12_throw_away

2,349 karmajoined 4 anni fa

good luck, i'm behind 7 proxies

Submissions

Are insecure code completions in PyCharm a vulnerability?

sethmlarson.dev

47 pointsby 12_throw_awaymese scorso16 comments

User Interface Hall of Fame (1999)

hallofshame.gp.co.at

17 pointsby 12_throw_away4 mesi fa1 comments

Judge orders Krafton to re-hire Subnautica studio founders

rockpapershotgun.com

5 pointsby 12_throw_away4 mesi fa0 comments

[untitled]

1 pointsby 12_throw_away5 mesi fa0 comments

CSB: Fatal Explosions at Didion Milling [video]

4 pointsby 12_throw_away5 mesi fa3 comments

"t-strings" and `string.templatelib`: new in Python 3.14

docs.python.org

3 pointsby 12_throw_away9 mesi fa1 comments

WinRing0: 20-year-old foundational library that's insecure by design [video]

2 pointsby 12_throw_away9 mesi fa1 comments

[untitled]

1 pointsby 12_throw_away10 mesi fa0 comments

Why do some gamers and laparoscopic surgeons use inverted y-axis controls?

pmc.ncbi.nlm.nih.gov

4 pointsby 12_throw_away10 mesi fa0 comments

comments

5 giorni fadiscuss

  I want XBOX to be one of the few companies that entertains more than a billion people each day

"more than a billion"? What are we doing here? Do you have any idea what your target market is? Surely someone in your organization can provide you with a good stretch goal ... >10% of all humans using an XBox daily is not that.

Seriously, I cannot fathom why you would say this. Innumeracy? Narcissistic delusion? Stealth launch for a new industrial human cloning project?

5 giorni fadiscuss

[flagged]

5 giorni fadiscuss

IMO it's worth watching the video rather than reading someone's writeup of it. My favorite part is this (written) list of everything that went wrong [1] ... and how much of it is due to the intersection of hardware, software, vendors, and linux

[1] https://youtu.be/EYRrUiM_A6g?si=T60tAChuo-GNtfqW&t=921

9 giorni fadiscuss

Literally some of the first advertised uses for LLMs were both "You can feed it bullet points and it will compose an entire email" and "You can take long emails and condense them into bullet points!" They've been doing this since day 1.

9 giorni fadiscuss

> for people to hold different definitions of words [...] is the single biggest reason for all arguments in recent human history.

IMO this extremely, extraordinarily true. And in my experience, it's somehow even more true for disagreements among scientists. Even though every scientific field is, in some sense, about defining a shared set of extremely precise jargon. (I recall two very well-respected scientists screaming at each other about the definition of "acidity" for instance)

11 giorni fadiscuss

Corollary: If a computer makes a business decision, the person who delegated the decision to the computer must be held accountable.

Consequence: All business decisions will eventually be delegated to computers via sufficiently convoluted and untraceable processes such that no manager can ever be held accountable.

11 giorni fadiscuss

> If you don't want people to optimize against your filtering process, you have to make it somewhat nondeterministic.

I'm sorry, I'm not following this at all. When you say "better candidates are exponentially more likely to pass the filter", we're still are talking about a metric, yes? A metric that can be optimized? Why would switching from a hard cutoff to some sort of stochastic filter weighted by this metric discourage optimization?

11 giorni fadiscuss

This is, indeed, the next generation of AOP: they've managed to evolve it from "extremely complex and hard to understand runtime behavior" into "completely undefined runtime behavior". UB as a service. True innovation!

14 giorni fadiscuss

> it could be reviewers and others intentionally trying to sabotage other people's work

Anyone with even a tiny bit of knowledge about the topic being discussed would know that everything I described happens after reviewers and editors approve the manuscript. At this stage, they have all long since ended their involvement in the process.

As with most conspiracy theories, this "theory" reveals much about you, but fails to say anything at all about the topic at hand. [1]

[1] https://www.youtube.com/watch?v=DzDztyS09CU

14 giorni fadiscuss

Right? A lot of journals make a big deal about submitting your manuscript in the proper format (sometimes even LaTeX if you're lucky) and then you get the galley proofs back and half the equations and citations now have typos in them.

The entire publishing process often feels like a chain of "you had ONE job"-type errors from the journals (presumably because they're wildly underpaying and overworking the people whose one job these things should have been).

14 giorni fadiscuss

Well, maybe no competition when we're talking about institutional access rights on dry land. However, for everyone else, there's quite a bit of competition out there in the high seas.

16 giorni fadiscuss

I am unable to comprehend the state of mind that would lead one to ask this question.

16 giorni fadiscuss

As a legal theory, "this default judgement against an anonymous AliExpress seller is binding on literally everyone in the world" kinda reminds me of the Dune nft bros' "we bought a book about Dune and therefore now own the intellectual property rights to Dune."

Except this one is apparently coming from actual accredited lawyers? (Who knows, I'm not a lawyer, maybe it really does work that way and Fender is the first company to figure out how to exploit this)

18 giorni fadiscuss

Whether you're a consumer, employee, or enterprise IT, in today's data-rich world, it can be difficult to tell if your information is really being stored securely.

I'm therefore proud to announce my new, AI-powered startup to help you navigate the modern privacy landscape. It's called Nope™. If you're concerned about privacy, make sure Nope™ is the first thing you think of.

It's easy to get started! Whenever you find yourself asking, "Is this company really storing my data securely?", your next thought should be: "Nope™"

18 giorni fadiscuss

So, um, the phrase "legally forced" really does not go with "happened easily".

Ignoring what "legally" forced entails, and just focusing on the mechanics of renaming stuff: Even if your code base can be updated with a simple "s/old/new/" regex (and, IME, this never, ever works as well as you thought it would, and always creates a long tail of manual grepping and editing), you still have to fix the docs, deal with breaking API changes, support multiple versions, provide upgrade paths, deal with confused users, etc. etc. etc.

And, um, confusingly, the examples you cite are actually even more complicated and work-intensive than what is being talked about in this article, because they also require URL / domain name updates, legal document changes, regulatory filings, etc. etc.

TLDR: the contention that "name changes happen easily" is not an accurate description of the situations you are citing.

19 giorni fadiscuss

Dunno, the old school "we'll get you started in the mailroom, and you can work your way up by gaining knowledge about the organization while demonstrating competence and professionalism" sounds like a pretty solid hiring strategy.

Although tbf I kind of doubt if this was ever really the case - probably this is imagined nostalgia for idealized bygone times. And given that this is a strategy that requires, y'know, long-term investment and planning, it's not like it's going to start happening anytime in the near future

20 giorni fadiscuss

tbh I think it's probably just commented out (and is about as likely to still work as any other commented-out code)

20 giorni fadiscuss

In contrast, I don't know that much about VMs.

But if you're making a big fundamental change to a system, I do know that it shouldn't start with a single "+279,276 -4,272" PR. It starts with a small patch with the core of the change so that everyone can understand what it does and how it works. (I mean, ideally, a change like this starts with documentation, discussion, diagrams, surveys of existing implementations, etc, before you start writing code)

You don't cram everything into a single 270K line PR, even (especially) with an LLM, unless you specifically don't want anyone else to look too closely at what you did.

20 giorni fadiscuss

> liked the idea of the book and used the bulk of its text as lorem ipsum in a demo

I'm sorry, what? What exactly do you think is happening here?

21 giorni fadiscuss

> did we inadvertently train AIs on idiotspeak.

Nope! That is - training on lowest-common-denominator, low-signal high-noise "idiotspeak" was not at all inadvertent.