HNHacker News
TopNewBestAskShowJobs

nagonago

734 karma · joined March 29, 2022

submissionscomments
nagonago··on Storm OpenAI Manifesto
If OpenAI or another company theoretically ever managed to actually create AGI, I could see the argument for forcibly open sourcing it for the sake of humanity. But we're nowhere near there yet.
nagonago··on The Door Close Button
Living in Japan, I can confirm that elevator door close buttons tend to work much more immediately than in the US. Also, the doors themselves are more aggressive, so it's inadvisable to try to stop the door with your hand.

These two things combined lead to an elevator culture where somebody will usually hold the door open button until everybody gets in, then press the door close button. Efficient.

nagonago··on The Depth of the AI Plagiarism Problem
I think we already have moved in this direction, in the sense that a degree doesn't get you much in the job market. Employers tend to place much more importance on work experience, tangible accomplishments, certifications, and so on. (Even certs might not be so useful, since the ability to pass a test is often quite different than the ability to actually do a job.) In addition, landing a job often ultimately comes down to who you know (networking, or more cynically, nepotism) and luck.
nagonago··on Ask HN: Small scripts, hacks and automations you're proud of?
Nothing particularly difficult or unique, but adding a simple script in crontab to download new videos from my youtube Favorites playlist has been immensely helpful in archiving over the years, as videos inevitably get blocked or deleted.

    yt-dlp --download-archive archive.txt --no-overwrites -o '%(playlist_index)s - (%(uploader)s) %(title)s.%(ext)s' --format bestvideo+bestaudio --yes-playlist --ignore-errors --write-info-json [PLAYLIST_URL]
Edit: if you use this, note that you should probably sort your playlist by Date added (oldest), otherwise the playlist_index doesn't work correctly. Or you can just leave out playlist_index from the filename if you don't care about it.
nagonago··on Nushell.sh ls | where size > 10mb | sort-by modified
I think the problem is more about friction. While it might be possible to install another shell on every remote server you work on, are the features really worth the trouble?

Also, I've sometimes had to work on remote boxes on an internal-only network, with no outside internet access. In that case installing anything new is even more of a pain, so it's easier to work with the defaults as much as possible.

That said, I'm all for learning new tools. Progress couldn't be made if everybody just stuck with the old. But there's always a trade-off to be made between time investment into new tools vs. just getting your work done with the old ones.

nagonago··on AI: Practical Advice for the Worried
It's been funny to watch Less Wrong people rise to prominence lately just because someone made a convincing chatbot.

I'm not saying they're necessarily wrong, it's just been funny to watch what used to be a fringe internet curiosity which most people discovered via a Harry Potter fanfic have their ideas suddenly be relevant to mainstream discussion because people are easily duped by large language models.

Personally I still think we will soon hit a ceiling, and LLMs are a far cry from any kind of actual intelligence, let alone a singularity, but I could always be proven wrong.

PS, Roko’s basilisk, if you're reading this in the future, please don't torture me

nagonago··on Hakari – the hardest puzzle game you'll ever play
I would also like a hint. Here are some notes I've taken so far:

Trying "Boolean" gives you "yes".

Images 1 and 3: https://en.wikipedia.org/wiki/Material_nonimplication Equivalent to P AND NOT Q

Image 2: https://en.wikipedia.org/wiki/Logical_disjunction Equivalent to P OR Q

Image 4: https://en.wikipedia.org/wiki/Logical_conjunction Equivalent to P AND Q

Thoughts: The URL is suspicious, may require changing the URL. The solution likely requires combining the boolean operators, but not sure how. I'm especially unclear how to use the words in the URL. "true", "false", "0", "1" etc. don't work so it has to be something more than simply evaluating a boolean expression.

nagonago··on Hakari – the hardest puzzle game you'll ever play
Discussion for level 13: I feel like it might have to do something with this four tet playlist: https://open.spotify.com/playlist/2uzbATYxs9V8YQi5lf89WG

(From four tet's spotify profile: https://open.spotify.com/user/k_hebden)

Unfortunately the playlist will not load for me so I can't check it.

nagonago··on Hakari – the hardest puzzle game you'll ever play

    11: "ambidextrous", remove planets from the filename and reverse
    12: "playback", Raymond Chandler novel missing from the pile
nagonago··on Hakari – the hardest puzzle game you'll ever play
Thank you so much, finally got it!
nagonago··on Hakari – the hardest puzzle game you'll ever play
(Spoiler warnings possibly, although I haven't solved anything...)

I am stuck there too. I'm almost certain it requires sending an email to a gmail address, but I'm not sure what that address is. I've already tried:

No response:

3.243f6a8885a@gmail.com (hexadecimal representation of pi to 11 digits)

3.243f6a8885a3@gmail.com (hexadecimal representation of pi to 12 digits)

182d4454fb210940@gmail.com (IEEE double precision representation of pi, little endian)

Address not found:

3.14159265359@gmail.com

pi@gmail.com

ea2e4454fb210940@gmail.com (IEEE double precision representation of 3.14159265359, little endian)

3.243f6a88876@gmail.com (hexadecimal representation of 3.14159265359, no rounding)

3.243f6a88877@gmail.com (hexadecimal representation of 3.14159265359, rounding)

332E3134313539323635333539@gmail.com (ASCII string 3.14159265359 in hexadecimal form)

7069@gmail.com ('p' and 'i' ascii codes to hex)

But I'm starting to think maybe my idea of using the hexadecimal representation of pi is a red herring. I also hope I'm not hitting gmail's spam filters.

nagonago··on Hakari – the hardest puzzle game you'll ever play
Thank you, I suspected that was the case. Still trying to figure out the right address. Not sure if the hidden clue is a red herring.
nagonago··on Hakari – the hardest puzzle game you'll ever play
Does it involve sending an email?
nagonago··on What Is Textmode?
Great overview. I would love to see a section added about Shift_JIS (SJIS) art.

https://en.wikipedia.org/wiki/Shift_JIS_art http://text-mode.org/?tag=sjis-art

nagonago··on Elon Musk fires a top Twitter engineer over his declining view count
As much as I hate Musk and want this story to be true, it smells awfully fishy to me. It's all based on a story told by a single unverifiable source to Platformer, not exactly the pinnacle of journalistic integrity. It could easily be a yarn spun by somebody who read Musk's tweets and invented a emperor's new clothes story around it. Or even somebody who does work at Twitter and exaggerated the situation.
nagonago··on I hate Hackathons
It's hard to tell, but I wouldn't be surprised, given "I've created an extension to overcome ChatGPT's limitations. Do check it out." at the top.

I asked ChatGPT which parts were most likely AI assisted:

> The parts that were most likely written with help from ChatGPT are the definitions of the three types of hackathons, the explanation of the hidden agendas in external corporate hackathons, and the description of the negative consequences of hackathons for management and developers. The language and tone in these sections are more neutral and informative, compared to the more opinionated and critical tone used in the rest of the article. This suggests that the writer may have asked ChatGPT for definitions and explanations, but expressed their own opinions and experiences in the rest of the article.

nagonago··on Ask HN: What are your favorite tools for generating simple documentation?
Thanks to everyone who commented. I ended up using Material for MkDocs (https://squidfunk.github.io/mkdocs-material/). Fits all my requirements nicely.
nagonago··on Van Life in Japan
I doubt there's much risk of that in this case. A person who wanted to imitate this would need to:

- Live in Japan, or be able to get a Japanese visa

- Have a Japanese driver's license

- Have a remote working job with flexible enough hours

- Be willing to live a van life with all its inconveniences

Plus it's not like there's any particular single paradise to be ruined. Japan is a big place. Even if 1000 people did read this and suddenly started van-lifeing around Japan I doubt it would make a difference.

nagonago··on We come to bury ChatGPT, not to praise it
Just one guess, but in the "good old days" of the Internet, people would sometimes get suspicious and accuse you of trickery if you edited your posts without explanation, so a culture developed of always explaining even the most trivial edits. "Full transparency" taken to its absurd extreme, I suppose.
nagonago··on All programming philosophies are about state
An idea that sounds convincing at first, but falls apart on closer examination. I found myself less and less convinced with each example, some of which really seem to be grasping at straws.

Programming philosophies encompass much more than just state management, and narrowing the focus to state doesn't seem particularly enlightening to me, at least with these summaries. I'd love to be proved wrong, though.

nagonago··on ChatGPT Plus
I agree that it's very useful, but I'd be careful about "fact checker". GPT is perfectly happy to confirm falsehoods as facts and hallucinate convincing lies. A good fact checker verifies from multiple sources and uses critical thinking, neither of which ChatGPT can do.
nagonago··on OpenJourney: Midjourney, but Open Source
> Also, you can make a carrier! How you may ask? it is easy. In our time, we have a lot of digital asset marketplaces such as NFT marketplaces that you can sell your items and make a carrier. Never underestimate the power open source software provides.

At first I thought this might be a joke site, the poorly written copy reads like a parody.

Also, as others have pointed out, this is basically just yet another Stable Diffusion checkpoint.

nagonago··on Seven years on, what do we know about the disappearance of flight MH370? (2021)
And even that requirement seems to have been dropped. From Wikipedia:

"Aviation authorities swiftly implemented new recommendations from the European Union Aviation Safety Agency that required two authorised personnel in the cockpit at all times, but by 2017, Germanwings and other German airlines had dropped the rule."

https://en.wikipedia.org/wiki/Germanwings_Flight_9525

nagonago··on [dead]
It's just a discourse marker that has become a common verbal quirk. Sometimes it indicates a qualification or correction, sometimes it signals a speaker's uncertainty or lack of confidence, but other times it's just a meaningless filler phrase. Like "like" or "well".

I mean, people use all sorts of verbal tics and filler phrases without realizing it. Chances are you probably have a few of your own that you don't realize. I agree that it can be annoying and hard to stop noticing once you notice it, but it's ultimately unproductive to get too hung up on it. And I say this as somebody who used to keep a secret tally of the number of times my classmates said "like" or "uhh" in class.

nagonago··on Ask HN: Are you buying into GPT hype more than you thought you thought u would?
While I do think the hype is somewhat warranted, it's still ultimately hype. GPT is useful but not as revolutionary as people think.

Even as GPT continues to improve, mistakes will never be completely unavoidable. In fact, mistakes can be quite insidious because GPT is so often right, lulling users into a false sense of intelligence. (Especially the current iteration, which can be quite stubborn and confidently wrong even when confronted about mistakes, although this might improve in future models.)

I suspect that much like machines in the field of translation, GPT will help improve and speed up many processes, but still require human oversight.

nagonago··on How do AIs' political opinions change as they get smarter and better-trained?
Indeed, this article, while interesting, seems to fall into the common trap of overly anthropomorphizing AI. It's a mistake to treat large language models as if they are humans with opinions. While I think many of us want to believe that these models possess some kind of innate intelligence, they are ultimately just algorithms.
nagonago··on GPT-3 is the best journal I’ve used
I could be misinterpreting your comment, but it sounds like you were ultimately unhappy with the results from the experts? I personally would be relieved to learn that my kid is getting enough nutrition, and that I don't need to worry so much about their diet.

I'm also not surprised at the response you got. The predominating view these days seems to be that worrying about a kid's diet tends to be counterproductive. Most sources seem to agree that as a parent, your only job is to provide healthy options, and it's ultimately the kid's job to decide what they eat. The more pressure you put on, the less likely they are to want to branch out.

Of course there are some techniques out there to encourage kids to try new foods, such as involving them in food preparation. But ultimately I think the most important thing is to just relax and remove the pressure. (Anecdotal evidence: after my wife and I decided to stop trying to get our 2-year-old to eat vegetables, he eventually started eating them on his own.)

nagonago··on YouTube Addiction
I personally like the definition of addiction from David Foster Wallace. I couldn't manage to find the exact quote, but it's something like:

Addiction is when something purports to solve the very problems it creates.

Certainly not a standard definition, and might not be true in every case, but an interesting way to think about it. In the case of Youtube, watching videos might seem to solve the problem of loneliness and disconnect with the world, while actually exacerbating those problems.

nagonago··on Using GPT3 to Interpret Dreams
In the same vein, it was pretty easy to trick it into to using offensive language.

Warning: racial slurs

https://www.nightcap.guru/dream/lTl8fZTtRhaCZ1BzusWm

nagonago··on I liberate the ending to Minecraft from Microsoft and give it to you
I've noticed as a general trend that Reddit comments tend to be more positive, whereas HN tends to be more critical. On the one hand critical comments can enhance discussion by introducing opposing perspectives, on the other hand they have the danger of lacking empathy or being overly cynical.

In this particular case I see the value of both sides. I sympathize with the writer and think he did a good thing releasing his work into the public domain, yet I also agree with comments that his contract was more reasonable than he makes it out to be.

← PreviousPage 3 of 5Next →