HN Debrief

Chrome is looking to permanently drop MV2 extension

  • Privacy
  • Security
  • Browsers
  • Open Source
  • Infrastructure

The post points to Chrome’s long-running Manifest V2 shutdown reaching its final stage. In plain terms, Chrome is removing the extension capabilities that let tools like uBlock Origin inspect and block requests dynamically. Manifest V3 does not kill all blockers, but it does force them into a narrower model, which is why uBlock Origin Lite exists and why many people still say it is not the same product. That distinction drove most of the reaction. People were not confused about the technical mechanism. They were angry at the incentives behind it. The dominant read was simple: Google sells ads, so of course it eventually constrained the best ad blocker on the dominant browser engine.

If ad blocking, privacy controls, or browser-based enterprise security tooling matter to you, stop treating Chromium as a stable foundation. Test a Firefox-based fallback now, or move blocking into the browser itself or a local/network layer before upstream changes force your hand.

Discussion mood

Mostly angry and cynical. People saw the move as an expected power play by an ad company, not a principled security cleanup, while also admitting that escaping Chromium is harder than slogans suggest because Firefox still has real compatibility and performance gaps for some users.

Key insights

  1. 01

    MV2 also powered enterprise security controls

    Beyond consumer ad blocking, Manifest V2 enabled browser security products to inspect requests, maintain background state, and enforce controls inside regulated and enterprise environments. The argument here is that removing those hooks pushes organizations toward Google-managed enterprise products and premium Chrome features, which shifts control from user-chosen extensions to the browser vendor.

    If you rely on browser extensions for security enforcement, audit those dependencies now. The risk is not just weaker ad blocking but losing entire control points your compliance or security stack assumed would stay available.

      Attribution:
    • apimade #1
  2. 02

    Firefox kept the blocker capability Chrome removed

    The useful distinction is not simply 'MV2 good, MV3 bad'. Firefox supports Manifest V3 while still allowing the blocking WebRequest API that powerful content blockers depend on, whereas Chrome channels extensions into declarativeNetRequest. That means Chrome made an additional policy choice beyond the manifest version bump, and that choice is what breaks parity.

    Do not reduce this to a version-number debate in your own planning. Ask which extension APIs a browser actually preserves, because two browsers can both say 'MV3' and still offer very different blocker capabilities.

      Attribution:
    • Sayrus #1
    • doikor #1
  3. 03

    uBlock Origin Lite is fine until it isn't

    A lot of people said uBlock Origin Lite feels identical in normal browsing, and for many sites that is probably true. The higher-signal point was that the original author's FAQ explicitly says Lite is less capable on anti-adblock sites, on breakage handling, and in cases where rules cannot be converted into declarative form. So the real difference is uneven and shows up exactly where the web is most adversarial.

    If you stay on Chrome, test the specific sites that matter to you instead of assuming 'no ads so it works'. The failures are likely to appear on high-friction properties like video, anti-adblock pages, and sites that constantly change their ad delivery.

      Attribution:
    • nolist_policy #1
    • pseudalopex #1 #2
    • michaelmrose #1
  4. 04

    Network-level blocking cannot replace in-browser filtering

    Pi-hole, AdGuard Home, and similar DNS or proxy setups still help, but they cannot fully substitute for a browser blocker that can inspect the rendered page and manipulate the DOM. They miss cosmetic cleanup, element picking, and many cases where ads or annoyances are injected dynamically from first-party domains or client-side scripts.

    Treat DNS and proxy blocking as another layer, not a replacement. If your user experience or threat model depends on cleaning up pages after they render, you still need browser-level controls.

      Attribution:
    • pseudalopex #1
    • nizbit #1
    • varenc #1
    • Markoff #1
  5. 05

    Chrome stays dominant because the web is built for it

    The strongest practical defense of Chrome was not affection for Google. It was that developers test against Chromium first, some sites still only work there, and many users keep Chrome installed as the browser of last resort for government, banking, Google, or enterprise workflows. That self-reinforcing compatibility loop is why Google can make unpopular platform moves without immediate mass fallout.

    If you run a product team, include Firefox and non-Chromium testing in your release process or you are deepening the lock-in you complain about. If you are a buyer, assume market share alone will not protect user choice once engine diversity drops low enough.

      Attribution:
    • michaelt #1
    • dijit #1
    • protoster #1
    • jesuslop #1

Against the grain

  1. 01

    Firefox still loses many users on speed and compatibility

    For a lot of people the blocker story is not enough to outweigh everyday friction. They report worse battery life, slower canvas or WebGL performance, lag on heavy sites, weaker devtools, and occasional breakage on line-of-business apps or mainstream sites. That does not make Chrome benevolent, but it does explain why many technically literate users keep Chromium around or never leave.

    If you want to move a team or customer base off Chromium, do not pitch this as a pure values switch. Benchmark the actual workflows they care about first, especially graphics-heavy apps, enterprise sites, and development tasks.

      Attribution:
    • mbmbn #1
    • prmph #1
    • Markoff #1
    • m-schuetz #1
    • djfergus #1
    • batperson #1
  2. 02

    Alternative browsers are not drop-in rescues

    Orion got attention because its PM said it supports Chrome and Firefox extensions including uBlock Origin, but the replies immediately filled with caveats. People called out performance problems, memory blowups, unclear Linux support, and the fact that on iOS it cannot make uBlock Origin truly work because all browsers sit on top of WebKit. The broader point is that 'supports extensions' marketing can hide major platform and quality limits.

    Vet replacement browsers the same way you would any other dependency. Check platform coverage, stability, and whether key features work in practice on each OS instead of trusting a headline claim.

      Attribution:
    • yannicklesuisse #1
    • monitron #1
    • mead5432 #1
    • jayofdoom #1
    • red_alpacalypse #1
    • TingPing #1
  3. 03

    For many people Lite is good enough

    Despite all the objections, a real minority said their daily experience on uBlock Origin Lite or AdGuard MV3 has barely changed. If your browsing is mostly mainstream sites and you do not hit aggressive anti-adblock setups, the reduced capability may be invisible. That weakens the idea that every Chrome user will feel enough pain to switch.

    Expect uneven migration pressure. Power users and security-conscious users will feel this first, but plenty of mainstream users may not move until specific sites or workflows degrade.

      Attribution:
    • TiredOfLife #1
    • AltruisticGapHN #1 #2
    • HDBaseT #1

In plain english

AdGuard Home
A self-hosted network filtering tool that blocks ads and trackers mainly through DNS-based filtering.
declarativeNetRequest
A more limited browser API where extensions supply preset blocking rules and the browser applies them, instead of letting the extension inspect each request dynamically.
DNS
Domain Name System, the internet service that turns names like example.com into the network addresses computers use.
DOM
Document Object Model, the browser's in-memory representation of a web page that JavaScript can read and modify.
MV2
Manifest V2, the older Chrome extension platform model that gave extensions broader powers such as request blocking and persistent background pages.
MV3
Manifest V3, the newer browser extension model that changes how extensions run and restricts some capabilities used by advanced blockers and security tools.
Pi-hole
A self-hosted network tool that blocks ads and tracking domains at the DNS layer for devices on a network.
uBlock Origin
A popular open-source browser extension that blocks ads, trackers, and other unwanted page elements using powerful network and page-level filtering rules.
uBlock Origin Lite
A reduced-capability version of uBlock Origin built to fit Manifest V3 restrictions.
WebGL
A web standard for rendering 2D and 3D graphics in the browser using the computer’s graphics hardware.
WebKit
The browser engine used by Safari and required for browsers on iOS and iPadOS.
WebRequest
A browser API that lets extensions observe and block network requests in real time, which is important for powerful ad and tracker blockers.

Reference links

Browser vendor positions and technical references

Enterprise and government security guidance

Alternative browsers and projects

  • Orion Browser
    Mentioned by Orion's PM as a WebKit-based alternative that supports Chrome and Firefox extensions.
  • Orion GTK early beta
    Linked to clarify that Linux support exists only as an early beta Flatpak.
  • Ungoogled Chromium
    Suggested as a Chromium variant likely to preserve or patch back MV2-related functionality.
  • Helium
    Recommended as an Ungoogled Chromium-based browser that ships with uBlock Origin by default.

Compatibility and ecosystem tools

  • webcompat.com
    Shared as the place to report sites that block or break on Firefox so Mozilla can pursue fixes or workarounds.
  • Microsoft Edge Workspaces
    Used to clarify the platform and account requirements for an Edge feature one commenter liked enough to stay on Chromium.
  • Netflix 1080p for Firefox add-on
    Mentioned as a workaround for lower Netflix resolution on Firefox/Linux.