Hacker Newsnew | past | comments | ask | show | jobs | submit | 6thbit's commentslogin

It works, it is popular, sure. Claude's code may be barely old enough to have suffered through its true long-term maintainability problems. They probably also haven't had a lot of rotation/attrition in their staff.

I still can’t believe they don’t let you search videos within a channel for example.

Or filter out music playlist from video ones.

Or search within transcripts.

It’s peak irony a company owned by the search overlord.


I made a little TUI last month for searching within a channel! It supports before: / after:, fuzzy/exact/regex matching, lets you order by upload date/views/duration, lets you search over just a video's titles or descriptions, etc: https://github.com/nolenroyalty/yt-browse

The vast majority of my youtube watching is "go to a specific channel and try to find a certain kind of video" so it drives me nuts that youtube channel search is so bad (and afaik you can't search a channel on mobile?). I end up using my tool to find a bunch of videos and get them into my history to watch on my ipad.

n.b. my tool downloads all video metadata for a channel and then searches over it locally, so it's pretty slow the first time you search a channel (results are cached for 24 hours though).


You can search videos within a channel, go to the channel page and look for the magnifying glass all the way at the end of the nav bar that has

Home | Videos | Shorts | Playlists | Posts | *Magnifying glass here*

Well at least in browser its there, I can't find it on mobile for whatever reason.


you're right! i was only looking at the mobile version. Sucks it's not there cause I typically don't have my pc around when infront of the TV and wanna search something.

It’s peak irony a company owned by the search overlord.

...whose search engine has itself become noticeably less of a search engine and more of a recommendation/sheeple-herding engine over time.


Google of today would absolutely get steamrolled by any of the search engines it used to compete against. Now granted the web of today is mostly a toxic waste pile vs more of a cluttered basement back then.

> I still can’t believe they don’t let you search videos within a channel for example.

Uh, yeah, they do.

https://www.youtube.com/@PuddleOfMuddTV/search?query=blurry

> Or search within transcripts.

Yeah, I also wish this were possible using the normal CTRL+F just doesn't work properly


I've had ctrl+f work for searching within the transcript on the page recently. I assumed it wouldn't due to lazy loading, but was surprised because the video I tried it on was quite long.

filmot.com exists too (found it on here, currently can't get past the cloudflare captcha to double check), but I have no idea how much of youtube's transcripts it has archived.


> I've had ctrl+f work for searching within the transcript on the page recently. I assumed it wouldn't due to lazy loading, but was surprised because the video I tried it on was quite long.

That was previously the case for me, none of the results outside of the current view would show up.

I just went to try, and I noticed that you can actually search in a transcript now!? There's a search bar


Is there an annotated picture mapping city “light dots” to city names?

Looks like they're now playing catchup.

What's the pitch for using Cursor now a days?


Good autocomplete for those of us who still write code.

Me too, I have the bad feeling autocomplete will be sunsetted sooner or later, it clearly isn't the path they're getting into. Also it started to get worse lately, it tries too hard to predict, it wasn't like that some time ago, hopefully you know what I'm saying

Given enough eyeballs, all prohibited items are shallow.


To focus on other things.

100x of a small number could be less than 1x of a really large number from a product that they are more comfortable monetizing.


OpenAI is growing fast, pivoting is only to be expected. It would normally be something HN folk would typically value and encourage on startups.

They have clearly been lacking focus and now finally they seem to be working towards a narrower direction, which is usually highly valued by investors.

This article doesn't explore the depth of the decisions and only regurgitate what you may find your neighbor complaining about on X but with a better stylesheet.


I don’t buy the “wait 7 days” being thrown around as a guard.

Wouldn’t that just encourage the bad actors to delay the activation of their payloads a few days or even remotely activated on a switch?


Of course the "wait 7 days" are not a silver bullet, but it gives automated scanners plenty of time to do their work. Those automated scanners surely catch this `eval(base64.decode("..."))` stuff that some of those attacks used so in my book this dependency cooldown is a net win. I guess the skilled malicious actors will then up their game but I think it's okay to kick off an arms race between them and the security scanners in the dependency world.

That's a good point. In some level I'd prefer the delay to happen on publication of the package itself. Do any of these scanners have cryptographic attestations or similar?

> published manually via a stolen npm access token with no OIDC binding and no gitHead

So this and litellm one would’ve been preventable by proper config of OIDC Trusted Publishers.


Quotes should be around “journalism”.

Let’s recognize those bait posts as what they are, which for sure is not journalism.


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

Search: