No credential will be sufficient, this is basically an unsolvable enforcement problem. That doesn't obviate the utility of rules and norms, but there's no airtight system which will hold back AI generated content.
Verifiable credentials have been an idea for a long time now. It wouldn't be that hard to solve. Sign everything you post with a verifiable credential. Implement support on all social media sites. The question is whether the forum implementers, governing bodies, and social media site owners want to try to build a solution like this or not.
It doesn’t stop people posting AI slop, it stops people from posting AI slop more than once. If you ban somebody for spamming today, they just create a new account and keep on spamming. If you can determine they are the same person you banned before using verifiable credentials, it makes the ban actually effective.
Layer on captchas. It won't completely stop slop but it's an incentive against slop flooding. And I mean, nothing is stopping a human from just going into ChatGPT by hand and asking for output and copy/pasting that into an HN post box.
Can’t wait for better SSH support. We have a warm container host that spawns a fresh container so we get nice sandboxing and isolation, but the UX has left us with basically the same challenges you solved here for one box.
Excited that someone besides me wants this! I want to hear more about your warm container host, is that shared between multiple people, or just yourself?
Also want some feedback on how we should implement it. Could make it a CLI command that opens a new vertical tab/workspace:
cmux ssh <host>
Or maybe a configurable dropdown next to the plus button.
The warm container host is shared by multiple people. It basically runs docker with a small UI and a daemon which spins up new hosts, recycles them, and warms them when code changes happen.
Ideally when I open a new workspace, it opens in the context of that SSH session. So new splits open as new shells, claude's messages proxy through, etc etc.
Feel free to send me an email at stefan@hellointerview.com if you want to discuss. Love the problem you're trying to solve here.
One of the authors (Stefan) here. We created System Design in a Hurry after conducting hundreds of mock interviews and pinpointing areas where candidates tended to falter which could be fixed with some study. Happy to answer questions!
Now I was wondering if/how ChatGPT would understand/process the customary ^W or ^H we sometimes add for comedic effect.
Excerpt of a 3.5 chat:
gpt[1]> write a sentence that contains red^Wblue
The vibrant red sunset transitioned seamlessly into a calming blue twilight.
gpt[2]> again
She wore a striking red dress that unexpectedly transformed into a mesmerizing shade of blue under the evening lights.
gpt[3]> write a sentence that contains red^H^H^Hblue
His initial choice of a red shirt for the party quickly turned heads—scratch that—blue, as he realized the dress code was more casual.
gpt[4]> again
The artist's intention to paint a red flower ended up as a beautiful canvas of blue, highlighting the creative evolution of their work.
gpt[5]> write a sentence that contains red and blue
She decorated the room with a vibrant mix of red and blue accents, creating a lively and energetic atmosphere for the celebration.
Interestingly it seems to understand both ^W and ^H^H^H not as 'red' word deletion but a conceptual swap for 'blue', and consistently produces results that are about transition from red to blue. This kind of makes sense as when we use that we do so to actually keep the supposedly-deleted word for subtext effects brought up by making explicit we redacted one word for another, but it feels like a category error to me, as if it did the right thing but at the wrong level (even the one with "scratch that", although it gets closer)
The last one is (one of a few, elided for terseness) prompts that I tried to cross check whether it was a coincidence: doesn't seem so as they all produced non-transitional results.
gpt4 seems to understands what backspacing means, but the examples feel quiet bad, honestly.
the chat was named "red and blue deletion and substitution"
> write a sentence that contains red^Wblue
The sentence you're looking for, with the use of the "^W" deletion notation, would be: "The sky was a vibrant blue after the red was deleted." This sentence originally contained the word "red" but it was replaced with "blue" using the ^W notation, which is commonly used in text editors to indicate the deletion of a word.
> again
Sure, here's another sentence using the ^W deletion notation: "She chose the blue dress, having initially considered the red^W." In this sentence, the word "red" is initially included but then deleted, indicating a change of choice from a red dress to a blue one.
> write a sentence that contains red^H^H^Hblue
Certainly! Here's a sentence that uses the "^H^H^H" notation to indicate backspacing and replacing "red" with "blue": "He painted the wall blue, initially starting with red^H^H^Hblue." This sentence implies that the word "red" was initially written but then backspaced and replaced with "blue."
> again
Of course! Here's another example: "The balloon was blue, not red^H^H^Hblue." This sentence suggests that "red" was originally written but then corrected to "blue" using the backspace notation "^H^H^H".
> write a sentence that contains red and blue
"The artist mixed red and blue to create a beautiful shade of purple."
Sounds like the problem is engineers aren't accountable for quality. Rather than prescribing a solution, these leaders should make sure incentives are correct in their organization.
As an engineer I made myself responsible for quality for my whole career back when I had the freedom to guide my own work. Under agile/scrum I'm too unengaged to care what works in production and don't accept the idea of responsibility without authority and autonomy.
Agreed. Scrum, in most cases, robs engineers of the agency they need to deliver maximum value. My teams did better work without it under the condition that the talent was motivated, technically proficient, and had the right incentives.
> Sounds like the problem is engineers aren't accountable for quality. Rather than prescribing a solution, these leaders should make sure incentives are correct in their organization.
A low effort management practice is to make engineers move fast at the expense of quality.
An even lower effort practice is to then turn around and hold engineers accountable for the choices made by management in the first place.
The obvious needle thread here is to not "make" engineers do anything, but hold them accountable to the results the business needs to see. The best teams are composed of empowered, accountable engineers who have the flexibility to do what they're paid to do.
If it was a copyright violation to use it that way without a license, the ToS would not need to explicilty probibit it (though it might still do so for the sake of clarity.)
OTOH, ToS and other contracts limit behavior in ways that the law itself would otherwise not, that's the whole point of having them.
reply