open source drama
project I contributed to and use regularly seems to implode due to the original maintainer (inactive for a long time, but holding the signing keys) suddenly deciding to be a jerk to the person who was actually maintaining the code
I’m starting to like ieee standards a bit more: they are on the ieee digital library and have a doi, ergo they are available on sci-hub
of course, the fact that they aren’t actually open access is still a travesty…
'90s: Information wants to be free! Heck yeah I would DOWNLOAD A CAR if I could.
'20s: Information must be monetized. You can buy a hash of this JPEG of a cartoon car, for a price and CO2 emissions of a real car!
why are there people distributing software as a 24GiB container image?
another similar (scientific prototype) tools is distributed as a 5GiB vm image, which is somehow less cursed…
re: logic programming, functional programming
Also, there seems to be a relatively straightforward (if a bit esoteric) way to map functional programs to logic programs. It even seems to be possible to do this with Datalog-like systems, at least for first-order functional programs, e.g., with
eval expressions in VIATRA. I guess you could do a limited form of higher-order functional programming in Datalog with defunctionalization. So you could (theoretically) get all the goodies with relational semantics and logic programming, but still write functional programs.
On the distributed / blockchain-but-done-right side, the Unison language tries to use a content-addressed store that hashes all program terms to identify them in a distributed system in an immutable way. Maybe if this could be extended to all terms at runtime it could form a basis of the system you describe. However, it is unclear how to incorporate relational / logic programming semantics. A “nondeterminism monad” (interpreting functions
A -> B as something like
A -> Set[B]) would be easiest, but of course can’t really be used with unbound variables – or you have to do something like µKanren.
@jonny One more reason to open papers from the authors’ (static html) website or straight from SciHub, even if we have an institutional subscription.
@aral @harald It still surprises me why the language that starts with a J and ends with avaScript is viewed so widely as beginner friendly. Among numerous eslint rulesets, typescript annotations, and a package manager that strictly enforces dependency declarations, I’m just starting to entertain the hope that some parts of my code might lack silly bugs…
Nevertheless, it’s what people use nowadays, so that’s what we end up using to reach a wide audience, too.
2021: scammers are copying websites from well-known news outlets and buying Facebook advertising space. Don't say victims are dumb, this is pretty convincing.
re: Messing with Matrix/Dendrite (long)
@trysdyn Isn’t presence turned on most smaller Synapse instances due to ridiculous memory usage, anyways? So no presence support doesn’t sound like that big of a deal.
I just want you guys to know that the multiple *years* I've had my music on Spotify and iTunes have still made me less money than the first two weeks of the same albums on Bandcamp.
Please support artists directly if you can.
yesterday I had to record and post-process 5 minutes of narration for a demo video. editing took over 2 hours for me, ouch…
I might have created something very cursed: https://github.com/siouan/frontend-gradle-plugin/issues/175
.cjs file into git?
gradle-wrapper.jar was bad enough, but this is on another level
[…] you would set the front panel switches to a number and run
dsw, which would then read that number of entries from the current directory, print on the terminal the name of the last one it had read, and crash itself, leaving a core dump. You could then, if you wanted, remove the file it had named by reexecuting the
This is legitimately terrifying.