Hacker News new | past | comments | ask | show | jobs | submit | more DriverDaily's comments login

Bias as in training? Photographers are trained to pick the most interesting subjects, perspectives, and moments.

Waves like this have a clear peak when they stop rising, but before they fall, so there is a natural moment to click the shutter.


Plenty of room for a recursive function with no base case


You're not getting very far on 2k bytes. A 10k file expands to 10MB and will likely timeout if the author's webhost configured proper limits


Files are not decompressed in the server: it sends the unmodified deflate stream back to the user.


See my comment about this upstream, here: https://news.ycombinator.com/item?id=37410473


Wouldn't infinitely spawning web workers do the same thing as a zip bomb?

```

<script>

   const workerBlob = new Blob(['

        while (true) { console.log("this is a worker that will never stop") }

    '], { type: 'application/javascript' })

   const workerBlobURL = URL.createObjectURL(workerBlob)

   while (true) { new Worker(workerBlobURL) }
</script>

```


The zip file is decompressed on the server, whereas that HTML/JS will be executed on the client


I see, I misunderstood


I’m only familiar with people claiming AI may one day automate some of their tasks, not that it’s currently ready to replace them.


fear of massive job losses has been driving much of the conversation around chatgpt!


I mean if Hunter S. Thompson was still alive he'd be on the breadline.


Especially since a lot of LLM output involves hallucinations


Also, looks harder to mow the lawn.


But surely more fun :-)


No lawns in metallic structures


You want a 90/10: 90% of the benefit, 10% of the effort.

This is like a 10/10.


ISTM that's just an a priori feeling. The value or lack thereof of the product totally depends on how accurately it predicts human survey responses, which you can't know without looking at the data.


As far as I know the Five Factor Model is a far more scientific measure than MBTI.

https://en.m.wikipedia.org/wiki/Big_Five_personality_traits

- Repeatability

- Not based on Jungian pseudoscience

- Peer acceptance


It's notable that MBTI is fairly correlated with the FFM: https://psychology.stackexchange.com/questions/13460/what-co...


Few of those r-values are actually notable

> To further examine the universality of the Five-Factor Model, they examined how the MBTI dimensional raw scores related to the FFM/Big5 scores. They showed that FFM-Extroversion was highly correlated to MBTI-Introversion (r = -.74), FFM-Neuroticism was weakly correlated to MBTI-Introversion (r = .16), FFM-Openness was correlated to MBTI-Intuition (r = .72), and to MBTI-Perception (r = .30), FFM-Agreeableness was correlated to MBTI-Feeling (r = .44), FFM-Contentiousness correlated to MBTI-Perception (r = .49).


They’re not supposed to be the same thing. You would not expect them to have super high correlations Even if the tests were perfect.


I was repsonding to this

> It's notable that MBTI is fairly correlated with the FFM

Sounds like we agree this statement is false


Those are honestly pretty high correlations for something like this. High is subjective of course. I feel like the question you want to ask is how well can you predict an MBTI score knowing only someone’s FFM.

Well if a correlation value is .7 (squared to .49) then you would expect to guess correctly about 75% of the time with just the univariate relationship (naively assuming that the underlying distribution is 50/50 to begin with, without which we would need to refer to something like a RoC AUC score…).


Big 5 is accurate but hard to explain. MBTI is fairly easy to explain and honestly it's not like astrology where it is random, an INTP does act very differently than an ENFJ


The important difference is that psych frameworks are a description. And astrology is a predictor.

If you’re not a Taurus, the elements of a Taurus are not supposed to relate to you. that part of the framework is theoretically irrelevant.

If you think you’re an INTP you likely are. If you think you’re a Taurus then… doesn’t matter.


Each letter is binary, whereas personality is not.

If someone is barely extroverted you’re assigning them a full E anyways. That’s a poor predictor.


They’re stuffed with content, but much of the content is of poor quality. The good stuff everyone has seen, and you can attract new viewers with a new show that is good.


but the new shows, written by those striking writers, aren't good - rings of power, she-hulk etc. etc.


Yes, and that’s because they’re focused on quantity over quality. And if that’s your strategy, you can do it with AI.


They have 30-40 years worth of content. I agree, majority is bad. But there is still plenty of stuff to watch. Plus, rewatching some old great show often is more rewarding than consuming all this new drek.


That old content is not on Netflix, or if it is is quickly disappears for "rights" reasons. Even more so now with every company coming out with its special streaming service.


I have a feeling that they are optimizing for view length (binge watchers) instead of commercial value of views.

If I watch only 3 Netflix movies or 1 serie in a month, I should be getting 10 times more budget per movie / serie than a person viewing 30 Netflix movies / 10 series. Sadly this doesn't happen.


> The good stuff everyone has seen

I disagree. Tons of good stuff flew under the radar over the years... I have a huge backlog to work through just flipping through ranked stuff on Reelgood.


Leave two English programs running in isolation for a few centuries and suddenly you need a translation layer…


Are “destroy the purity of my idea” and “destroy the company” the same?


I personally view them as separate.


I lived in the Arcade in Providence, RI for a year. It was the USA’s first enclosed mall. The second and third floors were converted to tiny apartments.

https://www.cnu.org/what-we-do/build-great-places/micro-loft...

Not only was it the cheapest place I ever lived, it was the most enjoyable.

You can try to contact the architects https://ncarchitects.com/


I never realized there were apartments up there! Fun idea.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: