Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

uBlock Origin Lite: Description

github.com

Post reply on HN
Print view View original

61–70 of 155 posts

Re: uBlock Origin Lite: Description

#61

Post by hombre_fatal »

novok wrote: post #54
hombre_fatal wrote: post #28

Earlier quoted context omitted.

Trivia: 1Blocker convinced apple to raise the limit from 50k to 150k per list for Safari content blockers.

My question is why have limits

Unbounded lists rarely seems like a thoughtful way to start out.

But there are probably some DX and UX improvements over unbounded lists. e.g. recompilation takes time, so you set up developers for better UX when their dynamic lists are shorter and separate from their static lists, and they can compile multiple lists at once. (A 50k rule list took seconds on my old Macbook Air). Also encourages breaking lists into logical sublists which is better UX (e.g. regional lists)

Also might discourage append-only accumulation where there’s no real incentive to prune the list (i.e. see EasyList). Some limit can help here.

Kind of like deciding on a queue size vs unbounded: makes sense to set a limit, see who hits it, and cross that bridge as needed. In Apples case, they 3x’d the limit after a real world product made a real world case.

Just some thoughts having built my own Safari content blocker.

Re: uBlock Origin Lite: Description

#62

Post by deathanatos »

hombre_fatal wrote: post #27
jjulius wrote: post #10

Earlier quoted context omitted.

>... it's often a positive signifier that describes low resource usage/costs Third sentence: >This means that uBOL itself does not consume CPU/memory resources while content blocking is ongoing -- uBOL's service worker process is required only when you interact with the popup panel or the option pages.

Sounds like a plus for manifest v3 then. Fewer resources, fewer permissions, to do most of the same task.

Adding support for a less-resource intensive API to filter is fine. The problem people have is the destruction of the more powerful APIs to permit the fine control that UBO requires, to do the job proper. There was no need for that, there was no trade-off¹ being made.

¹of technical concerns.

Re: uBlock Origin Lite: Description

#64

Post by hombre_fatal »

staticassertion wrote: post #39
ck2 wrote: post #12

Isn't the biggest problem with MV3 the 5000 rule limit? Or are there ways around that by compiling it into something internal? This is how adguard did it: https://adguard.com/en/blog/adguard-mv3.html I'd suggest stop trying to support chrome and let it die? It won't work on mobile anyway, only Firefox supports ublock origin on mobile?

If no one were building an adblocker for Chrome I'd just build it myself. I'm sure I'd do a worse job than the people doing it today but it'd be better than nothing.

Also, there are other adblockers on Chrome. They just aren’t as good and are sketchier. By pulling the extension you just enrich worse actors at the expense of innocent users.

Re: uBlock Origin Lite: Description

#65

Post by SteeCee »

cptaj wrote: post #49
baal80spam wrote: post #48

Earlier quoted context omitted.

You bet he should! To me, there are 3 projects I'd throw my money to: - uBlock Origin - ytdl - The Internet Web Archive

Not wikipedia? It seems way more important

If you've ever seen how topics are treated in a Wikipedia discussion thread you'd feel like the whole website has zero credibility. There's a reason academics don't let you use it.

Re: uBlock Origin Lite: Description

#66

Post by bornfreddy »

I'm really really really hoping that the non-crippled version, which works in Firefox, doesn't die the way uMatrix did (loved it, still miss it). That said, huge thanks to gorhill and all the others working on this project!

Re: uBlock Origin Lite: Description

#67

Post by forgotusername6 »

nottorp wrote: post #41
skyfalldev wrote: post #40

Earlier quoted context omitted.

Since MV3 adds quite a lot of limits on what uBO can do, you have to grant permission for each site to have cosmetic filtering.

So we need to build something external to the browser that will detect uBlock Origin asking for permissions and automatically grant them :) Edit: doesn't this mean that any site you visit will get a chance to track you the first time, until you manage to click to grant permission?

From the blog post on how adguard built their v3 extension the ads themselves are blocked but aren't removed from the page if the cosmetic filters aren't run

Re: uBlock Origin Lite: Description

#68

Post by nightpool »

I'm really happy that Gorhill has listened to feedback on the branding and explanation of this, and added back in the ability to "opt in" to extended permissions on individual sites. Thank you for your work, and I'm sorry if my comments in the preview HN thread came off as overly hostile, because given the original comments you made about MV3, I couldn't imagine this sort of outcome. I hope that you or someone else still works to release an MV3-compatible "Full site permissions" version of uBlock as well, since I think there are a lot of potential options for improvement by using declarativeNetRequestFeedback, and it's certainly more convenient to not have to opt-in. And I'm still worried that without an MV3-compatible version of full uBlock Origin, users are still going to assume that "MV3 means that ad blockers are broken".

Re: uBlock Origin Lite: Description

#69

Post by runjake »

baal80spam wrote: post #48
Ligma123 wrote: post #11

Gorhill should have a statue somewhere for making the web browseable.

You bet he should! To me, there are 3 projects I'd throw my money to: - uBlock Origin - ytdl - The Internet Web Archive

But, do you donate?

https://archive.org/donate?origin=wbwww-TopNavDonateButton

(Couldn't find links for uBo and ytdl. I've asked gorhill where to donate in the past and they've refused donations.)

Re: uBlock Origin Lite: Description

#70

Post by longtimelistnr »

SteeCee wrote: post #65
cptaj wrote: post #49

Earlier quoted context omitted.

Not wikipedia? It seems way more important

If you've ever seen how topics are treated in a Wikipedia discussion thread you'd feel like the whole website has zero credibility. There's a reason academics don't let you use it.

The reason is that it is not a direct source of course because the authors cannot be validated. Never met a college professor that said don’t use Wikipedia (if anything it was encouraged) as long as you’re pulling from the citations. I had plenty of high school teachers talk about how awful it is though and I think it was because they just didn’t understand it
Post reply on HN