Today, I’m proud to announce Homebrew 7.0.0. The most significant changes since 6.0.0 are faster installations and upgrades, stronger sandboxing, a native macOS app, built-in vulnerability checks and an advisory database, the end of macOS 10.15 support and Intel Macs moving to Tier 3 (announced last year).
Literally just upgraded before seeing this post; I had 31 outdated packages and the first thing I saw was how blazing <i>fast</i> it was! I looked online to see if something changed and saw this post.<p>Speed was my biggest complaint about brew, so I'm glad to cross that one off my list of things I don't like about brew.<p>Thank you for your hard work BTW, brew's been a lifesaver on macOS and immutable distros.
Thank you for Linux Homebrew! It works so damn well.<p>For me the distro package manager is for system packages, Homebrew and Flatpak for the user facing apps.
Where do you draw the line between system packages and user facing apps? Some software defies such an easy categorization. If your default install doesn't come with docker and you install docker later for development, does that make docker a user-facing app? What about language toolchains like golang, rust, npm, etc?
It's an interesting question on Linux. On macOS entire toolchains would be fair game. I just ran `brew install colima`, `brew install llvm`, etc.<p>But clearly anything that would conflict with distro-specific opinionated decisions is out. Or DE-specific opinionated decisions.
<i>> Where do you draw the line between system packages and user facing apps?</i><p>If it works in Homebrew, I almost always pick Homebrew. :) I have a pretty good feeling for what works since for the past few years I've mostly used an atomic distro (Aurora, based on Universal Blue, based on Fedora). It just comes naturally for me on Fedora too.<p>I've found that this way you can get many of the stability pros of using an atomic distro even on a non-atomic one.
On Linux I use mise to install CLI tools
I don't understand the distinction you use here.<p>Then again I use versioned AppDirs on Linux since +20 years anyway, so I am
not really into any arbitrary disctinction random linux distributions try
to push down onto the (downstream) userbase. Besides, if you compile from source, why
would you want to rely on the distribution package manager to begin with?<p>None of them allow for versioned AppDirs by default as far as I know; NixOS
uses a hashed name, so that is the only exception I can think of (and GoboLinux
of course), but as far as I know if you are on e. g. a debian system, you can
not use it for a versioned AppDir layout.
<i>> I don't understand the distinction you use here.</i><p>Distinction: <a href="https://news.ycombinator.com/item?id=49683258">https://news.ycombinator.com/item?id=49683258</a><p>I use "dnf" to upgrade my system and "flatpak update" and "brew upgrade" to upgrade the apps I've installed.<p>An app like VirtualBox does not install using Flatpak or Homebrew, so I would use dnf for that. It's usually an app or two that doesn't work via Flatpak/Homebrew/AppImage that I need to install using dnf.
> NixOS uses a hashed name, so that is the only exception I can think of (and GoboLinux of course)<p>Also Guix, which is inspired by Nix. Don’t know about AppDir support, though. Are AppDirs more of a general concept or a formalised standard? In what capacity are you using them?
Congrats and thank you to you and your team for maintaining critical infrastructure we all take for granted.
I don't know what I would do without brew on my Macs. Thank you and everyone who contributes!
Silly question - but how do you apply the cooldown to brew itself ?<p>I would think this is one of the most critical dependency that you would not want to get right away at let it rest for a couple of days
You can’t, by design. We instead apply if for you on upstream packages from NPM, PyPi, etc.<p>Our model is very different to those where cooldowns exist and make sense. As-is if would just delay security updates too.
would you please share what amount of the new dev (work done on brew) is AI/LLM-assisted. this major version bump was very quick to arrive compared to when v6 got released?
I think at one point you said you’re working on a rust rewrite? Is that still in the cards?
I tried it. It ended up being slower on most non-synthetic benchmarks (like repeatedly installing the same thing with warm caches).<p>The lessons learned were instead used to make the Ruby frontend much faster.
<a href="https://fishshell.com/blog/rustport/" rel="nofollow">https://fishshell.com/blog/rustport/</a><p>Fish 4.0: The Fish of Theseus (fishshell.com)<p>906 points by jdxcode on Dec 28, 2024 | hide | past | favorite | 198 comments<p><a href="https://news.ycombinator.com/item?id=42535217">https://news.ycombinator.com/item?id=42535217</a>
They experimented with it but ultimately decided to focus on perf with their existing ruby codebase.<p><a href="https://github.com/Homebrew/brew/issues/7755#issuecomment-5145260312" rel="nofollow">https://github.com/Homebrew/brew/issues/7755#issuecomment-51...</a>
I found that Mise handles all my needs for development and it is scoped so it doesn't try to update Python and break all my virtual env when I install something new /shrug
I also started using Mise for global CLI tools instead of brew and it’s working really well<p>Eg: mise use -g gcloud instead of brew install xxx<p>It can even do that for npm packages! Like mise use -g npm:xxx
Why would you even use python without uv anymore, and have a system python binary and virtual envs linked to it?
I split my usage. Homebrew for OS things mise for the various tooling.<p>The only problem some things still have a dependency on requiring python and others on the system. The problem being that they’re there, mise works just fine.
my split: brew for casks, mise for tools
Mise is really sloppily vibe coded these days
TIL Homebrew has its own sandbox mechanism - looks like it's built around their own sandbox-exec wrapper, at least on macOS: <a href="https://github.com/Homebrew/brew/blob/d79ef822ab8136e393ed5f86e2b56afc68d04874/Library/Homebrew/sandbox.rb#L227" rel="nofollow">https://github.com/Homebrew/brew/blob/d79ef822ab8136e393ed5f...</a>
Yup! We’ve been using it for a really long time at this point.<p>Were using Bubblewrap on Linux and moved to Landlock this release.<p>P.S. HUGE fan of your blog and writing Simon, keep up the great work. It’s the #1 resource I recommend to anyone in the industry wanting to learn more about LLMs (and pelicans).
I've been using homebrew for so long, I can't tell you how much it has helped me to spend more hours developing and less managing packages and updates, etc. Thank you so much, congratulations on the update!
I prefer to use Mise for everything nowadays. homebrew bootstrap specifically. A lot easier to manage all of my packages in one single file, homebrew, Node packages, etc.<p><a href="https://mise.jdx.dev/bootstrap/packages/brew.html" rel="nofollow">https://mise.jdx.dev/bootstrap/packages/brew.html</a>
Always a joy to see security and performance as a big highlight on a release. Brew is notably slow compared to the other package managers I use, but never in an unbearable way.<p>I've always used it on my Mac work machines, but always used my native package managers on Linux. What are the reasons to run homebrew on Linux? Better newer package support when you're on something slower moving like a Debian distro?<p>I've standardized my config scripts on using language managers for those tools (go, cargo, uv), which isn't perfect, so maybe brew is worth a chance.
As a MacPorts users (generally light-weight, just a few things here and there):<p>Has anyone gone from MP to HB, or vice versa? Why did you switch from one to the other (and perhaps back again)? What did you find as the pros/cons of each?
I switched from Homebrew to MacPorts a few years ago because of Homebrew's aggressive dropping of macOS versions they consider "too old." Even in the top-comment announcement here[1] they are making sure to include that they are removing support, as if they're proud of it!<p>MacPorts generally doesn't care what version of macOS you run, so here I am. I'll probably never go back to Homebrew. MP is great!<p>1: <a href="https://news.ycombinator.com/item?id=49681546">https://news.ycombinator.com/item?id=49681546</a>
Awesome, I just checked and had already upgraded at some point. I run the alias command below every now and then which keeps everything up to date.<p><pre><code> alias u="brew update && brew upgrade --greedy -y && brew upgrade --cask -y && brew cleanup"</code></pre>
Does installation still require root and a dedicated user account on Linux?<p>That really put me off.
See the release notes: we now have experimental support for using any prefix shorter than the Linux default. We are aiming to eventually fully support (Tier 1) any prefix under 64 bytes long.
The dedicated user isn't really required; it just expects that /home/linuxbrew/.linuxbrew exists. That can just be a symlink to your home directory. But yes, creating that symlink requires root on most systems. Brew itself doesn't require root or that pathname; you can put packages anywhere, but then many will have to be built from scratch since the pre-built packages don't work. And you need bubblewrap installed (which requires root) to use the sandbox, but again that is optional.<p>All that said, when you are running without a sandbox installing to a nonstandard location, not everything works consistently. I run this mode all the time. But it has been getting better.
I always download the install script and change the directory to my user home directory, ~/.brew
The dedicated user account is optional.<p>It uses root once to chown its prefix directory /home/linuxbrew/.linuxbrew
Farewell, Homebrew. It's been a good run.<p>- 2019 Intel iMac user.
Sorry we couldn’t support this for longer :(<p>From the release notes:<p>> The Intel support decision reflects the limits of a volunteer-run project: Apple have dropped Intel x86_64 support from macOS 27 Golden Gate and GitHub Actions will retire Intel macOS runners in autumn 2027. If Apple and Microsoft’s GitHub, two of the world’s largest technology companies, cannot continue supporting macOS Intel x86_64, sadly neither can Homebrew. MacPorts still supports macOS Intel x86_64 and is likely to provide better results on this platform.
My personal opinion, but I'd rather see wider macOS support than Linux support, if it's a sheer matter of where time is allocated.<p>Linux already has world class package managers.
Linux Homebrew users would disagree.<p>Supporting macOS Intel has been significantly harder, slower and more expensive for us the last few years than adding ARM Linux support. If it was easy and free we’d have kept it for longer (until Apple and GitHub killed theirs at least which is coming in less than a year).
How come that MacPorts supports macOS versions all the way back to 10.5 Leopard in the latest version?
That's an excellent question for the maintainers of the MacPorts project! It can supplement the very clear and polite answer you've already received from a maintainer of Homebrew.
Different philosophy of dependency management.
Thank you for all of your hard work, I’ve been very happily using Homebrew on OS X / macOS since 2010.<p>No need to be sorry, the situation is understandable.
very disappointed about this
We've all gone through it :-( Happy users of the (really good) software until our hardware gets "too old" and the inevitable rug pull.
MacPorts is still there for us.
Same boat: 2018 Intel Mac mini, stuck on macOS 10.15.<p>When I eventually upgrade my hardware, I’ll happily use Homebrew again. In the meantime, MacPorts!
I am quite disappointed by their decision to remove wine[1] from brew. The utility of it is obvious, and it's very popular software.<p>[1] <a href="https://formulae.brew.sh/cask/wine-stable" rel="nofollow">https://formulae.brew.sh/cask/wine-stable</a>
Not usually a “+1” or “Me too!” kinda guy but homebrew is incredible and makes dev on Mac quite seamless. Huge thanks to the team.
Thanks for the hard work, but the Tahoe limitation for the GUI app is absolutely ridiculous.
The GUI is pretty sharp, but I don’t like it uses emoji instead of SF symbols.<p>Is it Claude or Codex built?
I didn’t build it. The emoji were my idea. I (and Homebrew) have used them heavily before LLMs did. Maybe LLMs loving them is my fault. Sorry if so.
I don’t get these questions asking if it’s codex or Claude – I see them often. Does it matter? There are other options out there also, it feels strange to ask this. Like asking “is your car a Toyota or a BMW?“
If you say "AI tools", some vibe-coders try to equivocate between tools that generate the whole program for you and, like, an editor that uses an scoring algorithm to decide which method to show at the top of an autocomplete list. But if you say "vibe coding", some vibe-coders try to claim that what they're doing technically isn't vibe coding because they applied some non-zero amount of testing or review during the process.<p>If you ask "did you build this using Codex or Claude", it removes most of the wiggle room. And it's worded pretty neutrally, so it will often get vibe-coders who aren't technically using either of those tools to say something like, "Actually, I've built a custom harness for DeepSeek..." instead of an outright denial.
I was surprised to see my Homebrew consider my macOS a Tier 2 for using the very latest macOS version and Xcode version… But I think it's because I don't have Xcode 27.0, which is presumably releasing tomorrow?
You can download Xcode 27 RC now from <a href="https://developer.apple.com/download/applications/" rel="nofollow">https://developer.apple.com/download/applications/</a>.
This is a bug, sorry. Working on it.
What is the impact of being tier 2?<p>I’ll update Xcode, but I will definitely not install macOS 27.0
I guess the Hackintosh era is (finally) facing its death...
the gui homebrew manager is a nice addition!<p>it can be installed with:<p>brew install homebrew-app
And that's the end of Homebrew for me as a user. I like the app, but my old Intel MacBook apparently can't handle it anymore.<p>I'm back to the old installation methods !
I recommend MacPorts, the migration is easy, you copy the package names you installed with brew, uninstall homebrew (there's an unisntall script on brew.sh FAQ section, it will also uninstall all homebrew packages (but not the casks)) then install macports and reinstall the packages with sudo port. Although the selection is smaller on macports and newer packages might not have a prebuilt pkg for your OS so you still might end up building manually (My poor i5 4core had to attempt building nodejs26 because there wasn't a pkg for Ventura)
Actually, Homebrew 7 is supported on your Mac, you just have to install Linux to use it. :)<p>If you want to continue using the same MacBook that will be only path forward when Apple decides to EOL it (if it hasn't already happened).<p>I'm a long time Homebrew on Linux user, it works really well. I always prefer it over the distro supplied package manager for user facing CLI apps.
How long do you expect them to support Intel processors though? It’s been like 6 years since the last MacBook release had any Intel processor, and Apple doesn’t make OS updates anymore, so it doesn’t surprise me that Homebrew stopped too.
Are you implying that a 6 year old hardware is as good as a paperweight? Outside of the Apple fairyland this is just bonkers.<p>I'm not blaming the Homebrew project/developers here, but the fact that many people are okay with deprecating perfectly fine hardware.
No, he’s asking what you think a reasonable support timeline is. An open source project supporting hardware 6 years old is pretty awesome.
6 years is really not a long time at all. You can probably find heaps of open source software out there that compiles and runs perfectly on 20 year old PCs. It's not like the maintainer has to do much to retain support--they just have to <i>not</i> make the software dependent on new operating systems.<p>Bits don't rot.
NixOS is an open source project that runs great on my 2013 Dell laptop. I think the real point of distinction is that brew depends on Apple's OS and SDK, which somewhat ties their hands, while other OSs don't.
There's been a huge step up in basically everything going from intel to apple silicon. It's not a comparable hardware in any sense, and as such, it really does feel like it became a paperweight? I know I never wanted to see an intel macbook ever again the moment I tried M1. It's maybe hard to comprehend how different it is if you haven't used both. And at the same time I'm happy with my x86 desktop for 10+ years. Only thing that really improved in that span was storage.<p>So I wouldn't say this applies globally to apple users/products, but the post-intel era is a special case. If nothing else, they are helping the environment by not feeding that (literally) hot garbage with more electricity.
You're on macOS. you should know full well that they do not care about those values.
I have one of the last Intel Macs and works perfectly fine (on Sonoma). How long should I expect working tech being supported?<p>I am already looking at Linux, but still need to get out of the Apple ecosystem. I am not going to spend +2500 Euros every 5 years on a laptop
Sonoma will be getting its last security update imminently (possibly tomorrow).
What does saying "working tech" do for you? If I have a working Samsung CRT from 25 years ago do I ping them about smart TV support? Nowadays it's a shitty situation with planned obsolescence; but 6 years for an open source project dedicating resources to a dead end is more than enough and appreciated.
And given the performance jump from Intel to Apple Silicon, the perceived difference is 10+ years even.
Sure, but 2019s Intel Macs are aboslutely usable for most tasks, and given much of the agentic dev. happens... somewhere remotely, it is more an opportunity to reuse this equipment for various tasks, rather than pollute the already dirty Earth with more electronic waste.
MacPorts has been there this whole time.
I've used MacPorts on macOS versions that don't support the latest OS, which per homebrew policy seems to be versions upto N-2.<p>Apart from the issue of having to compile the downloaded software leading to spinning fans, I'm not too concerned.
Same here, but I understand their position. Meanwhile I rely on packages publishing builds for Intel Mac in GitHub (surprisingly a lot) or building from source code. I just asked AI to build me install system.
FYI Macos will also stop with major updates for Intel macs. You can just keep using brew 6 on macos 26.
I’m installing Linux on my 2019 16” today.
Have you tried Mac Ports?
You do realize Intel support is gone, right? You won't be installing anything on your Intel Mac very soon...
[flagged]
Fantastic, thx for the effort!
Time to `brew update && brew upgrade` later today. Always a little nervous, but Homebrew usually makes it painless.
Unfortunately this project has a case of dread with every update to me. To see what is being force changed, worse support or even removed features.<p>Looks like this release breaks Intel Macs. Very very sad. Not even 5+ year old hardware is apparently supportable to some devs
Thank you so much for doing this, for years you make me feel like home, or the year 2010. Please don't ever stop!
Does it still need root on Linux? That makes it an instant no go.
still needs fucking 'sudo'
[dead]
[flagged]
"That front page" isn't the front page, it's the release notes for version 7.0.0.<p>The actual front page has "The Package Manager for Everywhere" as its subtitle (and that exact description as part of its <title>), and that is basically all you need to know. It's a staple for Unix-y power users on macOS (but supports Linux too, including WSL), and version 1.0 will be ten years old next week. You're not very far off from complaining that Postgres release notes don't explain that Postgres is a database.
If you actually clicked through to their homepage, I imagine you would find the “what does homebrew do” section quite enlightening.
It's a package manager.<p>The second header on the main page:<p>What Does Homebrew Do?<p>Homebrew installs the command-line tools and applications you need across macOS, Linux and WSL.
It’s a package manager for Mac.
It’s just another software package manager for installing dev tools
Just remember that brew still must never be installed on a machine you need to be able to trust.<p>The supply chain security policy of brew is basically non existent and optimized for low-friction contributions. Think wikipedia. No enforced commit signing, review signing, or multi-party release signing, and thus everything is honor system.<p>Do not put brew anywhere near systems that access production or even on systems used to review production-bound code.
<a href="https://docs.brew.sh/Homebrew-Security-and-Supply-Chain" rel="nofollow">https://docs.brew.sh/Homebrew-Security-and-Supply-Chain</a><p>We take supply chain security very seriously, moreso than many package managers.
How is that any different from PyPI, npm, cargo, etc?
These release notes appear generated by LLM, with the usual extreme verbosity. Wish a human had done a better job editing these.
An LLM helped with the initial tedium of rewording 1000s of PRs from maintainer centric language into user centric language. It was reviewed and edited (by a human) probably >20 times. Go look at the PR if you don’t believe me.<p>I used to do these all by hand and they took me multiple hours. I still probably spent over an hour, including a bunch of time on a family holiday today. Comments like this make me wonder why I bother sometimes.
FWIW, I also got a bit of an LLM vibe skimming the changelog. In particular, the section about Linux sandboxing jumped out at me:<p>> Status in 7.0.0: kernels without Landlock continue working without Linux sandboxing in the less secure pre-6.0.0 configuration; brew doctor reports missing protection as an advisory.<p>> No replacement opt-out; unavailable Landlock remains advisory.<p>With that being said, I can kind of see why "eliminate all traces of AI writing styles" wouldn't be a goal of the review/editing process. I've been out of the Apple ecosystem for a while, but I imagine Homebrew has enthusiastically adopted LLMs for a lot of tasks. If the public communications reflect that, then people who don't like LLMs can bounce immediately, rather than getting invested in the project and feeling betrayed later.