Lightnews — Scholar-powered news

Reposted by andrew blinn

Matthew Kay @mjskay.com · 7d

to appear at #alt.vis 2025: THEORY IS SHAPES

sure "theory figures" are great, but @matthew.wiki, @maryamhed.bsky.social, me, and Carolina Nobre wonder: why always a 2D plane or a flowchart? why not icebergs, horseshoes, Möbius strips, or BLT sandwiches?
arxiv.org/abs/2510.01382

#ieeevis #hci

Abstract and teaser figure of the paper "Theory is Shapes" by Matthew Varona, Maryam Hedayati, Matthew Kay, and Carolina Nobre.

The teaser figure is an "iceberg" theory figure, showing four levels: the tip of the iceberg, containing a nested set diagram and a Cartesian plane. The next level down (just under the water) has a complicated flow charts and a combination flow chart and matrix. The next level down has the iceberg figure itself and a horseshoe. The final level contains a Möbius strip and a BLT sandwich.

The abstract reads:

"Theory figures" are a staple of theoretical visualization research. Common shapes such as Cartesian planes and flowcharts can be used not only to explain conceptual contributions, but to think through and refine the contribution itself. Yet, theory figures tend to be limited to a set of standard shapes, limiting the creative and expressive potential of visualization theory. In this work, we explore how the shapes used in theory figures afford different understandings and explanations of their underlying phenomena. We speculate on the value of visualizing theories using more expressive configurations, such as icebergs, horseshoes, Möbius strips, and BLT sandwiches. By reflecting on figure-making's generative role in the practice of theorizing, we conclude that theory is, in fact, shapes.

4 5 23

Reposted by andrew blinn

Alexander Obenauer @alexanderobenauer.com · 1d

This article is now public! Link in reply.

I recreated Jef Raskin's 1987 'Canon Cat' user environment and observed myself as I lived in it for a week.

Implementation details, surprises, and more in the essay.

(Why touch grass when you can read 3.2k words on user environments?)

1 4 26

Reposted by andrew blinn

Automerge @automerge.org · 1d

🎮 Like CRDTs and videogames? @inkandswitch.com is ✨hiring✨ for a project that combines @automerge.org with @godotengine.org to make the next generation of collaboration tools for game development!

More detail here: inkandswitch.com/jobs/godot-i...

(Fully remote 🌍🌎🌏 contract, late Nov to April)

Godot IDE Engineer

Help build native, visual version control for collaborative game development in Godot

inkandswitch.com

2 13 23

andrew blinn @disconcision.com · 2d

was flying back from LHR last month where guy next to me had audio on the whole time. not just reels but like a notification ding every ~20 seconds over. it's a wild world out there

1

Reposted by andrew blinn

Alexander Bandukwala @bandukwala.me · 2d

Adding a keypress editor

1 1 8

Reposted by andrew blinn

Programming for the Planet @propl.dev · 3d

That’s a wrap for PROPL25 at @icfp-conference.bsky.social! Here’s the tired but happy organising team @anil.recoil.org @dorchard.bsky.social @kcsrk.info with @neurocy.bsky.social celebrating! Stay tuned for a writeup and enjoy our proceedings in the meantime dl.acm.org/doi/proceedi...

4 8

Reposted by andrew blinn

Programming for the Planet @propl.dev · 3d

@neurocy.bsky.social defines the core problem as bad defaults in scientific code tools like Jupyter that make it hard to be FAIR without lots of work #icfpsplash25

1 2 3

andrew blinn @disconcision.com · 3d

replmode: live inline eval of selected intermediate expressions, where selection can be steered via user-entered linebreaks). the trace cursor can be used to navigate between evals inside function definitions

1 1 14

Reposted by andrew blinn

Kate @katef.bsky.social · 3d

i am sad, so i made a 4-channel passive mult. i am still sad but now i have a mult. I RECOMMEND DOING THINGS

okay listen up i am not going to explain what a mult is. it's clearly some synthesizer stuff. but i am going to describe the hell out of my photos. this is the back of my mult, half way through construction, and we got a eurorack panel with a bunch of jacks and things mounted to it, and on the back i've started soldering diodes bridging between the jacks, and various wires and things cool

here's the finished mult installed in a eurorack setup, doing its job. it's patched alongside some other modules. it has four leds, a couple are currently lit. it mostly looks like it belongs with the other modules, except the panel is unlabelled

1 5 94

Reposted by andrew blinn

𝕮 @chrisshank.com · 4d

Some notes by @akkartik.name (2025)

akkartik.name/post/2025-06...

"I'm very dissatisfied with the current state of the world in two regards:

𝐀. Computer owners can't easily understand or modify programs they run.
𝐁. Computer vendors compete to limit permissions for modifying programs.

Both 𝐀 and 𝐁 have happened through the efforts of often idealistic people, but they've brought us to a world that it seems clear is going in the wrong direction:

For 𝐀: Every tool (IDE, debugger, programming language) and process (Waterfall, TDD, Agile) arose out of dissatisfaction with some current state of the world, but the "pragmatic" ones that gained traction were the ones that vendors get to control.
For 𝐁: "Trusted computing", browser sandboxing and app stores are all "pragmatic" solutions that have centralized power in vendors and failed to keep computer owners safe."

2 8

Reposted by andrew blinn

Szymon Kaliski @szymonkaliski.com · 6d

⚫⚫⚫

4 16

Reposted by andrew blinn

Tomas Petricek @tomasp.net · 8d

I'm teaching 𝗪𝗿𝗶𝘁𝗲 𝘆𝗼𝘂𝗿 𝗼𝘄𝗻 𝘁𝗶𝗻𝘆 𝗽𝗿𝗼𝗴𝗿𝗮𝗺𝗺𝗶𝗻𝗴 𝘀𝘆𝘀𝘁𝗲𝗺(𝘀)! again. I'll be posting the videos & tasks on YouTube too.

In the first lecture, I explain what's a tiny system, why write one and show plenty of demos!

🎞️ Playlist: www.youtube.com/playlist?lis...
👉 More info: d3s.mff.cuni.cz/teaching/npr...

Write your own tiny programming system(s)! - YouTube

The goal of this course is to teach how fundamental programming language techniques, algorithms and systems work by writing their miniature versions. The cou...

www.youtube.com

17 49

Reposted by andrew blinn

Duende Software @duendesoftware.com · 7d

We're happy to start sponsoring the Spectre.Console project!

It is a #dotnet library that makes it easier to create beautiful console applications by giving you access to standard components you may find in a CLI experience.

Go check it out! duende.link/sp3ctr3

2 18 52

Reposted by andrew blinn

𝕮 @chrisshank.com · 8d

i like a nice higraph

dslab.konkuk.ac.kr/class/2012/1...

3 9

Reposted by andrew blinn

thebes @vgel.me · 8d

if you're interesting in gaining a better intuition for how llms behave at inference time, you should try logitloom🌱, the open-source tool i made for exploring token trajectory trees (aka looming) on base and instruct models! more info in thread

🌱 vgel.me/logitloom
💻 github.com/vgel/logitloom

5 25 100

andrew blinn @disconcision.com · 8d

dont we get free student access

2 3

Reposted by andrew blinn

Alexander Bandukwala @bandukwala.me · 9d

I can finally do some real work.

Hazel programming language with a parsed csv and table of pokemon

1 4 23

andrew blinn @disconcision.com · 9d

maybe. convincing fakes have been around since the dawn of photography. scale x quality makes a difference but for many we're deep in the singularity already. photoreality is relative. many seemed and seem already (perhaps always) bedazzled by the spawn of photoshop 2.0

2

andrew blinn @disconcision.com · 9d

> Video will become like other forms of testimony

is this not already the case? not convinced that widespread ai image or video generation changes much legally

1 3

andrew blinn @disconcision.com · 9d

generally my feed here is pretty pleasant but as soon as i get even slightly off manifold things take a turn

2

andrew blinn @disconcision.com · 9d

somewhat aesthetic bug

3

Reposted by andrew blinn

Hillelogram @hillelwayne.com · 10d

Any abstract data type that supports the flipping, bopping, twisting, and pulling operations

luna @imlunahey.com · 12d

so ah.. what's a linked list?

Hillelogram @hillelwayne.com · 12d

If you call yourself a "software engineer" you better be able to:

1. Flip a linked list
2. Flip it again
3. Bop it
4. Twist it
5. Twist it
6. Pull it
7. Bop it

2 3 28

Reposted by andrew blinn

Emilia (halloween name) @lazerwalker.com · 11d

So if you visualize LLMs as not "artificial consciousness" but "an artificial connection to the collective unconscious":

1. They become philosophically way more interesting
2. It's way more forgivable they suck at math
3. It becomes extremely clear they are infinitely less useful to capitalism

Emilia (halloween name) @lazerwalker.com · 12d

A thing I am starting to embody in my extremely weird healing journey, and that a lot of AI people don’t seem to get, is that consciousness is *by far* the least interesting or impressive part of human intelligence

6 14 100

Reposted by andrew blinn

𝕮 @chrisshank.com · 10d

You can just make things on the web collaborative by adding a single HTML attribute...

1 12 57

Reposted by andrew blinn

thebes @vgel.me · 11d

new blog post! why do LLMs freak out over the seahorse emoji? i put llama-3.3-70b through its paces with the logit lens to find out, and explain what the logit lens (everyone's favorite underrated interpretability tool) is in the process.

link in reply!

8 47 210