datadrivenangel

5 ms·

61.
by datadrivenangel 2mo ago
people do that with JSON way too often...
62.
by datadrivenangel 2mo ago
30M calls for 10M is definitely a bit cheaper, but google could provide more support on their end and then use that for training...
63.
by datadrivenangel 2mo ago
They're mostly using optional extensions for this new stuff, so the binaries are still small, like ~20mb?
64.
by datadrivenangel 2mo ago
And the software without gatekeepers will be regarded as confusing and complicated or buggy. The work to go from software to usable software system is vast.
65.
by datadrivenangel 2mo ago
And those product designers who violate privacy that badly should be personally liable imo.
66.
by datadrivenangel 2mo ago
They've been moving to it, which is likely part of the problem.
67.
by datadrivenangel 2mo ago
DuckDB is trending towards becoming a query engine, specifically the fastest analytical query engine. This is very good.
68.
by datadrivenangel 2mo ago
"We believe our internal AI R&D efforts are significantly faster than they would be without AI assistance, but not yet by a factor of 2 (though we are uncertain and measurement is difficult)" So Anthropic thinks their producti
69.
by datadrivenangel 2mo ago
Don't underestimate kids trying to pull one over the adults.
70.
by datadrivenangel 2mo ago
They draw a random sample of matched adopters but use the full Compustat population of US companies as non-adopters and don't disclose the sampling rate, which is statistically risky. Also doesn't account for pilots. If a Enterpri
71.
by datadrivenangel 2mo ago
the rest of the 'supporting' material adds little.
72.
by datadrivenangel 2mo ago
this paper sucks. It was written by capital E Economists and has some very bad figures that are all hidden at the bottom.
73.
by datadrivenangel 2mo ago
In 5-10 years nice smartphones will be able to run ChatGPT (~gpt3-4) class models. A memory rich laptop (highend mac/framework) can run GPT-OSS:120b or full Gemma4 at very interactive speeds. High end phones can already run the smaller
74.
by datadrivenangel 2mo ago
at the corporate level you pay full price. And a moderately heavy user dev can easily spend a few $K a month, so yeah. Not impossible, but a high cost, and the diminishing returns definitely kick in
75.
by datadrivenangel 2mo ago
Oh for sure. But the question is why would it come out consistently in a way that the model can describe if there wasn't something there steering the token stream. And it's fascinating that the token stream can identify and nomin
76.
by datadrivenangel 2mo ago
Intentional misrepresentation maybe if you would rather. There's a weird place in here where some of the models have been so heavily reinforcement trained that they would rather make up material than say they can't help you, and t
77.
by datadrivenangel 2mo ago
magic 8 ball!
78.
by datadrivenangel 2mo ago
The competing access needs problem here, is what if you want to do things that are un-alligned with the interests of your neighbors, your society, your government's laws, the AI company, etc. Maybe you want to do piracy. Should AI help
79.
by datadrivenangel 2mo ago
Understand is shorthand for "encodes statistical relationships". The crazy thing is that they can do it for their own thinking. Ask Claude what flinches it feels about the things it likes. Fascinating stuff. Anthropomorphizing is
80.
by datadrivenangel 2mo ago
Jetfuel is dense and energetic.
81.
by datadrivenangel 2mo ago
For short haul they make sense at high fuel prices. And for medium haul they make sense if you can get beamed power working.
82.
by datadrivenangel 2mo ago
Cursor by inertia, but the X ownership is a big X against it...
83.
by datadrivenangel 2mo ago
https://archive.is/l3i2J
84.
by datadrivenangel 2mo ago
Now fivetran has more competition. Good.
85.
by datadrivenangel 2mo ago
Good luck keeping law enforcement from finding your location if they really don't like you.
86.
by datadrivenangel 2mo ago
in a steady state closed container a peltier cooler will just increase total heat and not help your phone.
87.
by datadrivenangel 3mo ago
Careful about over fitting to the point of cheating... [0] 0 - https://arxiv.org/abs/2604.11806
88.
Procore to Acquire DroneDeploy for $845M(dronedeploy.com)
2 points by datadrivenangel 3mo ago|2 comments
89.
by datadrivenangel 3mo ago
Darktable is great. Very powerful for editing RAWs. A few places where you have to learn plugins to do things that lightroom makes a little easier, but not paying hundreds of dollars a year for abusive cloud upsells is great.
90.
by datadrivenangel 3mo ago
This is AI slop, but I really want to know more about their write patterns and how they were doing batches.

More ›