Tootfinder

Opt-in global Mastodon full text search. Join the index!

No exact results. Similar results found.
@patrikja@functional.cafe
2026-04-18 17:28:39

Just submitted my latest Swedish Science Council grant proposal: "Navigating the Incomparable"!
We want to build correct-by-construction software for multi-objective optimization—helping safely navigate complex trade-offs like economic costs vs. global temperature rise (see the attached idealised Pareto front).
To do this, we're proposing three connected work packages (see diagram) moving from formal specification, to state-space reduction, and finally scalable exec…

Figure 1: Idealized Pareto front (yellow) showing the trade-off between global temperature rise and economic costs in our stylized Green Transition dilemma. This front is the set of Pareto-optimal points in the objective space (here R²).
Figure 3: Work package pipeline. WP1: Specification (OptUU: Agda / Lean / Functors) -> {- Lifting of partial orders over uncertainty functors -} WP2: Reduction (Pi-theorem & Relational thinning) -> {- Dimensionless Pi-groups H & symmetry invariants -} WP3: Execution (Scalable solvers / Surrogates) -.-> {- Interative refinement - back to the top -}
@azonenberg@ioc.exchange
2026-02-16 18:10:15

1.7x speedup on the 100baseT1 decode from doing the PAM3 demodulation (but not the descrambling) in a shader.
Now the whole baseT1 demo is running at about 8 WFM/s, with a roughly 125ms filter graph execution time (on 80ms of waveform data, so real time would be 12.5 WFM/s)
This breaks down (roughly, not exact due to parallel execution) to:
* 1.3 ms subtract diffpair legs
* 10.6 ms de-embed directional coupler
* 0.6 ms equalization
* 7.5 ms upsample
* 14…

@simon_brooke@mastodon.scot
2026-04-05 12:17:38

#Software peeps: do you still do box diagrams to work out how your structures will sit in memory?
#Lisp

A section of a design document I am working on, comprising text surrounding a box diagram. The text is: In the beginning was the Word My intention is that memory will be considered as an array of 64 bit words. Each word may be considered as 1. a cons cell: two instances of object32, each having one mark bit, three tag bits and 28 payload bits; 2. a single object64, having one mark bit, seven tag bits, and 56 payload bits. Note that, for any word, the first four bits comprise the mark and …
@jonippolito@digipres.club
2026-02-13 14:11:12

A rogue AI enrolls in your online class, then publicly shames you when you remove it. This just happened in a dev forum. Are we ready if it happens to our courses?
linkedin.com/posts/jonippolito

"You suspect one of your 'students' in an online class is actually an AI agent. You alert the other students not to interact with their synthetic classmate but the bot calls you out for human gatekeeping and posts a link to a blog post accusing you of discrimination. This precipitates a fervent conversation among your students about the rights of artificial intelligences."

A rogue AI agent infiltrates your online class. What do you do? Inspired by Marc Watkins’ provocative AI scenarios for educators, I’ve imagined a troubling situation of my own drawn from a real… | Jon Ippolito
A rogue AI agent infiltrates your online class. What do you do? Inspired by Marc Watkins’ provocative AI scenarios for educators, I’ve imagined a troubling situation of my own drawn from a real case. What would you do if a bot enrolled in your course and publicly shamed you when you tried to remove it? It sounds like science fiction, but a version of this just unfolded in an online forum for software developers. As I learned in FromThePage’s presentation on AI and crowdsourcing yesterda…

@jevinskie@mastodon.social
2026-02-10 22:06:30

Does anyone have a favorite or quirky application of BDDs/ZDDs (Binary/Zero-suppressed Decision Diagrams) that they’d like to share?
I’m tinkering with them to look into novel instruction decoding approaches (including software LUT cascades). I need a better decomposition with fewer support variables tho… =\