415168 - Disable <a ping> for Firefox 3

Status: RESOLVED FIXED

(Core :: DOM: Core & HTML, defect)

Description

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

Attachment: Turn it off

The spec has changed under us so we're now at a state of no UI, no good proposal for a UI, and no compliant implementation. Let's just disable it and see if we can get a good implementation for next firefox instead.

(note that this change is firefox specific. It's already disabled by default for all other gecko apps, and they will still be able to enable it if they so wish)

Comment 1

[Vladimir Vukicevic [:vlad] [:vladv] (needinfo me, slow to respond)]

Comment on attachment 300754 [details] [diff] [review]
Turn it off

Jonas was supposed to ask beltzner for r/a+, setting approval request here.

Comment 2

[Mike Beltzner [:beltzner, not reading bugmail]]

Comment on attachment 300754 [details] [diff] [review]
Turn it off

a=beltzner, alas!

Comment 3

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

Checked in

Comment 4

[Reed Loden [:reed]]

This caused the unit test from bug 375314 to fail. sicking said to just remove <a ping> testing from the test, so I did so. If this ever comes back, the test for <a ping> needs to be re-added.

http://bonsai.mozilla.org/cvsview2.cgi?diff_mode=context&whitespace_mode=show&subdir=mozilla/content/base/test&command=DIFF_FRAMESET&file=test_bug375314.html&rev1=1.1&rev2=1.2&root=/cvsroot

Comment 5

[Peter Gasston]

Google have recently begun using ping to speed up mobile search results: https://plus.google.com/+IlyaGrigorik/posts/fPJNzUf76Nx

Comment 6

[Randell Jesup [:jesup] (needinfo me)]

And also pushing this as a wished-for performance improvement generally, not just on mobile, though it's a win there:
http://calendar.perfplanet.com/2013/browser-wishlist-2013/

I'd assume the argument is that any <a ping= href=...></a> would be replaced by the website that wants to track people following the link with a redirect, and thus there's no true privacy difference here.  If anything, ping support is a privacy win (even with no UI) in that if sites start using ping, then the user *can* stop sending them (and a bigger win as ping use spreads and if UI is added).  If ping isn't supported and the sites use redirects, the user has no option.

Comment 7

[Jonas Sicking (:sicking) No longer reading bugmail consistently]

Me and Doug recently worked with Microsoft and Google to specify a "beacon" spec which should be a more general, easier to implement and less controversial feature.

Spec is here: https://dvcs.w3.org/hg/webperf/raw-file/tip/specs/Beacon/Overview.html

I think Doug is working on implementing, but I don't know if he has a bug yet.

Comment 8

[Olli Pettay [:smaug][[e-mail]]]

bug 936340

Comment 9

[Brad Lassey [:blassey] (use needinfo?)]

(In reply to Randell Jesup [:jesup] from comment #6)
> And also pushing this as a wished-for performance improvement generally, not
> just on mobile, though it's a win there:
> http://calendar.perfplanet.com/2013/browser-wishlist-2013/
> 
> I'd assume the argument is that any <a ping= href=...></a> would be replaced
> by the website that wants to track people following the link with a
> redirect, and thus there's no true privacy difference here.  If anything,
> ping support is a privacy win (even with no UI) in that if sites start using
> ping, then the user *can* stop sending them (and a bigger win as ping use
> spreads and if UI is added).  If ping isn't supported and the sites use
> redirects, the user has no option.

I generally agree that at worst this is a wash privacy-wise.

Doug, care to expand on what Jonas said?

Comment 10

[Doug Turner (:dougt)]

yeah, we're working on it.

Comment 11

[Brad Lassey [:blassey] (use needinfo?)]

(In reply to Doug Turner (:dougt) from comment #10)
> yeah, we're working on it.

was more looking for a bug number where this is being discussed

Comment 12

[Doug Turner (:dougt)]

comment #8.