Self-host Firefox Sync

[nose_gnome]

It would be great if you could self host your own sync server for Firefox, so you don't need a Firefox Account.

It could either work by only allowing users to sync when connected to the LAN the self-hosted sync server is running on, or accessible anywhere by setting up port-forwarding.

This would give the following advantages over the current system:

• Better security and privacy as our data isn't leaving any of our device.
• Allows syncing while the internet is down but LAN is still working

This wouldn't be a replacement for the current system but should be provided as and additional option.

[Jon]

Thanks for submitting an idea to the Mozilla Connect community! Your idea is now open to votes (aka kudos) and comments.

[nose_gnome]

It could also be an idea to create an "app" for Netgear's ReadyNas, allowing you to host Firefox Sync off a Nasbox

[nclm]

I seem to remember that this used to be possible when syncing was introduced. But is it possible that support for self hosting was removed?

[Stopi]

+1 to get a self-hosted solution, fully working without any need to reach Mozilla's servers.
This requirements are a REAL step in the direction of keeping our data private.

I mean: one of Firefox's mission is to protect our privacy, right?
https://www.mozilla.org/en-US/firefox/privacy/

And what data should be kept more private than our browsing experience (history, password, etc.)?

Various documentation can be found on the web:

• https://mozilla-services.readthedocs.io/en/latest/howtos/run-sync-1.5.html
• https://mozilla.github.io/ecosystem-platform/
• https://github.com/mozilla/fxa
• https://github.com/mozilla-services/syncstorage-rs/
• https://news.ycombinator.com/item?id=34674569

The problem is they're featuring a lot of moving parts, only partially documented and some of it are clearly outdated.

@nose_gnomethe ReadyNas idea is very niche and should be implemented by Netgear, not by Mozilla.
Unless suddenly a big chunk of the users decided to adopt this NAS over any other.

[KZ]

Every six months or so I do a quick check on the state of self-hosted FF sync/accounts, been doing so  for years. There's always something to learn and sometimes signs of recurrent interest, but ultimately never any joy. I realize this is partly by design.

I also struggle to put into convincing words why it's important, for us & for anyone, to have the option to self-host infrastructure instead of using Mozilla.

At the core, we already have network identities--actually, too many of them--which are less optional than Mozilla's identities. It's like Mozilla and our other vendor, who shall remain nameless but we all know who it is, are engaged in this epic game of chicken, to see who will dominate identity workflow. That makes us the road, scuffed up with tire tracks. Not exactly the bright future of computing we all thought would happen.

Our existing identities mirror our policies, which Mozilla's cannot, that a quick check can reveal. We have enough "shadow I.T." of people signing up for random services, but then coming to I.T. begging for help with something we didn't know existed.

So-and-so has left on bad terms and holds their work resources hostage (or absconds with them) within their Mozilla identity, and leadership needs them now. Which is easier? 1) 'UPDATE foo SET pw = "";' then log in and set a new password, or 2) a ticket/authentication/reset cycle with Mozilla in an age when live people in support is too expensive?

In other contexts, like web or components or languages, competing vendors strive to dominate a space and then arrive at a grudging detente, converging on a vaguely neutral standard. I wish I understood how that process worked, how to fast-forward past the winnowing period to something we can explain to our users that we won't have to rename or requalify or reexplain in six months' (or six weeks') time.