcube webmail. (I do not particularly like Roundcube, but I almost never actually have to use it).

Mutt is wonderful, though I found it highly nontrivial to initially use. K9 is fine. One feature of K9 I like is that it has a surprisingly capable search capability given that I use it purely over IMAP --- it dispatches a search to my mail server and parses the results.

I haven't looked at that before. But for what it's worth, I confirm that if I navigate to the web app on my phone, navigate to the HTML version, and repeatedly click to ignore gmail's suggestions to use the mobile app, then it is in fact possible to send plaintext email from my phone.

I don't think this is a suitable replacement for an email app. For example, this is an extremely inconvenient way to check email, and having to use a different process to check and read email is too much friction. But this does seem like a workaround to send a plaintext email.

I first began to use K9 when gmail's android app stopped allowing me to send plaintext email. I still find this very confusing. I routinely write emails to a developer using an ereader, for whom plaintext emails are trivially easy to work with and for whom HTML emails are a world of pain.

That's what started my migration away from google.

Having said that, K9 is fine. It's good enough that I'm not looking for a replacement.

Which podcasts were you listening to? They sound interesting to me, and I know nothing there. Would you recommend them?

Having written this, I am actually completely incapable of retrieving the page number over dbus. (As you had mentioned). I'm a bit surprised that forward and reverse search can still be done.

On recent-ish versions of evince it is possible to do a bit more via dbus. For example, forward and reverse synctex search between evince and vim over dbus is done at [1]. I use this when I use evince, but I acknowledge that it has stupid flaws.

[1]: https://github.com/peterbjorgensen/sved

As an aside (if the author happens to read this), I would happily subscribe to an RSS or atom feed for this site, but I don't see one.

This is an advertisement for Authorea (which I'd never heard of). I extract two passages that stand out to me.

> What is the single most important factor that has prevented the arXiv to quickly innovate? We believe it is LaTeX. The same technological advancement that has allowed the arXiv to flourish, is also, incredibly, its most important shortcoming. Indeed, the reliance of the arXiv on LaTeX is the source of all the weaknesses listed below.

> The research products hosted by the arXiv are PDFs. A title, abstract, and author list are provided by the authors upon submission as metadata, which is posted alongside the PDF, and is rendered in HTML to aid article discoverability.

It's interesting to me that the authors ignore that it is possible to read the source tex for most papers on the arxiv. The arxiv prefers to be given tex and source files, and then to compile and serve the pdf --- when this is done, you can read the source. In this way the arxiv is a repository of both the plain text source of the document and a formatted output.

In some of my papers, I deliberately include comments or extra data in the source for others. I'm not alone here; I've used the code embedded in this paper [1], for example.

While I think there would be some advantages if the arxiv required all papers to be compilable tex source files, I understand that the arxiv also accepts other formats to not exclude potential writers who do not know tex. [The other formats are pdfs (e.g. converted from Word) or HTML with jpg/png/gif images (which I have never seen in practice)].

[1]: https://arxiv.org/abs/1607.07827

These are very beautiful visualizations!

What do you use for debugging instead?

I'm impressed at the relatively high efficiency of the system. Pedalling at 150 watts and recovering 100 watts of electricity is higher than I would have expected from such a simple system.

I'm very curious how much better the system would perform if the flywheel were attached directly to the drive-train instead of via friction roller. In practice this might not give much higher efficiency, but I would guess that the chain and primary sprocket would wear out slower than a tire friction-running a flywheel.

(They mention that they don't do this because this because it would be harder to build. I believe that. But I'm still curious.)

What does "MIXING" mean in this context?