Hacker Newsnew | past | comments | ask | show | jobs | submit | sevg's commentslogin

Maybe instead of shooting people down in “What are you working on” threads (I see you have priors), you can instead share something cool that you’re working on.

> My, that was a yummy slime mold!

This is a NetHack reference for anyone unfamiliar.


Does anyone know if it’s possible to have “ultrathink” be the default instead of saying it in every prompt?

https://x.com/bcherny/status/2007892431031988385?s=20 Seems to be moved to the default now. PSA for anyone who didn't see

Put it in CLAUDE.md because that just gets added to the prompt

Ah, thank you! Now I feel like an idiot. I guess I was thinking “ultrathink” was a specially interpreted command within claude code (sort of like a slash command).

I believe it’s to do with supporting shift+enter (to do multiline prompts).

https://github.com/anthropics/claude-code/issues/1259#issuec...


Oh that’s neat. Thanks for highlighting


Hmm, no.

It’s just like all the other postgres extensions named “pg_foo”, and the clear and obvious choice for “foo” in this case is “clickhouse”.

Unless this is some bad joke that has flown over my head.


I will never un-see it now, tbh


definitely a joke, not even that bad


Thanks for sharing this.

I thought this comment was strange at the end of Catfriend1’s post:

> I’ll review the progress from time to time and if I find anything malicious going on, I’ll let you know here.

That’s absolutely not something you say when you trust the person you’re handing things over to :s


Seems like a statement to reassure users who don't necessarily have any trust in the new maintainer. And even if the users trust the new maintainers, it's better to have the reassurance of previous maintainer on top.

Trust is not transitive, nor should it be. We (the users) trust the previous maintainer. They trust the new one. We don't (naturally). The old maintainer says they'll review the new one's work, so we'll have trust the old maintainer (mostly).

Not that the whole trust system can't improve in various ways in general. But for now we have to trust someone.


> Seems like a statement to reassure users who don't necessarily have any trust in the new maintainer.

The statement didn’t seem reassuring.

It’d have been reassuring to hear something like “This person has been a committer for X period, and has demonstrated Y and Z.”

> They trust the new one.

Well my point is it doesn’t sound like they actually do trust the new maintainer. Maybe just poor choice of words, but it didn’t fill me with confidence.


‘I’ll keep an eye on the project and speak up if I discover my trust was misplaced’ is a kind reassurance to the anxious community, but anxiety will just use it as a launchpad for more anxiety. Nice of them to try, though.

I suspect a lot of folks would be horrified at how typical the former maintainer’s approach to trust is in actual reality. It ends up being necessary because there are maybe a single digit number of people in the world who are willing to commit to long-term project maintenance (beyond their own pet peeves, anyways) at all, and with the general hostility towards compensating anyone for their work in software, it’s not like a maintainer can afford to hire and develop a protégé. This is how maintainership worked in CPAN for decades and, barring a culture shift towards paying project maintainers for their maintenance effort, it’s how it’s going to continue working in most projects as us maintainers grow tired and fade out.


I agree - the statement could've been much more convincing. But it's above the threshold for me.

Although I agree if the new maintainer had some creds, it would've been better to use them in a similar reassurance like in your example. But it's hard to really vouch for someone, even if they've made X commits for the past Y years, etc.. Lots of examples here.

If it's still a random/(pseudo-anonymous) account you're trusting, unless there have been some real life appearances or if it's an account that's been proving itself for years, you can only trust them so much.

Basically I agree the message could be interpreted as "I don't trust them, so I'll be on the lookout for anything malicious", but, honestly, at first I just read it as "I trust it, but you can't really trust anyone, so I'll still be on the lookout".


There is an uncounted amount of trusted people who turned to malice, especially in vulnerable situations. Even if someone initially was trustable, they can always have a change of motivation for whatever reason. And that's leaving out accidental fuckups turning harmful. At this point it's clear that even in open source, blind trust can be harmful long term.


Lack of trust is not the same as distrust.


I didn’t mention “distrust” in my comment :)


I think you’re getting downvoted because you’re reporting the typo in an odd and likely unproductive place.

I’m not sure what you expect HN readers to do about the typo. There is a comment section on the blog itself :)


It's not unusual that the author (or someone of the team) see the trafic peak an appears in HN to reply the questions.


Sure, that happens.

But instead of just reporting it directly, we instead get this unsubstantive comment (“Cool! Great! Btw you spelled a word wrong.”). Essentially just noise, nothing that provokes curiosity or interesting discussion.


I bind ctrl-q to lock screen in gnome, which is less bad than losing the whole browser.

I suppose you could also bind it to a noop.


You can tell firefox to ask before quitting ...


You can also tell Firefox to ignore it completely:

    browser.quitShortcut.disabled
As well as to warn:

    browser.warnOnQuit
    browser.warnOnQuitShortcut
Well, apparently I once was aware of these because I have it set in my custom user.js. But I guess ctrl-q will always be lock screen for me, old habits die hard.


> browser.quitShortcut.disabled

The setting is there, it's editable, but Firefox seems to completely ignore it on Linux.

At least up to 145.0.1.


Hmm, maybe it’s still necessary to rebind ctrl-q in some cases.

This says gnome has ctrl-q to quit applications:

https://wiki.gnome.org/Design/OS/KeyboardShortcuts

I don’t know if the above is still the default, but I have this in my gnome setup scripts anyway:

    gsettings set org.gnome.settings-daemon.plugins.media-keys screensaver “[‘<Control>q’]”
(IIRC it’s a terminal command because trying to bind ctrl-q in Settings will quit the Settings window. And you can’t unbind ctrl-q completely, so you have to bind it to something else. You could maybe add it as a custom launcher that just runs /bin/true.)


I use 144.0.2 on Linux and it works entirely fine here.


and testing it now, it asks me by default, and I never changed that setting


> it's not the sort of thing that anyone actually comes out of happy

The lawyers do! :sweat_smile:


Yes please!


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

Search: