Skip to content
This repository was archived by the owner on Nov 6, 2023. It is now read-only.

update Jabber.at.xml - #4792

Merged
fuglede merged 1 commit into
EFForg:masterfrom
J0WI:jabber.at
May 15, 2016
Merged

fuglede merged 1 commit into
EFForg:masterfrom
J0WI:jabber.at

Conversation

@J0WI

@J0WI J0WI commented May 14, 2016

Copy link
Copy Markdown
Contributor

No description provided.

@fuglede

fuglede commented May 15, 2016

Copy link
Copy Markdown
Contributor

A wildcard for the subdomains would also work here, given that the thing is already HSTS preloaded in some browsers.

@fuglede
fuglede merged commit 483ed46 into EFForg:master May 15, 2016
@J0WI
J0WI deleted the jabber.at branch May 15, 2016 11:48
@J0WI

J0WI commented May 15, 2016

Copy link
Copy Markdown
Contributor Author

Oh sure, we really need a test for that.

@fuglede

fuglede commented May 15, 2016

Copy link
Copy Markdown
Contributor

For catching preloadables, you mean? And if so, what should it do? In this case, preloadability has not yet descended all the way through to ESR, so current consensus appears to be that it should be treated on the same footing as all other rulesets.

What I could see being useful would be to have a setup similar to what we use on Wikipedia where general maintenance scripts take care of things that we like but probably don't want to check on every single change; I'm thinking stuff like killing completely preloaded target hosts (which I'm still not certain I'm personally in favor of), enforcing a particular ruleset style, running https-everywhere-checker, etc.

@J0WI

J0WI commented May 15, 2016

Copy link
Copy Markdown
Contributor Author

The test should check if a host sends the HSTS preload attribute and if the hosts is in the list.
The list gets uplifted to ESR and TOR, but the list on SSL Labs is not up to date (TOR lacks a stable branch where the list can be pulled from automatically).

So we both agree that we don't want to maintain HSTS rulesets. A script could generate rules automatically during the build process out of the list (so they don't need to be in the repo).
But I'm also fine if they would just be deleted. Because the preloaded list of browsers if has a much better performance than HTTPS Everywhere and is not affected by false mixed content blocking issues. It's also well maintained by browser vendors and webmasters.

@fuglede

fuglede commented May 15, 2016

Copy link
Copy Markdown
Contributor

But how would we check if a host is in all of the relevant lists (without running in to the same problem as SSL Labs)?

@J0WI

J0WI commented May 16, 2016

Copy link
Copy Markdown
Contributor Author

I think the list of Chrome and Firefox (if there are any differences) are enough, because all other are based on them.
I don't know if the list of Edge is public available.

@J0WI

J0WI commented May 16, 2016

Copy link
Copy Markdown
Contributor Author

#4462 looks interesting btw.

@fuglede

fuglede commented May 16, 2016

Copy link
Copy Markdown
Contributor

Based on them, but as I see that you are already aware, only actually including them at some point in the (potentially far) future.

@J0WI J0WI mentioned this pull request May 16, 2016
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants