Nanobrew: The fastest macOS package manager compatible with brew
127 points
7 hours ago
| 16 comments
| nanobrew.trilok.ai
| HN
mikemcquaid
5 hours ago
[-]
If it doesn’t ever execute Ruby: it cannot be compatible with Homebrew. “Compatible” is doing a bit of work here when it also means “implicitly relies on Homebrew’s CDN, CI, packaging infrastructure and maintainers who keep all this running”.

There’s a new vibe coded Homebrew frontend with partial compatibility and improved speed every few weeks.

Homebrew is working on an official Rust frontend that will actually have full compatibility. Hopefully this will help share effort across the wider ecosystem.

reply
runjake
2 hours ago
[-]
Context for those unaware: the commenter, mikemcquaid, is the project lead for Homebrew.
reply
Xunjin
1 hour ago
[-]
Thank you, his arguments totally makes sense, only the part that makes me icky is:

> There’s a new vibe coded Homebrew frontend with partial compatibility and improved speed every few weeks.

People are free and probably do this because it is slow. Alternatives often are not a bad thing.

reply
alwillis
1 hour ago
[-]
Since I enabled HOMEBREW_DOWNLOAD_CONCURRENCY, downloads have improved for me to the point where download speed is no longer an issue.
reply
runjake
1 hour ago
[-]
Point noted! I took it as a tongue-in-cheek phrasing of "agentically coded". Hopefully, that's right.
reply
pxc
4 hours ago
[-]
It is really coll that Homebrew provides a comprehensive enough JSON API to let people build on Homebrew in useful ways without directly running Ruby, despite everything being built in a Ruby DSL. That really does seem like a "best of both worlds" deal, and it's cool that alternative clients can take advantage of that.

I didn't know about the pending, official Rust frontend! That's very interesting.

reply
SOLAR_FIELDS
4 hours ago
[-]
Wow they are finally getting away from Ruby? Awesome. The speed will be a nice boon
reply
petcat
3 hours ago
[-]
Yeah I don't know why people are saying that speed doesn't matter. I use Homebrew and it is slow.

It's like yum vs apt in the Linux world. APT (C++) is fast and yum (Python) was slow. Both work fine, but yum would just add a few seconds, or a minute, of little frustrations multiple times a day. It adds up. They finally fixed it with dnf (C++) and now yum is deprecated.

Glad to hear a Rust rewrite is coming to Homebrew soon.

reply
mhurron
31 minutes ago
[-]
> Yeah I don't know why people are saying that speed doesn't matter. I use Homebrew and it is slow

Because how often are you running it where it's not anything but a opportunity to take a little breather in a day? And I do mean little, the speedups being touted here are seconds.

I have the same response to the obsession with boot times, how often are you booting your machine where it is actually impacting anything? How often are you installing packages?

Do you have the same time revulsion for going to the bathroom? Or getting a glass of water? or basically everything in life that isn't instantaneous?

reply
kelvie
1 hour ago
[-]
One of the reasons I switched to arch from debian based distros was precisely how much faster pacman was compared to APT -- system updates shouldn't take over half an hour when I have a (multi)gigabit connection and an SSD.

It was mostly precipitated by when containers came in and I was honestly shocked at how fast apk installs packages on alpine compared to my Ubuntu boxes (using apt)

reply
akdev1l
39 minutes ago
[-]
pacman is faster simply because it does less things and it supports less use cases.

For example pacman does not need to validate the system for partial upgrades because those are unsupported on Arch and if the system is borked then it’s yours to fix.

reply
akdev1l
43 minutes ago
[-]
yum was slow not because of python but because of the algorithm used to solve dependencies

Anyway the python program would call into libsolv which is implemented in C.

dnf5 is much faster but the authors of the program credit the algorithmic changes and not because it is written in C++

dnf < 5 was still performing similarly to yum (and it was also implemented in python)

reply
novok
1 hour ago
[-]
Ruby doesn't have to be the slow part, bazel uses starlark which is mostly python and it's very fast.
reply
pxc
54 minutes ago
[-]
I would guess this change builds on the existing json endpoints for package metadata but that the Ruby DSL is remaining intact.

I think how to marry the Ruby formulas and a Rust frontend is something the Homebrew devs can figure out and I'm interested to see where it goes, but I don't really care whether Ruby "goes away" from Homebrew in the end or not. It's a lovely language, so if they can keep it for their DSL but improve client performance I think that's great.

reply
atonse
3 hours ago
[-]
Heyyyy, who are you to tell us what is and isn't compatible with homebrew?

(Just kidding, thank you for creating homebrew and your continued work on it!)

reply
orf
3 hours ago
[-]
reply
tfrancisl
3 hours ago
[-]
I appreciate the push for an official rust frontend. I've personally been migrating (slowly) to using nix to manage my Mac's software, but there are a ton of limitations which lead me to rely on homebrew anyway. The speed ups will be appreciated.
reply
halapro
3 hours ago
[-]
Makes no sense, the wording suggests it can use Homebrew's backend, not that it's a complete alternative to Homebrew. Nobody is confused about that.
reply
akdev1l
37 minutes ago
[-]
The recipes for building and installing homebrew packages are written in Ruby

You cannot really be compatible with this unless you run the Ruby as the install scripts could do whatever arbitrary computations

In reality most recipes contain a simple declarative config but nothing stops you from doing Ruby in there.

Hence to achieve total compatibility one would need to run Ruby

reply
nozzlegear
3 hours ago
[-]
I mean, I'm confused about it. The nanobrew homepage says this:

> nanobrew

> The fastest macOS package manager. Written in Zig.

> 3.5ms warm install time

> 7,000x faster than Homebrew · faster than echo

It presents itself as an alternative to Homebrew.

reply
halapro
2 hours ago
[-]
There are many such examples for npm as well: many "compatible" managers, one registry.
reply
nozzlegear
2 hours ago
[-]
Sorry, examples of what? Package managers that present themselves as replacements for other package managers? Or package managers that aren't compatible with the registry they're supposed to be compatible with? Your use of scare quotes is confusing.
reply
0x457
1 hour ago
[-]
pnmp, npm, yard all have different lockfiles, all use the same registry format (and the same registry itself), all try to stay compatible in other ways.

You won't be having situation where one uses yarn and someone uses pnpm on the same project tho.

reply
boobsbr
3 hours ago
[-]
Please, don't remove bottles and casks that are blocked by Gatekeeper. :˜(
reply
drob518
4 hours ago
[-]
This feels like a solution looking for a problem. I have a couple hundred brew packages on my system and I’ve never sat there thinking “If this was only 2 seconds faster…” while doing an update. I’m sure the Homebrew folks could mine this for a few ideas of how to further optimize brew, but I don’t think I’ll be adopting it anytime soon. Compatibility is more important than speed in this case.
reply
swiftcoder
4 hours ago
[-]
> I’ve never sat there thinking “If this was only 2 seconds faster…” while doing an update

I definitely have thought something along those lines (mostly when I go to install a small tool, and get hit with 20 minutes of auto-updates first).

Pretty sure I also will not be adopting this particular solution, however

reply
bombcar
4 hours ago
[-]
I've never thought "only 2 seconds faster" - I've certainly thought "why is this taking half the time it takes Gentoo to recompile an entire server".
reply
joshstrange
4 hours ago
[-]
But you can turn that behavior off, IIRC it tells you the environment variable to set if you don’t want it to do that every time it runs.

I agree it’s annoying, but I haven’t turned it off because it’s only annoying because I’m not keeping my computer (brew packages) up-to-date normally (aka, it’s my own fault).

reply
swiftcoder
3 hours ago
[-]
I'd be much happier if it were on a background job, than arbitrarily running when I invoke a command
reply
slackfan
3 hours ago
[-]
Terrible default behavior is a great reason to abandon a software package.
reply
SOLAR_FIELDS
4 hours ago
[-]
FWIW this seems to have improved in recent years. Back in the dark times of non parallelized downloads I would purposefully wait to end of day and fire the thing off before leaving
reply
rconti
1 hour ago
[-]
I've been a lightweight homebrew user for many, many, many years now. I just use it to download or update a thing I need, once every 3-6mo.

It constantly blows my mind how insanely long it takes just to do a few simple things on the fastest hardware I've ever owned in my life.

reply
pxc
4 hours ago
[-]
If you use the Homebrew module for Nix-Darwin, running `brew` against the generated brewfile becomes the slowest part of a `darwin-rebuild switch` by far. In the fast cases, it turns something that could take 1 second into something that takes 10, which is definitely annoying when running that command is part of your process for configuration changes even when you don't update anything. Homebrew no-ops against an unchanging Brewfile are really slow.
reply
ziml77
3 hours ago
[-]
Agreed here. The speed bottleneck I run into is simply that there's often a lot of packages that need updating, so there's a lot to download. And if anything needs to be compiled from source then the time that takes will dominate (though I think everything I currently run is thankfully pre-built)
reply
noahbp
3 hours ago
[-]
The same criticism has been said of Deno and Pnpm and bun, and yet, despite all these years since their respective releases, node and npm remain slower than all three options.
reply
fleebee
2 hours ago
[-]
Yeah, but do they work? Last time I gave bun a chance their runtime had serious issues with frequent crashes. Faster package installation or spin-up time is meaningless if it comes at the cost of stability and compatibility.
reply
alwillis
1 hour ago
[-]
bun is my go to for npm packages; it’s so much better and faster than npm, it’s not funny.

Never had any issues.

reply
never_inline
3 hours ago
[-]
Well, pnpm solves the storage issue, which is a more pressing reason to use it. (I don't know about deno/bun)
reply
dilap
4 hours ago
[-]
Horses for courses, but I've stopped using brew 'cuz it's too slow, so this might bring me back!

Edit: no, it won't...

reply
drob518
4 hours ago
[-]
Agreed on horses for courses. Different people have different tolerances. And yea, all things being equal, faster is better, but they are almost never equal. If you don’t mind me asking, what does “too slow” mean for you in this context? Do you have a particularly complex setup? And what do you use now as an alternative and how has that impacted the update speed?
reply
dilap
58 minutes ago
[-]
I wish I could remember the details -- I know I got annoyed with things being slow and when I got a new computer decided to go the no-homebrew route. I'm using nix, and it seems fine so far, but I also really don't understand it at all, which is a little concerning. :-)
reply
motorpixel
3 hours ago
[-]
If I have to deal with even the mention of another package manager in the cross-platform dev ecosystem I am going to snap
reply
staticassertion
2 hours ago
[-]
I've wanted brew to be faster. It would be a nice QoL for me.
reply
mproud
4 hours ago
[-]
My brew update/upgrade takes forever
reply
password4321
2 hours ago
[-]
reply
ryandrake
3 hours ago
[-]
What would be great is a Homebrew-compatible system that doesn't cut off support for older machines. I have a 3.8 GHz Quad core i5 iMac that still crushes, yet Homebrew has determined that I'm just too old and icky[1] to work with anymore. I had to move over to MacPorts, which is surprisingly nice, but I still miss brew.

Yea, I know. It's open source. They can do what they want. Still sucks.

1: https://docs.brew.sh/Support-Tiers

reply
happyopossum
2 hours ago
[-]
To be fair, Apple stopped providing security fixes for Mojave ~4+ years ago, and there have been 7 or 8 new os releases since then…

I don’t think it’s reasonable to expect an open source project to support everything

reply
gabagool
1 hour ago
[-]
I agree in principle but Homebrew only supports the latest 3 versions of macOS. Right now Ventura 13 which came out in October 2022 is unsupported.
reply
dewey
1 hour ago
[-]
I still think that's entirely fair for a power user tool like homebrew. With the upgrade rates of macOS that probably means that's 98% of the users would be covered. Expecting an open source project to accept bug requests from a bigger variety of versions that then would need test devices on these versions to replicate issues sounds unrealistic. Bigger companies, or Apple itself I would hold to much higher standards when it comes to that.
reply
bsagdiyev
36 minutes ago
[-]
> power user tool like homebrew.

That makes no sense then. A power user may still want to run older OS versions for a reason. Take the training wheels off it and then it'll be a power user tool.

reply
dewey
33 minutes ago
[-]
> A power user may still want to run older OS versions for a reason.

No doubt there are edge cases like that, but I don't fault a project for not catering to the < 1% of users who would fall into that bucket and would probably be the ones that cause trickier support cases. These would maybe also be the user that could just install it without homebrew then, it's not like homebrew is the only way to install software.

reply
ksherlock
36 minutes ago
[-]
brew used to say, more or less, "This OS is old and unsupported. Don't submit bug reports. If you have problems, too bad. If you submit a PR to fix something, we might merge it". Fair enough, right? Now it just says, "Go fuck yourself, grandpa."
reply
yabutlivnWoods
1 hour ago
[-]
Run Linux on it. Apple has cut that OS off anyway. Would be safer security wise to have an OS that's updated
reply
password4321
2 hours ago
[-]
Yes MacPorts is the way. I switched after a new MacOS release meant mine was too old - brew update uninstalled a bunch of stuff I had been using then it stopped and let me know.

There's also https://github.com/dortania/OpenCore-Legacy-Patcher for the adventurous.

reply
maxkfranz
2 hours ago
[-]
You could use the OpenCode legacy patcher to upgrade to v15/Sequoia: https://dortania.github.io/OpenCore-Legacy-Patcher/
reply
ryandrake
1 hour ago
[-]
Sure, but this might win you a couple of years max. Homebrew's "Support Tiers" page, which I linked, also addresses OCLP users, going so far as to specify a minimum Intel architecture. So, even if you use OCLP to allow support for newer OS versions, eventually your CPU architecture will be too old and you're back in Tier 3.

Also, the writing is on the wall: Ultimately, Homebrew will be ARM-only, once Apple's legacy support becomes ARM-only. At which point it's game-over for Intel Macs.

Homebrew solves the "availability of software" problem in the Mac ecosystem, but it does not solve the "Need to stay on the new hardware treadmill" problem.

reply
themadsens
59 minutes ago
[-]
I naively assumed it would work on the already installed homebrew packages. No such luck.

After installing, 'nb list' and thus eg. 'nb outdated' will yield the empty list! I have absolutely no use for a competing homebrew installation that is mostly compatible ..

reply
chuckadams
5 hours ago
[-]
This might be a good thing for homebrew to adopt for the download/install process, but if it doesn't include a ruby interpreter, I have a hard time seeing how it's going to be compatible with anything but searching and installing bottles. I install most of my packages from a Brewfile, which itself is Ruby code.
reply
alwillis
1 hour ago
[-]
> I install most of my packages from a Brewfile, which itself is Ruby code.

Same. Whatever happens, the new version should support Brewfile.

reply
luizfelberti
4 hours ago
[-]
It might be good to explain how this differs from zerobrew [0], which is trying to accomplish the same thing

[0] https://github.com/lucasgelfond/zerobrew

reply
Alifatisk
2 hours ago
[-]
Zerobrew looks mature, I'll check it out.

Btw, I noted this:

> Zerobrew is experimental. We recommend running it alongside Homebrew rather than as a replacement, and do not recommend purging homebrew and replacing it with zerobrew unless you are absolutely sure about the implications of doing so.

So I guess its fine to run this alongside Homebrew and they don't conflict.

reply
tomComb
3 hours ago
[-]
And zerobrew, like the original Homebrew, is compatible with Linux.

It appears that Nanobrew is not.

I care about the light-weight efficiency of these new native code variants much more when I want to use brew on some little Linux container or VM or CI, than I do for my macOS development machine.

reply
maxloh
3 hours ago
[-]
How does this work? AFAIK Homebrew formulae are written in Ruby [0].

Do they use some kind of Ruby parser to parse formulae?

[0]: https://github.com/Homebrew/homebrew-core/blob/26-tahoe/Form...

reply
fny
2 hours ago
[-]
It uses the Homebrew API and uses its own dependency resolver and linker to pull Homebrew's precompiled packages.
reply
manlymuppet
3 hours ago
[-]
If we get the Bun-ification of every package manager and language ecosystem that would be an awesome thing. This is a great trend.
reply
alsetmusic
4 hours ago
[-]
I'm not a Python dev, but I appreciate the motivation uv has inspired across other package managers. I tried another brew replacement called zerobrew last month. It installed packages to a different directory from homebrew, so I didn't actually test drive after seeing that. Regardless, I look forward to the competition pushing mainstream tools to improve their performance.
reply
12_throw_away
1 hour ago
[-]
So, A) to what extent is this vibe coded? And B) what is "trilok.ai" where you download it from?
reply
tantalor
3 hours ago
[-]
And why does speed matter in this case?
reply
an0malous
5 hours ago
[-]
Does it reinstall postgres for every package install?
reply
ericcholis
5 hours ago
[-]
HOMEBREW_NO_AUTO_UPDATE=1 will disable this (annoying) behavior. Set it in your bashrc or zshrc.
reply
mitchitized
5 hours ago
[-]
(report card for an0malous): "Does not play nice with other students."
reply
an0malous
4 hours ago
[-]
It's true :')
reply
kassadin
4 hours ago
[-]
Do you choose compatibility or speed?

nb info --cask codex-app

nb: formula '--cask' not found

nb: formula 'codex-app' not found

reply
pxc
4 hours ago
[-]
I've been looking for something like this, especially to use only with casks now that Homebrew has removed support for not adding the quarantine bit. Looking forward to giving it a try!
reply
marksully
2 hours ago
[-]
what happens if I test this tool by installing some packages and then remove (the tool)? will I still be able to use Homebrew to manage these new packages?
reply
Onavo
2 hours ago
[-]
The current version of brew has a flaw where the installer can't install isolated dependency trees in a sterile manner. If you have packages A, B, C, and D that all have updates, and assuming A,B,C depend on each other and come out to a total of say 1MB, and D is 1000MB, brew works in a MapReduce manner where it will attempt to finish downloading everything in parallel (even though the real bottleneck is D) before doing any installation.

Since the first 3 has no dependency on D, a better way would be to install them in parallel while D is still downloading.

reply