https://noscript.net/

https://github.com/uutils/coreutils

Although I could see them benefitting.

> Rapid sending of messages

> Careful checking of the content of messages

> Ability to quickly target a specific geographic area

> Inability to mistakenly send a test message

> Requiring strong proof that the message is authentic before sending

> Resilient enough to work after significant damage to infrastructure

> That networks have the ability to vet and ignore

> That networks are compelled to send

> Which can only be used for good

> And cannot be used for evil.

Considering the last two, I'm assuming this must be tongue-in-cheek.

I think that the best approach for public utilities is to design them in such a way as to minimize, or preclude if possible, their ability to choose their customers or level of service. We want them to have a set of rules/instructions, written beforehand and publicly available, that everyone can point to after an event that says, "well, those were the things they were supposed to do, so they did them." The entities that write the set of rules are separate from the utility: the utility solely exists to implement the rules/instructions as efficiently as possible.

There's obviously a reasonableness factor here: it's impossible to foresee all situations. The effort spent on specifying exactly what should be done should be proportional to the potential impact of that possible decision.

https://github.com/maximelebreton/quick-javascript-switcher

By default, I don't run into these problems and save a lot of compute!

    # Block Anthropic (Claude)
    User-agent: ClaudeBot
    Disallow: /
    User-agent: Claude-SearchBot
    Disallow: /
    User-agent: Claude-User
    Disallow: /

    # Block OpenAI (ChatGPT)
    User-agent: GPTBot
    Disallow: /
    User-agent: OAI-SearchBot
    Disallow: /

    # Block Perplexity
    User-agent: PerplexityBot
    Disallow: /

    # Block Google's AI Training
    User-agent: Google-Extended
    Disallow: /
    User-agent: Google-Extended-Factual
    Disallow: /

    # Block Microsoft's Search & AI Crawler
    User-agent: Bingbot
    Disallow: /

Comments URL: https://news.ycombinator.com/item?id=49011610

Points: 3

# Comments: 0

As I noted in another comment:

Currently this is explicitly against the law[0]:

  (a)In General.—Any person who, in a circumstance described in subsection (d), knowingly produces, distributes, receives, or possesses with intent to distribute, a visual depiction of any kind, including a drawing, cartoon, sculpture, or painting, that—
    (1)
      (A)depicts a minor engaging in sexually explicit conduct; and
      (B)is obscene; or ...
  (b)...
  (c)Nonrequired Element of Offense.—
  It is not a required element of any offense under this section that the minor depicted actually exist.

[0]https://www.law.cornell.edu/uscode/text/18/1466A

  (a)In General.—Any person who, in a circumstance described in subsection (d), knowingly produces, distributes, receives, or possesses with intent to distribute, a visual depiction of any kind, including a drawing, cartoon, sculpture, or painting, that—
    (1)
      (A)depicts a minor engaging in sexually explicit conduct; and
      (B)is obscene; or ...
  (b)...
  (c)Nonrequired Element of Offense.—
  It is not a required element of any offense under this section that the minor depicted actually exist.

[0]https://www.law.cornell.edu/uscode/text/18/1466A

Mere possession of a substance is surely not what society cares about.

There is something ironic about US laws that attempt to prevent crime A by outlawing action B. For example:

  * A: physical sexual abuse of children. B: possession or distribution of CSAM
  * A: drug trafficking or tax evasion. B: structured cash withdrawals

It's my understanding that conviction of CSAM-related crimes do not require any physical act to have ever occurred to any real person: one can be convicted of CSAM-related crimes related to paintings/drawings/created_art of fictional people.

It's my understanding that one can be convicted of structured withdrawals that are not driven by, linked to, or in any way related to anything nefarious.

Whoever made this, please make a "proportion" column somehow!

I would like to know which has the most complexity per total volume or weight (it's less surprising that huge items have high complexity).

My intuition says that the output quality difference is marginal compared to the change in price especially when taking into account the effects of prompt/context engineering and harness differences.

Essentially: since opus 4.6, working through a model's quirks with prompt/context engineering and harness development will yield significantly better output than just switching models to the latest.

I calculate around a 3% drop in speed (from 60mpg) for holding an average sized book out of the window. That's surprising to me.

It's not quite right to use hazard ratios to calculate life expectancy. But if we force it, it looks like being in the top 20% of "regular" sleepers compared to the bottom 20% confers 3-4.5 years of extra life (from birth, assuming everything else equal, assuming USA, etc.). That's 3.8%-5.7% more life (79 year life expectancy at birth in the USA as of 2025). So the numbers are actually close.

I made a bad analogy :)

But you get my point!

Only if we know of an intervention that will likely slow the car down and the risks+cost of that intervention justify the benefit.

Otherwise, we worry without purpose.

EDIT: I will say that there is a philosophical question here related to "basic research" / "pure science" / "fundamental science." Usually just "knowing new things" eventually proves valuable, especially in a long timeline. So in that sense, TFA could be important.

You might measure the speed of your car by putting your hand out of the window and notice that the wind force on your hand is strong when the car goes fast.

Putting your hand out of the window and then blocking the wind with a book doesn't make the car slow down.

Keyword: "associated"

EDIT: I meant to communicate that it doesn't make the car slow down as much as your hand behind and blocked by the book (feeling almost no wind), would imply.

> So the same effort you would expend to get out of those activities on account of their risk, the same effort you should be willing to expend to get a full-body mri.

"get out of those activities"

> To calculate the earnings premium for undergraduate programs, the threshold is the median U.S. Census Bureau earnings of a working high school graduate, aged 25-34 who were not enrolled in postsecondary education during the year of the associated measured earnings, in the state in which the institution is located unless the institution enrolls more than 50% of its students from out of state, in which case the national median is used.

Details here: https://www.nasfaa.org/ge_2026

(For bachelors degrees): Each year, they will calculate the median earnings 4 years after graduation of graduates of every (Institution, CIP code) tuple. (CIP Code: 6 digit code that identifies a "program"; there are ~2,000).

Then they calculate the median earnings of high school grads in that Institution's state across all jobs.

If the high school grads earn more than the bachelors degree grads for 2 out of the past 3 years, the Feds won't offer student loans for that (Institution, CIP code) tuple.

Incidentally, the IRS already supplies this data in aggregated form to the US Dept. of Ed.

Check it out here: https://collegescorecard.ed.gov/

Here is MIT: https://collegescorecard.ed.gov/school/?166683-Massachusetts...