blog! “1KB JS Numbers Station”

Code Golf is the art/science of creating wonderful little demos in an artificially constrained environment. This year the js1024 competition was looking for entries with the theme of "Creepy".

I am not a serious bit-twiddler. I can't create JS shaders which produce intricate 3D worlds in a scrap of code. But I can use slightly obscure JavaScript…

Read more: https://shkspr.mobi/blog/2025/07/1kb-js-numbers-station/
⸻
#code #HTML #javascript #tts

the other side

Step 1: Write R code
Step 2: Share it
Step 3: Spend 2 hours helping someone install R and troubleshoot their environment

OR

Step 1: Write R code
Step 2: webR sharelink

https://webr.r-wasm.org/latest/?mode='editor-terminal-files'#code=eJx1klFr2zAQx7%2FKTS3YBidxwgZbWB7Gxh5GVkZLn%2BpiLo7iGOxTkE7dspJ99iHJduNuebDN3f1O%2Ft%2F%2F9PAsCFsplqLFmqa3IhUH5L1YitletXL2U24Ka6SeDWWWv1gsxRWsFW5hZ6nkWpEBpC1oS4CEzdHUJiejrC5lHA3M9DZKXtKlol1dhVxOV%2FBD6p3S7dkBW2SEjxPQErfT0jzFkcH20MjCFVzCtWppbMMO842%2FZWGwkSZ2TArhJ9daVrWiJKeDronj0JPkBCBSgZaVtiSWrK08pYMjZ7ov2jJmBm8%2BdTO8%2BJPTSJ7T29c6qUFjsasblhpWcHO%2FXifw7ETWO4jf1GZKtmniEZd0BEBvlvs%2BuFc3NKxW46NTeHQdJ%2Fdyj7Fti%2FpYGEb2upracBwOZcXYdIpXjvRar30iSQODT9VAtBLpP4gPis2xaBXxHlaAVaVlhSzj0PkHfCWFYIWxrV9O0mvUkq2meCQ1ycnPIM5W1t%2Bpi%2Fs6A4ZlffY5q9EtA4xkrqkyOQV25EhvKUQ3SvM%2B8vMZRs3uTkpXWGSLt5NsPsnmoVparSWVR1e7v%2FsS5ZS8Vv3qVl8U%2Fy%2FXzRDMC%2BJS72hO35BSLzKdv8uynL7KTR%2B%2Fd%2FF31H28cLHj75TlfbrIej7E8w8939UXWSZOj38Bg1du9Q%3D%3D&jz

Step 3: Profit

We're curating an exhibit on #code as culture for #UNESCO. Got a snippet & a story? Submit by Sept 8. https://www.softwareheritage.org/2025/07/07/code-exhibit-unesco-cfp/

Civilisation machinique et clôture algorithmique : capturer les subjectivités, neutraliser le dissensus

De #Marx à l’#IA, la #lutteDesClasses s’est déplacée. La #civilisationMachinique capture désormais les #subjectivités, à #vitesseAlgorithmique. Hors du #code, plus #rien n’existe. #technique #subjectivité #empireCognitif #accélération #mégamachine #postMarxisme #postPolis #philosophie Depuis les fondations des sociétés hydrauliques pyramidales, la civilisation…

https://homohortus31.wordpress.com/2025/07/15/civilisation-machinique-et-cloture-algorithmique-capturer-les-subjectivites-neutraliser-le-dissensus/

I did another post on use with() and within() from Base R, simply because it bears repeating. I don't think it gets enough eyeballs.

Post: https://www.spsanderson.com/steveondata/posts/2025-07-14/index.html

PRIVACY FOLKS: Stop fucking forgetting about accessibility. People with disabilities deserve privacy too - and in fact probably need it more than you do.

Signed,
So fucking tired of not being able to use privacy focused alternatives to apple and google, their native aps for various things from email to navigation, and suchlike.

Heh, you can add emoji into #GitLab labels.

I've been using Claude Code, and I like it. It's produced decent code and configuration files and everything, but I've only so far used it for "evergreen", fully vibe coded projects. So having Claude start from scratch.

Meanwhile, I *have* used Cursor on existing projects to add features, fix bugs, and add tests. And I found that to work pretty well too.

The problem I have is that with Cursor, I can see the diffs of the code in my editor, step by step, and approve or deny individual changes.

With Claude, it seems like it just prints a diff in the console and I have to accept or reject the whole thing there, with no context of the rest of my project, and no ability to tweak it.

Am I just doing something wrong? Is this the reason to stick to Cursor?

Looking for insights.

I know I'm late to the party on this, but when people say "headless CMS" do they kinda mean the database part of the "database publishing" of 20 years ago?

Like it is storing content separately from the HTML/CSS or app interface so one doesn't have to munge that stuff when making content, or munge content when making a site or app or something, right?

Or is there some new magic in "headless CMS" I'm missing?

"We ran a randomized controlled trial to see how much AI coding tools speed up experienced open-source developers. The results surprised us: Developers thought they were 20% faster with AI tools, but they were actually 19% slower when they had access to AI than when they didn't."

https://bsky.app/profile/metr.org/post/3ltn3t3amms2x

PHP people,

I'm trying to convert some python that I wrote many years ago to #PHP. The code takes a word (string) and converts it into IPA(International Phonetic Alphabet) based on an alphabet character=>IPA dictionary (array).

In Python every string is an array, but in PHP it's not.

How do I iterate over a string not just seeking for individual characters but also for groups of characters? For instance the word 'light' I'd want to seek for 'igh', which would be converted to 'aɪ'

TIA

#Programming #Python #ConLang #Code

Edit: With loads of help from lots of people, especially @inguin, I now have this sorted (I think). Thanks to everyone.

CfP: DHd2026 in Wien
„Nicht nur Text, nicht nur Daten“ – unter diesem Motto lädt die Jahrestagung der Digital Humanities im deutschsprachigen Raum vom 23.–27. Februar 2026 dazu ein, über Methoden, Daten, Software und deren Zusammenspiel in den Geisteswissenschaften zu diskutieren.

Deadline für Einreichungen: 01.08.2025
https://digitalhumanities.de/2025/06/20/call-for-papers-dhd2026/

Finally! Next #Laravel version will include a way to find a value between two table columns.

Say hello to Cavil-Qwen3-4B; the #LLM making #opensource #license detection easy & accessible!
Trained on 150k samples
Works on consumer #GPUs
From #code headers to docs, Cavil's here to help! #openSUSE #AI https://news.opensuse.org/2025/06/24/suse-refines-releases-os-llm-to-fuel-community-collaboration/

My latest blog post: Metronalmost

https://mikecoats.com/metronalmost/

This metronome, by design, will never, ever, tick exactly once per second. This infuriating object is designed to thwart the goals of @hackaday's One Hertz Challenge contest.

Got a bug report for @novelwriter from someone who uses Cuneiform text in their work. These are 4 byte Unicode symbols, and turned out to be very tricky to handle.

The app is built with Python, which will switch a string to UCS-4 when it contains such characters, so the characters always have a single index in the string.

However, the Qt library uses UTF-16. That means 4-byte characters use two slots, creating a mismatch in indices between the two representations.

The Sequencer: Detect one-dimensional sequences in complex datasets

The Sequencer reveals the main sequence in a dataset if one exists. To do so, it reorders objects within a set to produce the most elongated manifold describing their similarities which are measured in a multi-scale manner and using a collection of metrics. To be generic, it combines information from four different metrics: the Euclidean Distance, the Kullback-Leibler Divergence, the Monge-Wasserstein or Earth Mover Distance, and the Energy Distance. It considers different scales of the data by dividing each object in the input data into separate parts (chunks), and estimating pair-wise similarities between the chunks. It then aggregates the information in each of the chunks into a single estimator for each metric+scale.

https://github.com/dalya/Sequencer

http://sequencer.org/