xpe

27 ms·

331.
by xpe 11mo ago
I think the implied thought (?) is there is a similarity between central planning and oligopoly bandwagoning. To my eye, the causes and dynamics are different enough to warrant bucketing them separately.
332.
by xpe 11mo ago
I was both hoping and fearing that someone would correct my simplification. (That is more nested logs than I would’ve guessed however.)
333.
by xpe 11mo ago
> [PHP] is general purpose programming language. To be charitable, yes — PHP has access to low-level system details like the file system, sockets, and processes. > I know a person who wrote Linux X Desktop Environment using PHP. Worke
334.
by xpe 11mo ago
I am not a layer (IANAL), but here is what Gemini 3 Pro says: "You generally do not need to use the trademark symbol for CUDA in a blog post, unless you have a specific commercial relationship with NVIDIA." Now direct from actual
335.
by xpe 11mo ago
The obvious and naive method described above is O(sqrt(N)). For N ~= 2 ^ 127, that is about 2 ^ 64. / The Lucas-Lehmer method described in the article is better (how much better is an exercise for the reader).
336.
by xpe 11mo ago
Figuring out detailed requirements requires a lot of contact with reality. Specific details about not only the technical surface area but also the organizational and financial constraints. An AI model with the appropriate context would prob
337.
by xpe 11mo ago
> It's not a problem if they're wrong 30% of the time if they're able to gauge their own confidence like a human would. This is a case where I would not use human performance as the standard to beat. Training people to be
338.
by xpe 11mo ago
This makes me wonder if there are cases where one would want the LLM to generate a syntactically invalid response (which could be identified as such) rather than guarantee syntactic validity at the potential cost of semantic accuracy.
339.
by xpe 11mo ago
Yeah, people are different, but a lot of this difference results from various constraints, such as cultural practices around collaboration or technological options. Many of these limitations probably shouldn't be locked in by our tools
340.
by xpe 11mo ago
> Code by mass live committee. Yikes. Let's lean into the chaos and see what it might give us. Imagine a production application deployed directly from a non-version-controlled directory. Anyone on the team can edit the files, at any
341.
by xpe 11mo ago
> but it's notable that shipping next-gen editor software takes a lot more developer effort now than it did in the 2000s. Yes, the scope increase is vast, due to more languages, more tooling, more features, higher expectations, and
342.
by xpe 11mo ago
> Both OpenAI and NYT are bad. Both -1 and -1,000,000 are negative numbers. We need to be careful and mindful of our framing. Saying "X is bad" is a drastic oversimplification and not necessarily useful. Pointing at any one com
343.
by xpe 11mo ago
And webring buttons at the bottom!
344.
by xpe 1y ago
It is notable (but perhaps not surprising) that this is mostly about the people and the work itself. The writing is silent on the downstream impacts on the world. In contrast, there are fields (global development, medicine, etc.) where peop
345.
by xpe 1y ago
Nothing like piling transactions, analytics, and logs to the same database. /s
346.
by xpe 1y ago
I never disagreed with the utility and importance of understanding backprop. I'm glad the article exists. And it could be easily improved -- and all of us can gain [1] by acknowledging this rather than circling the wagons [2], so to sp
347.
by xpe 1y ago
Yes. No need to be apologetic or timid about it — it’s not a nit to push back against a flawed conceptual framing. I respect Karpathy’s contributions to the field, but often I find his writing and speaking to be more than imprecise — it is
348.
by xpe 1y ago
Karpathy is butchering the metaphor. There is no abstraction here. Backprop is an algorithm. Automatic differentiation is a technique. Neither promises to hide anything. I agree that understanding them is useful, but they are not abstractio
349.
by xpe 1y ago
Yep. Also, I don’t find the metaphorical connection to leaky abstractions useful at all. It feels strained.
350.
by xpe 1y ago
Hard to say if the above comment is serious or sarcastic. To my eye, seeing "Agentic Postgres" at the top of the page, in yellow, is not persuasive; it comes across as bandwagony. (About me: I try to be open but critical about new
351.
Apache Fory Rust: Serialization Framework(fory.apache.org)
3 points by xpe 1y ago|1 comments
352.
by xpe 1y ago
I'll start with a direct response, because otherwise I suspect my answer may come across as too ... complex. > How can I trust a therapist that has a financial incentive to keep me seeing them? The direct response: I hope the commen
353.
by xpe 1y ago
Here's a summary of XDG_STATE_HOME from [1]: The $XDG_STATE_HOME contains state data that should persist between (application) restarts, but that is not important or portable enough to the user that it should be stored in
354.
by xpe 1y ago
I'm sharing the article out of pain as I try to add only the relevant files from ~/.config into version control. Many programs partially implement the XDG Base Directory spec by using ~/.config. But then some also dump data
355.
XDG Base Directory Specification(alchemists.io)
5 points by xpe 1y ago|2 comments
356.
by xpe 1y ago
After carefully weighing, cataloging, chanting, and valuation, the custom is to skip them into river of their birth. (This is why river rocks are so smooth.)
357.
by xpe 1y ago
Looking over the comment chain as a whole, I still have some questions. Is it fair to say this is your main point?... > Also, Claude doesn’t “think” anything, I wish they’d stop with the anthropomorphizations. Parsing they above leads
358.
by xpe 1y ago
> ... a decent fraction of humanity ... can't easily distinguish between "Claude told me how it got its answer" and "the statistical model made up some text that looks like reasons but have nothing to do with what the
359.
by xpe 1y ago
Unrelated to wasp eradication, but interesting, still: Three Valuable Peptides from Bee and Wasp Venoms for Therapeutic and Biotechnological Use: Melittin, Apamin and Mastoparan [1] BBC: Wasp venom 'a weapon against cancer' (2015)
360.
by xpe 1y ago
What are we debating? Does anyone know? One claim seems to be “people should cease using any anthropocentric language when describing LLMs”? Most of the other claims seem either uncontested or a matter of one’s preferred definitions. My poi

More ›