Irreal

Three years ago, I wrote about marking the active window (update here ) so that it’s easy to pick it out from the other, inactive windows. I did that by making the mode line of the active window a distinct color instead of the slightly different shade of gray from the other windows. That’s worked out really well for me and I’ve never looked back.

Now Ray Zeitler over at Ray on Emacs takes another look at the problem and found a really nice solution. It’s a bit more in your face than mine so if you’re looking for something that makes the active window really stand out, you should take a look at his post. The TL;DR is that instead of changing the active window, he changes the background color of all the other windows.

Zeitler found his solution on a 10 year old Stack Exchange post so it’s pretty certain to work no matter what version of Emacs you’re using. The code is very simple and you can simply copy it to your init.el if you’re interested.

As I said, this solution seems a little aggressive to me so I doubt I’ll be implementing it but others may disagree. One thing for sure, it’s going to make your active window stand out. It’s a small enough change that you can try it out—say by copying it to your scratch buffer and executing it—to see if it’s something you’d like.

Over at the MIT Technology Review there’s a nice article on Emacs turning 50 this year. It gives a short history of Emacs starting from its TECO days. Most of that history is pretty well known to seasoned Emacs users but it’s still interesting.

What the article doesn’t give is the exact day of Emacs’ 50^th birthday. That’s not surprising. The early history of Emacs involved the researchers at the MIT Artificial Intelligence Laboratory agreeing on a standard set of TECO macros in order to improve collaboration. That was undoubtedly an ongoing process that didn’t have a well defined beginning or end so it’s easy to understand why the exact birthday of Emacs is a little murky.

Regardless, we can all celebrate Emacs’ 50^th year. It’s remarkable when you think about it. Very few programs last that long. There’s FORTRAN and Lisp, of course, but I can’t think of any other piece of software with that sort of longevity. There is, to be sure, a Ship of Theseus aspect to that longevity but it’s one shared by FORTRAN and Lisp as well.

Who knows what things will look like in another 50 years but if we still have editors, I’m putting my money on Emacs still being around. Wouldn’t that be something?

Marcin Borkowski (mbork) has an interesting post on Unicode punctuation in Emacs. His post is mainly concerned with the vagaries of dealing with English and Polish punctuation in the same document but a lot of the heavy lifting is accomplished with Electric Quote Mode.

The main problem is handling single and double quotes ‘like this’ and “like this”. Because I do almost all my writing in Org mode, I use the ':t option to turn the ASCII double quote (") into their curved typeset equivalents on export. That’s supposed to do the same with single quotes but I could never get it to work for me and always had to add then manually.

Fortunately, it’s easy to make everything work right in the buffer as you type. You merely turn electric-quote-mode on and it takes care of everything for you. By default it acts the way TeX used to by requiring “ for an opening double quote and ” for the closing double quote and similarly for single quotes. You can make it accept ” and ‘ instead by setting electric-quote-replace-double and electric-quote-context-sensitive to t.

Electric Quote Mode is global so turning it on enables it for all your buffers. You can use electric-quote-local-mode if you want it only in the current buffer. You can toggle either of them on or off by simply calling Meta+x electric-quote....

If you have to deal different quote notation for two or more languages, take a look at mbork’s post for a nifty solution.

Today I learned a new Emacs trick from Emacs Dwelling. Well, new to me. It turns out it’s been in Emacs since 28.1 so for about four and a half years.

Here’s the problem. When you call isearch it starts searching at the current point location to the bottom of the file. You can cause it to wrap around and continue searching at the top of the file but this isn’t always convenient. As a result, many people always start their search by moving to the top of the buffer first. But this isn’t always convenient either. For example, if you quit the search you go back to the top of the file, not where you started.

Sometimes you want a bit more control: perhaps you want to start at the first match or the last match and if you quit, you’d like to be back where you started. Well this is Emacs so of course you can do that. The nice thing is that it’s already built in. There are no packages to install, you just enable it my setting isearch-allow-motion to t.

You can start the search at any of:

1. The current point position
2. The first match in the file
3. The last match in the file
4. The first match before the window
5. The first match after the window

Take a look at the Emacs Dwelling post to see the details and how to invoke the extra behavior. I don’t know why this hasn’t become the default. If you don’t use the extra capabilities, everything works exactly as before.