Except mobile browsers, arguably the most critical place where you need blocking to get any kind of reasonably decent experience (load times, viewable area, battery life, etc.).
Extensions like uBlock target people who don't want to bother with editing hosts files and keeping them updated.
Agreed. My point was playing with hosts file is definitely not something easy on a mobile, or something most people would want to bother with. Especially when it comes to adding exceptions and getting some sites to work.
For computers and Android devices Firefox and uBO are far more capable solution than DNS blocking, at least when it comes to browsing.
On iOS, you can use DNSCloak[1] which has recently been open-sourced[2]. You can either setup a DNS server (DoH or dnscrypt) somewhere that include these filters, or you can use in-app Blacklist functionality to block domain (although it needs domains only-type of block file, instead of host file, but converting them is quiet trivial).
I use PiHole and firewall to manage adblocking at home, and uBlock on every device that supports it. uBlock is a far more effective solution than DNS based ones which is why I prefer to install Firefox+uBlock on Android than use hosts file.
I'll give DNSCloak for iOS a try, thanks for the recommendation.
If you use an Android browser that supports the Chrome Custom Tabs protocol [1] such as Lynket [2][3][4] it's possible replace the Chrome tab that appears when you click/tap a link with a completely differently browser such as Brave or Firefox (inc. Focus, Nightly or Preview). [5]
By doing this you inherit the features of the replacement browser like ad-blocking and tracking protection.
You can sign up at https://www.nextdns.io/ and then you can block ads at the DNS level. Simple to do, and updated, and without the overhead of running a local VPN.
I think that's a reason why DNS based ad-blocking is as effective a solution as any, though easily worked around against (I see that Amazon's android app is able to show ads despite DNS blacklist).
It would be great if you were right, but I don't think you are. Do you have sources?
I got the idea from this [1] reddit comment made by someone with an Mozilla Employee flair:
> [...] There is not yet a concrete plan (feature) to bring Web Extensions to Fenix. That does not mean it is not going to happen. It does mean we have some things to figure out first.
and this [2] github issue where they state
> Not for MVP, will look at it for further versions
I fear that what they'll want to figure out first is whether not implementing them will hurt adoption.
[1] https://raw.githubusercontent.com/StevenBlack/hosts/master/h...