summarity

8 ms·

61.
by summarity 11mo ago
In terms of GC quality, Nim comes to mind.
62.
AI identifies enzyme that can digest polyurethane with 95% yield(arstechnica.com)
2 points by summarity 11mo ago|0 comments
63.
by summarity 11mo ago
Yes, indeed. In my apps, which are mostly Nim, my pool manager ensures this always happens - along with a host of other optimizations. I often start with barebones SQLite and then later switch to LiteSync (distributed SQLite with multi-mast
64.
by summarity 11mo ago
I've always tried to avoid situations that could lead to SQLITE_BUSY. SQLITE_BUSY is an architecture smell. For standard SQLite in WAL, I usually structure an app with a read "connection" pool, and a single-entry write connec
65.
Apple will phase out Rosetta 2 in macOS 28(developer.apple.com)
318 points by summarity 1y ago|335 comments
66.
by summarity 1y ago
Yeah it’s just heat. Some e-bikes have a clutch to disengage the motor to prevent this motor breaking.
67.
by summarity 1y ago
If you want to get reliable automated fixes today, I'd encourage you to enable code scanning on your repo. It's free for open-source repos and includes Copilot Autofix (also for free). We've already seen more than 100,000 fix
68.
by summarity 1y ago
> Crazy amount at a crazy high speed That's 300GB/s slower than my old Mac Studio (M1 Ultra). Memory speeds in 2025 remain thouroughly unimpressive outside of high-end GPUs and fully integrated systems.
69.
by summarity 1y ago
Ah nice, I’ve been meaning to try TCC with Nim.
70.
by summarity 1y ago
For sure, some common ground is needed. Madrigal's is a famous example of LT that almost exclusively focuses on transfer itself, because English and Spanish are reasonably similar. LT's method goes a bit further though, hence why
71.
by summarity 1y ago
Immersion and transferring patterns you already know from other languages. Language Transfer is a good completely free resource: https://www.languagetransfer.org
72.
by summarity 1y ago
Lol I wish YT had a warning for that. In the meantime, good old Settings -> Accessibility -> Audio -> Play Stereo as Mono helps.
73.
by summarity 1y ago
I for one wonder if there’s any new hardware features. M3 introduced nested virtualisation.
74.
OpenVPN2 Security: Taming 2,500 compiler warnings with CodeQL(blog.trailofbits.com)
2 points by summarity 1y ago|0 comments
75.
by summarity 1y ago
Mozart not Kangaroos
76.
by summarity 1y ago
One thing you can do to become a tiny(!) bit more informed is take some psych classes yourself. PSYC1030 is a popular (free) one and gives you the first glimpses into social, clinical and developmental psychology. It doesn’t give you any pa
77.
by summarity 1y ago
Full text of EO: https://www.whitehouse.gov/presidential-actions/2025/09/rest...
78.
by summarity 1y ago
Does anyone remember the Person of Interest episode which saw the gang destroy a Samaritan-run startup to prevent it from usurping children’s education (via free tablets and AI powered teaching)? Anyway …
79.
Infecting Generative AI with Viruses [video](youtube.com)
2 points by summarity 1y ago|0 comments
80.
by summarity 1y ago
EVO ICL. Or even pinned ICL. Just got it recently, would recommend. LASIK/SMILE/etc were never and are not safe. At least ICL is fully reversible. Note that is the current version (EVO) I’m referring to, not the older (Visian) whi
81.
by summarity 1y ago
Sell the service via a merchant of record platform. They front the payment for you, and you’ll have wayyy fewer headaches
82.
by summarity 1y ago
Exactly, especially since some brands initially pushed back only to then recall products or fire their labs. Lies all the way to the bottom https://www.bbc.com/news/articles/c4gzl41rpdqo
83.
by summarity 1y ago
It does increase but slightly. The biggest increase comes from accidentally using copy semantics, which are Nims default. If you use ref everywhere (which makes sense for JS), Nim will not emit a bunch of deep copy helpers.
84.
by summarity 1y ago
Wrote a post about it here: https://summarity.com/nim-alpine
85.
by summarity 1y ago
This is just a variation of index time HyDE (Hypothetical Document Embedding). I used a similar strategy when building the index and search engine for findsight.ai
86.
CCFC: Core and Core-Full-Core Dual-Track Defense for LLM Jailbreak Protection(arxiv.org)
1 points by summarity 1y ago|0 comments
87.
by summarity 1y ago
Whenever I’m missing a library I’m usually just 5 mins away from successfully using a C library (or any library with a C API). In my years doing data analysis, signal processing, and just writing plain servers I’ve never once gotten stuck b
88.
by summarity 1y ago
I’m also an avid Nim developer (findsight.ai is almost pure Nim). I recently compiled some stats on Nims popularity on GitHub: https://code.tc/nim-stats-august-2025/ It’s growing steadily but I do qualitatively feel li
89.
GPT-5 doubles performance in offensive security benchmark(xbow.com)
27 points by summarity 1y ago|3 comments
90.
by summarity 1y ago
It literally is. LLM inference is almost entirely memory bound. In fact for naive inference (no batching), you can calculate the token throughput just based on the model size, context size and memory bandwidth.

More ›