Tell HN: Chromium forks for legacy platforms are disappearing from GitHub

Even though all of this should be perfectly legal and follows TOS, GitHub really isn’t the place to host anything possibly damaging to or annoying big tech.

You can add push URLs to your origin branch such that `git push` automatically mirrors your code on multiple remotes: `git remote set-url --add --push origin git://another/repo.git`. If you maintain an open source repository, I would encourage you to have mirrors across gitlab, codeberg, and/or a self hosted gitea instance.

I'm certainly not against mirrors. But—

> GitHub really isn’t the place to host anything possibly damaging to or annoying big tech.

Why would Google give a crap if a handful of people futzing around with ancient versions of Windows and macOS want to run up-to-date Chromium browsers?

If you wanted to be super conspiratorial, you could point to Supermium's statement[1] that they plan to retain Manifest V2 support. But that merely puts them in the same category as Brave and Firefox.

At the same time, I don't know how else to explain what is going on!

---

1: https://win32subsystem.live/supermium/#:~:text=Manifest%20V2....

Conspiracy hat: Google is fighting ad blockers via the new v3 manifest to keep the YouTube milk cow going. They will soon be rejecting v2 plugins and probably expect some flight to chromium.

Maybe take the hat off for a minute and consider how they'd possibly retroactively revoke the (non-revokable) open source licenses that all Chromium versions published to date have been licensed under.

Then ask yourself what the benefit of all that – already impossible – endeavor would be, given that Firefox also exists and isn't deprecating Manifest V2.

Finally, ask yourself which actual paths Google could take to still eventually force MFv3 on most web users without resorting to any such drastic and publicly-visible measures.

See? You didn't really need the hat!

they don’t have to revoke the license, they just add this for new versions

> what the benefit of all that endeavor would be

Breaking all adblockers for good, ensuring revenue. Firefox has almost no marketshare so it’s more or less irrelevant.

> which actual paths Google could take to still eventually force MFv3

Next steps are to break the web more and more for non-chromium browsers. Already many sites that don’t work well with FF/Safari. There will be reasonable reasons to do that (eg protect the kids) but the end result will be 1 browser dominance and ads tattooed into your eyeballs - mark my words!!1

> Next steps are to break the web more and more for non-chromium browsers

What stops people from switching to Brave, or some light fork that's basically "Chromium + Manifest V2"?

Google would need something akin to Web Attestation. And yes they did try that—so if you're going to put your conspiracy hat on I'd direct your attention there.

The ecosystem for V2 plugins will die when Chrome no longer supports them.

Stagnation is the plan, not fullfrontal assault. Programmers tend to want people using their products more than they want anything else.

Im keen to believe this is a genuine mistake on either GitHub or Google’s end. Perhaps some automated system they’re testing out.

https://forums.macrumors.com/threads/chromium-legacy-google-...

Looks like this is a step towards Google DRMing the internet.

This is bad.

How would Google push DRM by filing false [1] copyright claims against various Chromium forks, which presumably wouldn't even contain any hypothetical DRM keys?

Regular old Chrome does in fact include a Widevine DRM binary blob, but Chromium doesn't contain the source for it.

If Google really wanted to push something like browser attestation (like they potentially did with their Web Environment Integrity proposal), almost by definition that part couldn't be open sourced either and would not be part of Chromium.

[1] Chromium is licensed under various open source licenses that Google can't retroactively revoke. Even for code that they own the full copyright on, they can at most release future versions under a closed license; for GPL-ed dependencies, that wouldn't work at all (not sure if Chromium has any).

Why couldn't browser attestation code be OSS?

The code could be (but often isn't, to frustrate circumvention), but the signature keys can't, just like with all DRM.

What's the point of playing devil's advocate, or giving any benefit of doubt to Google? They have shown repeatedly that they intend to control the internet. Just because we can't see why yet, it's not far fetched to assume that this might be a step in that direction.

In the thread you posted:

"Bluebox doesn't know why his Github account disappeared, but it was not intentional and he has contacted Github support!"

"the Supermium GitHub repository is currently unavailable. Efforts are being made to restore the repository"

If this is true, it could be because of using Google API Keys in the software.

API keys are a good theory but as far as I can tell, at least in Chromium Legacy they're set up correctly per Google's policies. You can sign into a Google Account in the released builds of Chromium (so they are embedded in the binary) but when I compile the source myself the keys aren't there, so they haven't been disclosed in the source.

https://github.com/gdamdam/iagitup

It's a little late for that, sadly.

I'm not sure if GitHub is the right place for chromium mirrors. When you upload to GitHub, you are agreeing that the code is licensed to GitHub under certain conditions (in addition to the license specified in the repository).

Although a potential issue generally, it's unrelated here.

If it's already open source, GitHub can almost (depending on license) use it as they see fit anyway.

Isn't GitHub owned by Microsoft? Isn't Microsoft using GitHub to train AI? Doesn't Google also have AI ambitions?

I think it plausible that we are seeing the shadows cast by competitive maneuvering between two industry heavy-weights.

Working for me now, so I guess they got it sorted out.

Somebody needs to upload their local all over the internet to gitlab, codeberg, anywhere else.

Their local what?

Copy of the source code?

The beauty of git (everyone has a full copy of the repo). Although this being an installable browser targeted more towards end users I wouldn’t trust a repo from anyone other than the official maintainers.

Both up again. Probably an automatic spam filter, shadowbanning both users

It's notable that Chromium is an especially large repo, and it seems has been special cased in githubs repo size limits.

It's possible this is GitHub just enforcing their size limit

Could this be tied to Google's push to force everyone to use Manifest 3, and nerf Add Blockers?

How would it? Google can't retroactively un-license the Chromium source, which is under various open source licenses that I'm almost certain all include at a minimum the right to host a fork somewhere.

Additionally, if you are concerned about MF3, your best bet isn't some outdated Chromium fork – it's Firefox.

Just to be clear, Chromium Legacy and Supermium aren't outdated, that's what makes them great. They work on outdated operating systems, but they track up-to-date Chromium.

That is great!

I kind of assumed as much, but unfortunately that also most likely means that it'll lose Manifest V2 support at some point (not that upstream Chrome will keep it).

That seems to be the concern of some people commenting here, and I wanted to highlight that the two are largely orthogonal as far as I can tell.

Supermium has stated that they plan to retain Manifest V2. I agree with you for Chromium Legacy.

Google can't take back the source but they can make distribution of binaries hard? Most Windows users are very dependent on pre-built binaries, for example.

And doesn't Google have leverage over Mozilla by way of their funding?