I might be OK with running this bit of their code on my computer, but it's not an open bar for dropping anything executable onto my machine at will.
Don't install their apps then. There are lots of alternatives out there.
If you actually look at the competition, most of them do not have LAN sync or block sync (OneDrive, Google Drive), they do not have proper support for revisions (Resilio Sync), or are hard to set up for most people (Syncthing).
[1] So that if I change a small piece of a large file, it does not upload the complete file again.
seafile.com
It's open-source (Chromium project) - just compile it yourself?
Once you've run somebody else's binary, you're already conferring significant trust on them. And I might be biased, but the Chrome team has shown themselves to be very vigilant with security, and generally on-the-ball on all of these things. (Memory issues aside...although I'm not going to get into that...lol).
Many newer projects do this - e.g. Atom from Github auto-updates.
VS Code from Microsoft does this - I think it's awesome! =)
This is the "you agreed to a date, that means you agreed to go all the way" argument.
Even when a someone extends trust or tentative trust, that doesn't mean they should be expected to give up the right to maintain control of that choice at each successive moment in time.
EFF strongly discourages auto updates that can't be turned off, because they can be used for enabling DRM and curtailing freedom.
If you remove it, and then try to run any Google Software, it will first lie to you and say that it needs you to authenticate so it can "function" properly. If you refuse, it will try to install it's code in ~/Library instead. If you intercept that, it will stop asking and work properly.
Google software does not depend on its updater in any way. There is no technical reason to keep asking you to install it. They just don't respect your agency. And if you say no to root installation, they will settle for just a user installation.
A lot of software that asks for authentication on first run is doing it solely to install it's helpers.
Not just is it creepy asf and wrong, but sometimes features will be removed or changed without my consent.
Software that was free when you downloaded it just got "upgraded" to shareware. Or maybe it just irreversibly converted all of its saved data to the new format that isn't backwards compatible.
You would never have agreed to install that update if you actually knew what it was going to do. That makes it dishonest.
My impression was that at least in Debian based systems (Ubuntu and others), it updates with apt now?
EDIT: Sorry if this sounded accusative. Wasn't meant that way.
Would be fun trying to autoupdate deb packages though.
But unless you audited the chrome code (with a team of 1000 experts) you have already trusted them, both to write quality code and to prevent malicious code sneaking in. Why not trust them to patch your browser against malware?
Feel free to contribute to ungoogled chromium: https://github.com/Eloston/ungoogled-chromium
The Dropbox sync service, on the other hand, is literally just the client half of Dropbox's own proprietary backend. If Dropbox makes an ABI-incompatible change to their sync protocol on the server side, the non-updated client will just break until you update.
Wanting to disable updates to such a service, would be like wanting to load the old version of a webapp SPA whenever you visit the site. What's the point? It's not going to (have been designed to) work that way.
So... don't break ABI compatibility?