Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
mondrian
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
31.
▲
by
mondrian
1y ago
Related to this: https://x.com/karpathy/status/1835561952258723930
32.
▲
by
mondrian
1y ago
The "core reasoning" part of AI may be increasingly important to improve, and its "database of factual knowledge" aspects may be less and less important, maybe increasingly a hindrance. So more focused and specialized tr
33.
▲
by
mondrian
1y ago
Home building is interesting because I think a major blocker to monopoly-forming is the vastly heterogenous and complicated regulatory landscape, with building codes varying wildly from place to place. So you get a bunch of locally-speciali
34.
▲
by
mondrian
1y ago
It's a read/write protocol for making external data/services available to a LLM. You can write a tool/endpoint to the MCP protocol and plug it into Claude Desktop, for example. Claude Desktop has MCP support built-in and
35.
▲
by
mondrian
1y ago
Concentration is the default in an unregulated environment. Sure pure monopolies with 100% market control are rare but concentration is rampant. A handful of companies dominating tech, airlines, banks, media.
36.
▲
by
mondrian
1y ago
Because everyone worships Thiel's "competition is for losers" and dreams of being a monopoly. Monopolies being the logical outcome of a deregulated environment, for which these companies lobby.
37.
▲
by
mondrian
1y ago
It’s still unconvincing that the shift to AI is fundamentally different than the shift to compiled languages, the shift to high level languages, the shift to IDEs, etc. In each of those stages something important was presumably lost.
38.
▲
by
mondrian
1y ago
You have a card with a name, when you click it a detail popup opens and the name is in there. There's also a delete button saying "Delete <name>", which pops up a confirm dialog and the name is in there, too. When you u
39.
▲
by
mondrian
1y ago
The problem the name will have to be updated in 6 places in the UI.
40.
▲
by
mondrian
2y ago
You really think so? How about the alternative being he's fully pardoned of anything that might be brought against him and coasts to billionaire status by launching companies and having his funding rounds and paths to exits secured now
41.
▲
by
mondrian
2y ago
I mean it seems you’re dunking on a strawman yourself. Like if I said Denmark virtually denies no life saving claims, and when it happens it’s due to edge cases, you’ll insist just because it occurs it’s indistinguishable from a system wh
42.
▲
by
mondrian
2y ago
There are systems that cannot deny life saving care, and where everyone is necessarily insured. It’s facetious to compare those to a system where 30 million have zero coverage and the rest are systematically denied life saving care as a pro
43.
▲
by
mondrian
2y ago
That’s pretty interesting, thanks. It’s related to abstraction (interfaces) vs representation (implementation) in programming. To my eye, there’s no conflict that 1 \in 3 is true in one representation and not in the other. Trying to use \in
44.
▲
by
mondrian
2y ago
Tl;dr (a) program synthesis is provably hard and (b) English is too ambiguous as a specification language. Therefore we’re nowhere close. Ultimately not a convincing argument. We could use this same argument to argue that a human cannot wri
45.
▲
by
mondrian
2y ago
JD Vance is on record mentioning Yarvin’s ideas as inspirational for how republicans should run government.
46.
▲
by
mondrian
2y ago
This is a deep insight. It's a reactionary vs. establishment dynamic where the reactionaries get a free boost because they're fundamentally more provocative from a content perspective. I think it's more like "the reactio
47.
▲
by
mondrian
2y ago
The US is currently in proxy war with Russia on two fronts: Ukraine and Israel/Iran. Conceding Ukraine would help Iran, which is probably not what Israel wants. The idea that Trump would be friendly to Putin -- lift sanctions, give him
48.
▲
by
mondrian
2y ago
Curious about this statement, do you mind expanding?
49.
▲
by
mondrian
2y ago
Relatively speaking, Instagram and Twitter are far more interesting than LinkedIn. LinkedIn is a special kind of horrible.
50.
▲
by
mondrian
2y ago
There's a gradient of risk, though. Suppose someone is sitting on 5 billion dollars unrealized gains by holding the index fund VT. The idea of "risking 100% loss" in VT is ludicrous. Maybe unrealized gains tax can be formulat
51.
▲
by
mondrian
2y ago
Good points regarding test quality. One takeaway for me from this paper is that you can increase code coverage with LLMs without any human checking of LLM outputs, because it’s easy to make a fully automated checker. Pure coverage may not b
52.
▲
by
mondrian
2y ago
Recent work from Meta uses AI to automatically increase test coverage with zero human checking of AI outputs. They do this with a strong oracle for AI outputs: whether the AI-generated test compiles, runs, and hits yet-unhit lines of code i
53.
▲
by
mondrian
2y ago
I 95% agree, but "what people want" is probably not a strong indicator on the thresholds of paradigm shifts, since people don't know what's possible.
54.
▲
by
mondrian
2y ago
I fully agree, and that is my experience. I am just grappling with what looks like quite a bit of misunderstanding around aphantasia. Some people make it sound like it involves imaginary objects in the visual field, causing other people (wh
55.
▲
by
mondrian
2y ago
Yes I can relate to this. I can get deeply sucked into imagination while ignoring my actual visual field / surroundings.
56.
▲
by
mondrian
2y ago
You can imagine an object right now, and it's floating in front of your laptop, occluding this text? If so, that's wild. Edit: Oops I read "does appear" instead of "doesn't appear".
57.
▲
by
mondrian
2y ago
Yes! Apt description.
58.
▲
by
mondrian
2y ago
> How do those with aphantasia dream? Do they not see any visuals? Pretty sure that they have vivid dreams like everyone else. The aphantasia is limited to the topic of conjuring visual imaginations while awake.
59.
▲
by
mondrian
2y ago
Do you remember events visually? Does your memory work differently than imagination?
60.
▲
by
mondrian
2y ago
People who don't have aphantasia see objects in their visual field, occluding reality? That sounds like hallucination, and very unsafe unless you have a high degree of control over it. An imaginary horse pops into your visual field whi
More ›