(webdev Tory) @worldsendless@qoto.org

I use a massive org file to tangle my straigt.el emacs configuration. I just spent hours cleaning up a rampant `downcase-dwim`which was most certainly NOT `dwim`. I'm not sure when it happened, but somehow it got by my git commit screening and broke EVERYTHING. #emacs #LetterCasing

Why does my #emacs crash every couple hours of work? I can be doing whatever, even writing non-code. I can be in Gnome or in EXWM. Today it's occurred after about 2 hours of work, and then again 2.5 hours later. It doesn't seem to matter my window-load or my CPU load; I just hear my fans start whirring, my CPU usage goes way up, and I either freeze or even my cursor becomes sluggish. This didn't happen very often a month ago, back on emacs 28.2. What could be causing it now? Maybe #GarbageCollection?

One of the issues with the #REPL (and #emacs which is just a great big text-oriented repl) is that it is additive in nature; it usually takes major effort or a restart to REMOVE things once they've been added (thinking on plugins which modify app state). #Clojure

My life is so much better after I removed #emacs #UndoTree. Things that I just wrote off as failing before, and as freezing my process, just WORK now. For example, it turns out that "elfeed update" causes big buffer changes and that undotree was freezing the thread trying to track those changes. Same story with Telega startup. I had really suspected that my HD was failing; it didn't occur to me that I was getting sabotaged by a global buffer-monitoring plugin.

#emacs I was experiencing freezes of sometimes up to 30 seconds about 3/4 of the the time when I used my muscle memory to hit "undo", which I do as part of my regular "kill-line undo" combo. Something must have changed recently because I started to have show-stopping freezes of my emacs thread when I tried a routine "undo." I finally managed to squeeze a `toggle-debug-on-quit` and, with a bit of patience, got a C-g in during the freeze. The culprit in the resulting stack trace was my global undo-tree mode, which in combination with my other settings must have started failing. I turned it off by removing the line in my init.el and also running `global-undo-tree-mode` to toggle the mode and… far, no more funny freezes on undo. There are occasions when the undo tree is great and useful, but not at the cost of sometimes completely being a show stopper.

The fact that #emacs out-of-the-box includes "undo-in-region" is not only super cool, but also extremely useful. I just copied a bunch of stuff, changed it into org headings, updated the title, then realized that the stuff had multiple lines, so I undid just the heading part and removed the linebreaks before going back to headings -- without undoing the title change.

Today I am reminded that the difference between lazy "a la" and correct "à la" is called a "grave accent," not the pinyin 4th tone. We are doing french-english, not chinese-latin characters! #unicode In #emacs that's "LATIN SMALL LETTER A GRAVE"

No matter how much I required or load-file or check-parens, nothing explained why half a file of #emacs #elisp wasn't loading. Somehow the closing three `'")' had been stripped from two neighboring lines, and somehow the syntax had been balanced enough to satisfy the checkers. That was to hunt down.

In order to accomodate Github 2fa I've started using Jürgen Hötzel #emacs #totp https://github.com/juergenhoetzel/emacs-totp , which is a beta library leaning heavily on the ancient but unknown-to-me core #auth library https://www.gnu.org/software/emacs/manual/html_mono/auth.html#Overview . As if it wasn't enough to know "emacs can do that," it turns out "emacs did that decades ago."

Using #emacs bufler to group my browser windows (NOT tabs!) https://orys.us/vC

From the bygone era preceding #markdown and #StackOverflow, #emacs' message mode developed a marker for code areas within an email. With `message-mark-inserted-region`, by default in a message with a selection `C-c M-m`, you get something like this:

https://orys.us/vs