Mozilla to shut down Pocket on July 8
support.mozilla.org
That fractal that's been up on my wall for 12 years
chriskw.xyz
Does Earth have two high-tide bulges on opposite sides? (2014)
physics.stackexchange.com
Show HN: Defuddle, an HTML-to-Markdown alternative to Readability
github.com
How to cheat at settlers by loading the dice (2017)
izbicki.me
Trump administration halts Harvard's ability to enroll international students
nytimes.com
Loading Pydantic models from JSON without running out of memory
pythonspeed.com
Trade Secrecy in Willy Wonka's Chocolate Factory (2009)
papers.ssrn.com
Improving performance of rav1d video decoder
ohadravid.github.io
Fast Allocations in Ruby 3.5
railsatscale.com
The Annotated Kolmogorov-Arnold Network (Kan)
alexzhang13.github.io
Launch HN: WorkDone (YC X25) – AI Audit of Medical Charts
Show HN: Hsdlib – A C Library for Vector Similarity with SIMD Acceleration
Problems in AI alignment: A scale model
muldoon.cloud
A South Korean grand master on the art of the perfect soy sauce
theguardian.com
Engineers and AI: ramblings of a small startup founder
labadal.com
Show HN: SQLite JavaScript - extend your database with JavaScript
github.com
When good pseudorandom numbers go bad
blog.djnavarro.net
Planetfall
somethingaboutmaps.wordpress.com
Practicing graphical debugging using visualizations of the Hilbert curve
akkartik.name
Research Uncovers Parthenon Spectacular Lighting Effects for Athena in Antiquity
arkeonews.net
I Built My Own Audio Player
nexo.sh
We’ll be ending web hosting for your apps on Glitch
blog.glitch.com
How do people integrate these authz services with search, with its performance and filtering requirements? It seems like you have to offload search to authz, otherwise you have to deal with pagination-challenging pre- or post-filtering, or incorporate the authz logic into search. You need JOIN semantics, ideally with a context.