The editorial argues MV3 was designed around a specific threat model — malicious extensions exfiltrating browsing data — and addresses that model at the cost of the API surface that made high-quality content blocking possible. It frames Google's rationale (security, privacy, performance) as coherent but notes the tradeoff falls hardest on tools like uBlock Origin that depended on the blocking webRequest API.
The editorial emphasizes that Google announced the MV3 transition in 2019, delayed it repeatedly under developer pressure, began disabling MV2 extensions for a subset of users in June 2024, expanded that to enterprise customers, and has now closed the store side of the loop. It frames the store removal as the predictable final step in a multi-year schedule rather than a surprise.
Hill, the developer of uBlock Origin, has documented for years that MV3's declarativeNetRequest caps dynamic rules, restricts regex, and makes cosmetic filtering and scriptlet injection substantially harder to implement. He explicitly positions uBlock Origin Lite as a lesser tool, unable to match the effectiveness of the original against modern ad and tracker delivery.
By submitting the story with the framing 'Google Has Removed MV2 Extensions from the Chrome Web Store, Including UBO' and driving it to 622 points, the submitter highlights uBO specifically as the notable casualty — implying the store removal matters precisely because the MV3 replacement can't fill uBO's shoes.
Google has finished what it started in 2019. Manifest V2 extensions — the class that includes the original uBlock Origin, the gold standard of content blocking — have been pulled from the Chrome Web Store. Users landing on the uBlock Origin listing now see the extension is no longer available. Installed copies still exist on user machines for now, but Chrome has been progressively disabling them since mid-2024, and there is no path back to the store.
This is not a surprise, it is a schedule. Google announced the MV3 transition in 2019, delayed it repeatedly under developer pressure, began disabling MV2 extensions for a subset of users in June 2024, expanded that to enterprise customers earlier this year, and has now closed the store side of the loop. The company's stated rationale has always been the same triad: security, privacy, and performance. Extensions using the blocking `webRequest` API can inspect and modify every network request the browser makes, which is powerful and also, in Google's framing, a persistent supply-chain risk.
The replacement is Manifest V3's `declarativeNetRequest` API. Instead of an extension inspecting each request and deciding what to do, the extension registers static rules ahead of time and the browser enforces them. Raymond Hill, who writes uBlock Origin, has been documenting the gap for years: MV3 caps dynamic rules, restricts regex, and makes it materially harder to implement the kinds of cosmetic filtering and scriptlet injection that make uBO effective against modern ad and tracker delivery. His MV3-compatible port, uBlock Origin Lite, is explicitly presented as a lesser tool.
The technical debate has been litigated to death, so let's be direct about the substance. MV3 was designed with a specific threat model — malicious extensions exfiltrating browsing data — and it addresses that model at the cost of the API surface that made high-quality content blocking possible. You can argue those tradeoffs go the right way. You cannot argue they are neutral. The Electronic Frontier Foundation, Mozilla engineers, and Hill himself have all published detailed technical objections; Google has answered some and dismissed others. The rule caps have been raised over time in response to that pressure, which is itself an admission that the original numbers were not workable.
The timing lands in an interesting place. Chrome still owns roughly two-thirds of the desktop browser market. In the same window that MV2 is being removed, the EU's Digital Markets Act designated Chrome a gatekeeper, the DOJ won its search antitrust case against Google with remedies still being negotiated, and Firefox has been posting its first meaningful uptick in installs in years. A large fraction of Chrome's power-user base uses it for exactly one reason — the extension ecosystem — and Google has just made that ecosystem's most-cited extension worse.
Developer reaction on Hacker News, where this story hit 622 points, split along predictable lines. One camp reads this as the completion of a long-telegraphed platform decision that ships a defensible security model. The other reads it as the ad company that owns the browser degrading the ad blocker, and treats the security framing as post-hoc justification. Both readings can be partially correct. The security concerns about `webRequest` are real; extensions with that permission are a genuine attack surface. It is also true that the entity making this decision sells ads for a living, and that no other major browser vendor has felt the need to remove the same API on the same timeline. Firefox implements MV3 but kept blocking `webRequest`. Safari never had it in the same form. Brave forked Chromium and is patching MV2 support back in.
For extension authors beyond the ad-blocking world, the practical fallout is smaller but real. Password managers, session managers, developer tools, and privacy extensions all had to port. Most did. Some quietly shipped weaker versions. A few — particularly niche tools maintained by single developers — simply stopped updating and are now gone from the store along with the MV2 tier they lived on.
If you are a Chrome user who cared about content blocking, you have three honest options. Install uBlock Origin Lite and accept a measurably weaker filter; switch to Firefox where the original uBO still works with the full API; or move to a Chromium derivative like Brave that has committed to keeping MV2-style blocking alive. Edge and Opera are following Google's timeline, so they are not real escape hatches. Vivaldi and Arc are Chromium-based and inherit the same constraints. Chromium forks that keep MV2 support carry their own maintenance risk — you are trusting a smaller team to keep pace with upstream security patches.
If you build extensions, the migration window is over and the question now is what to build on top of the MV3 primitives. `declarativeNetRequest` is workable for most use cases if you were not doing sophisticated request rewriting. Service workers replacing background pages is the more annoying part of the port for most authors; the event-driven lifecycle breaks assumptions that persistent background pages allowed for years. Budget more time than you think for state management and wake-up handling.
If you ship a browser-based product to enterprises, revisit your extension recommendations. Any onboarding doc that tells users to install a MV2 extension is now dead copy. Any internal security policy that assumed Chrome plus uBO as the baseline needs to acknowledge that baseline no longer exists in its old form.
The interesting question is not whether MV3 will get better — it will, incrementally, because Google has to. The question is whether this ends the tacit deal Chrome had with power users: we give you the browser for free, you tolerate that we sell ads, and in exchange the platform stays open enough that you can block the ads if you want. That deal is now visibly one-sided. Whether that matters depends on how many of those power users actually leave, and whether the Firefox and Brave curves keep bending up. Google has spent a decade betting that the answer is not enough to matter. This is the quarter we find out if that bet still holds.
I know Firefox's share of browser share has dwindled to almost nothing. I know sooner or later, well within my lifetime, it will be gone.But I will keep using it until they pry it from my hands. And if they do, I'll use forks until those break down too. I can't stand Google, or what C
I remember in 2010 when Chrome made the web so much better for everyone. All the early adopters were singing its praise and encouraging friends and family to use it.Now, I encourage people to use anything but Chrome. Firefox is really great these days.
Few years ago when google announced their MV2 plan I moved to Firefox. I don't think in all this time I ever missed Chrome for anything in particular outside of maybe 1 or 2 websites that didn't have proper support.Switch to Firefox if you haven't already really.
The solution is to use Firefox. uBlock Origin always worked best in Firefox anyway:https://github.com/gorhill/uBlock/wiki/uBlock-Origin-works-b...
Top 10 dev stories every morning at 8am UTC. AI-curated. Retro terminal HTML email.
Ad blocking has become a safety issue. My parents are at an age where they fall for things. If a malicious ad pops up and offers to install McAfee, some other form of crapware, or outright scamware, they'll click that ad and install the thing. Then I'll get called over when the computer st