The contrapositive of "It is funny because it is true" is "It is not true because it is not funny", which I don't think has ever been taken seriously as a foundational approach in the study of logic.
Remote
pkal
@pkal@social.sdfeu.org
mastodon 4.7.3I happen to do use computers, that is what I mostly write about here.
What others have said about me:
"[Philip] has many of the attributes of a high-school bully" -- @flogth@mathstodon.xyz
277 Followers
117 Following
35 Posts
Joined October 12, 2025
Personal Website:
Emacs Work:
Languages:
en, de, hbs
Emacs 31 will be cut in one week (7th of May?): https://yhetil.org/emacs-devel/87zf2k5m0c.fsf@zephyr.silentflame.com/T/#u
If anyone has any outstanding patches, feature requests or non-trivial bugs, please ping the bug reports!
(I am planning to prepare a video demonstrating package.el-related features some time soon afterwards, to motivate people to try the non-negligable number of features we have added since Emacs 30 and submit bugs if encountered.)
For anyone interested in publishing an Emacs package on GNU or NonGNU ELPA, we just added a page on how to do that and what to keep in mind: https://elpa.gnu.org/contributing.html
If anything is not clear, please reach out! My hope is that this document can help clarify some of the misconceptions around ELPA and demystify the process.
Replying to
@chris_evelyn @mjg59 Isn't https://ethz.ch/en/news-and-events/eth-news/news/2025/09/press-release-apertus-a-fully-open-transparent-multilingual-language-model.html supposed to be something like that?
I guess I missed it, but it is kind of funny that the Epstein files include a version of the Bash manual from 2005: https://www.justice.gov/epstein/files/DataSet%209/EFTA00315849.pdf
Replying to
@torvalds@social.kernel.org Doesn't this mean that you don't get to exercise freedom 1: "The freedom to study how the program works, and change it so it does your computing as you wish (freedom 1)."? If you only can read the software but have no means of putting any changes into practice, then it doesn't seem unreasonable to complain about it from a software freedom perspective. Or am I missing something in the discussion?
Replying to
@janneke@todon.nl @jameshowell@fediscience.org @bbatsov@hachyderm.io @cwebber@social.coop In Emacs 31 there will be a presets theme with a lot of options that should make Emacs easier to use. As Emacs 31 still hasn't been feature frozen, any suggestions for new options are still welcome.
Just a reminder for people who blog on Emacs, there's been a blog carnival going on for the past few months, and this month I've been hosting the theme of "Mistakes and Misconceptions", in case anyone wants to contribute during this last week:
https://sdf.org/~pkal/blog/emacs/mistakes.html
FWIW I think that shorter posts on the Fediverse are also fine, as long as they are accessible without an account. Find out more here (also note, nobody signed up to host April yet!):
I just randomly remembered when in 9th or 10th grade a classmate wrote down a random string of 1s and 0s and wanted me to tell me what it means. I remain just as confused about that now as I was back then.
Replying to
@chris_evelyn @mjg59 I haven't taken a proper look at it either, so I don't know if it is open-washing as has been the case with a lot of other models, but if this means anything RMS has stated that it appears to "be free".
@kattascha Aus neugierde, läuft noch wer selbst hin, bestellt an der Theke und holt es selbst nach Hause?
Replying to
This is the cardinal sin of terminal interfaces: Every application is responsible for ensuring compatibility with terminal emulators, adding decoration and key/command handling.
The only reason there is any consistency with things like command line flags, is that getopt has made it easy, early on to do the right thing easily. And even that doesn't end up working all the time.
Replying to
@hell (cont.) the IRC channel has some wierd/annoying people IMO. One just has to know that people use it for general chat, which anyone is free to interrupt with a general question. You can join in from Emacs OOTB by invoking M-x irc.
I'd strongly advise against Reddit, not only because of the site as such, but also because they parrot a lot of cargoculted and outdated information that a lot of the users appear to have heard from second hand experiences or from people spamming self-promotion
Yesterday I heard that (non-technical) people have started abbreviating "ChatGPT" (meaning LLMs) to just "Chat".
Replying to
@paniash@mstdn.social That is probably the igc feature branch, see https://yhetil.org/emacs-devel/87v7d8ix3n.fsf@melete.silentflame.com/
Replying to
@cadadr @hywan Yeah, I think you can add a lot of branches to the lineage I sketched above.
Zed is made by the same people who worked on Atom, Gram forks Zed and removed the AI integration. VSCodium strips out telemetry out of VSCode. Cursor forked VSCode, adding AI and probably more telemetry. JetBrains had fleet that was supposed to displace Fleet, but that didn't work out.
Replying to
@tusharhero@mathstodon.xyz A Eli said, it depends on whether or not the feature is "safe and/or mature". I tried to read up on the thread a few weeks back, but IIUC there were still outstanding disagreements?
Replying to
@pjacock@fediscience.org @cks@mastodon.social INI or TOML as certainly become popular over the last decades, I guess it is intuitive to a certain point.
Replying to
@yosh@toot.yosh.is I think that anyone who looks at the state of software and software systems seriously, especially if you don't cloud your judgement by self-congratulating on being able to deal with glorified puzzling, has to admit that there is a lot of accidental complexity in how we deal with computers in general.
This also reminds me of the joke article about why Stroustrup make C++: https://harmful.cat-v.org/software/c++/I_did_it_for_you_all
Replying to
@oantolin @superketchup @RandamuMaki But yes, since GNU isn't yet certain of the legal state of LLM-created software, we don't accept patches to emacs.git or packages for elpa/gnu.git. We'd have to check if all the auto-sync'ed packages in GNU ELPA actually stand by this position.
Related, I was involved in some long standing dumb-jump issues that recently got resolved using AI agents. The code is OK I guess, but the workflow seems bizarre to a bystander: https://github.com/jacktasia/dumb-jump/pull/523
Replying to
Replying to
@sacha@social.sachachua.com This is great, from a first reading I don't have any real suggestions. If anything, I think it would be worthwhile to add your comment to the actual upstream file!
Re fido-vertical-mode, we have had comment relating to that, but the consensus was that this kind of input can still be confusing in certain edge-cases where you have to use M-j to input a partial input instead of a completion candidate. Perhaps icomplete-vertical-mode could be an option, but we weren't sure about that.
Replying to
@pjacock@fediscience.org @cks@mastodon.social Are INI files really "standard" in the UNIX world? I was under the impression that that was a convention from DOS. I am not aware of a really authoritative format (at least as conventional as command line arguments starting wiyh a dash), perhaps because for a long time many systems just used shell scripts to slurp out environmental variables out of a .foorc file?
Replying to
@levitte@mastodon.nu @bagder@mastodon.social As a user, I am always annoyed that the generated Makefile depends on Cmake. This bit me when I tried to prepare something on one system, and then build it on another, where I didn't have cmake installed.
Replying to
@oantolin @superketchup @RandamuMaki We have LLM clients on ELPA as well?
https://elpa.gnu.org/packages/ellama.html
https://elpa.nongnu.org/nongnu/gptel.html
Replying to
Replying to
@cesarb@mastodon.social @pjacock@fediscience.org @cks@mastodon.social But that emerged during the 1990s, if I am not mistaken? I was thinking more about pre-diaspora unix, so 1970s and early 1980s.
Replying to
@williampietri@sfba.social Are there any ideas how to practically legilate something like this, ideally on a global scale, considering that those opposing it will have plenty of resources to do so? IIUC this would be kind of a reversal of the NSFNET decommisioning from 1995, so perhaps the vector of attack here could be DNS again?
Replying to
@fabio@manganiello.eu By coincidence I stumbled upon this app on Fdroid that might be useful as well: https://f-droid.org/packages/nz.eloque.foss_wallet
Replying to
@zardoz03@mastodon.online Have you sent the patch to the bug tracker? If not, you'll have to wait at least until Emacs 32.
Replying to
@wolf480pl @leeloo Which is where the "motorised vehicle with wheels" analogy seems to not hold up, because what is the implied subtext in that case?