ledauphin

10 ms·

91.
by ledauphin 3y ago
hey, thanks for putting this out there and also for iterating on it! I will start using sending this to folks as a reference now that I know it's being maintained!
92.
by ledauphin 3y ago
The insight about thinking deeply is compelling, and I believe it's partly true. We could all benefit from sitting silently and thinking more often. But I also suspect that it overvalues individual insight and undervalues the normal pr
93.
by ledauphin 3y ago
the tables agree that microseconds are supported by both standards, but for some reason the Venn Diagram only displays 6-digit fractional seconds under ISO8601. This led me to do some head-scratching until I referenced the table, which seem
94.
by ledauphin 3y ago
i, too, wish that there were an actual way to pay per view on things like this. i read a lot, but i read widely, and I have absolutely no desire to pay for the entirety of what tNYT is into (nor do i want to deal with their dark patterns ar
95.
by ledauphin 3y ago
thanks for doing this. I could see this becoming a part of my workflow long term.
96.
by ledauphin 3y ago
yeah, this struck me as strange too. the ongoing maintenance on long-lasting industrially-manufactured drones might well cost more than rebuilding a cardboard one every few weeks.
97.
by ledauphin 3y ago
in Python, dicts are actually definitionally ordered, since 3.7, so this objection doesn't carry much weight. additionally, you might be interested to know that the implementation of dicts as ordered was found to be more performant tha
98.
by ledauphin 3y ago
100%. as an organization, there's value in _limiting_ the number of tools you adopt. But you'll nearly always want to adopt a few right-sized tools rather than using the same sledgehammer for every single thing that looks remotely
99.
by ledauphin 3y ago
agreed, but there's still differentiation in syntax, and I was responding to someone asking for no differentiation.
100.
by ledauphin 3y ago
I disagree - allowing for type inference on the closures will make for a much more pleasant language, whereas requiring type annotations on top level functions makes sense for lots of reasons.
101.
by ledauphin 3y ago
there will be any number of design issues for them to solve and this will certainly be one of them. it doesn't seem insurmountable, though.
102.
by ledauphin 3y ago
this was a common problem with the older tail-mounted engines. ice would come off the wings during takeoff and go straight back into them.
103.
by ledauphin 3y ago
"Moonbit makes programming easier with its automatic memory management" isn't absolutely nothing. they apparently have their own GC and intend to use the one that comes with WASM 2.0.
104.
by ledauphin 3y ago
The most obvious performance win for free threading in Python is where you have large read-only, immortal data that is referenced by all threads. The "memory traffic" you mention is a major issue for multiprocessing approaches, an
105.
by ledauphin 3y ago
you're actually blurring some of these images, huh? :D
106.
by ledauphin 3y ago
I appreciate this note - I have no exposure to the original games, but Dragon Age: Origins was one of the best games I've ever played.
107.
by ledauphin 3y ago
I looked up some basic numbers on the death toll of World War II in the Pacific theater. It seems to be in the ballpark of 30 _million_ people that died. I think it would be fair to say that bombing a civilian city into oblivion is immoral.
108.
by ledauphin 3y ago
it's not going to stop, and those who complain about it without curiosity about why it's such a valued language feature are doomed to suffer until the end of their careers. at the same time, I'm somewhat interested to hear a
109.
by ledauphin 3y ago
how do eSIMs enable locking arbitrary devices to a network?
110.
by ledauphin 3y ago
this downplays the possibility of having competitors. if your idea either already exists or can be cloned quickly, being stuck on one platform while somebody else hoovers up the broader userbase can absolutely kill you.
111.
by ledauphin 3y ago
wow.
112.
by ledauphin 3y ago
i think this is a strong point, and it's underscored by how little anyone is willing to center the discussion on something quite fundamental: is the behavior from the individuals in question acceptable even if it were legal? My answer
113.
by ledauphin 3y ago
yes, I use obsidian and I have several vaults, and each is its own directory that is synced independently using Syncthing to whichever devices need to be able to access each vault.
114.
by ledauphin 3y ago
great to hear that it works, but it is hilariously expensive for what it does. When the price comes down to $3/month, I'll start paying for it. Until then, I'll use Syncthing for free.
115.
by ledauphin 3y ago
it'll be very good. it won't resolve conflicts for you, though - just like Dropbox or Google Drive also wouldn't. It's last-write-wins, so you either need to keep your devices connected and syncing at all times, or you n
116.
by ledauphin 3y ago
I appreciate the correction and reference!
117.
by ledauphin 3y ago
in practice, they just make you sign a notice and consent form in order to receive care. if all the local hospitals do this (and they do) then you no longer have any option but to sign the form. we really need to eliminate this absolutely l
118.
by ledauphin 3y ago
this is exactly the software that came to mind for me.
119.
by ledauphin 3y ago
a fraction of the billions in wealth that had magically appeared during covid, largely because of government policy.
120.
by ledauphin 3y ago
i want something like this for lots of reasons, but the headliner is structual typing. It should be possible to get simple attribute access while not creating a new nominal type and not going outside the Python stdlib. I'm constantly u

More ›