413910 - Enable "Prevent sites from changing, moving or resizing windows" by default

Status: RESOLVED FIXED

(Camino Graveyard :: Annoyance Blocking, defect)

Description

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

[2:10pm] pinkerton: i think we should also default to "don't let sites change window sizes" on
[2:10pm] ss: I'd agree with that.
[2:11pm] ardissone|away: me too
[2:11pm] pinkerton: that's so totally useful and safari doesn't, and it's hidden away
[2:11pm] • ardissone|away tries to remember what sites it breaks

We have at least one old closed bug about a site that opens a tiny window, inserts an image, and then resizes the window to fit the image.  I seem to recall maybe one other site I've run across that is broken with this enabled, but I'm guessing not many major sites rely on those features.

Comment 1

[philippe (part-time)]

I don't remember ever running into problems with this, and I have it enabled since whenever that was implemented (that is, Mozilla 1.0 or something ?).

Comment 2

[Chris Lawson (gone)]

I remember running into lots of annoying problems with it *off*, and zero with it on, so I'd be in favour of this as well.

Comment 3

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

Confirming per the meeting; we'll at least do this on the trunk.  mento, do you have any strong opinion about doing this branch-and-trunk vs. trunk-only?  Stuart expressed concern at the meeting we wouldn't get enough exposure before 1.6.

Comment 4

[Mark Mentovai]

I've run with this turned on ever since it existed, but I understand the exposure thing, so I'm OK either way.  One thing to remember is that alphabeta testers are usually existing users, and so, they usually have existing profiles, and won't be affected by a default change no matter when we roll it out.

Comment 5

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

(In reply to comment #4)
> testers are usually existing users, and so, they usually have existing
> profiles, and won't be affected by a default change no matter when we roll it
> out.

Since prefs.js in 90% of the cases is just a diff against all.js/all-camino.js, the only users who wouldn't be moved are people who have a differing value set--and in this case (since it's a bool), they have the value we want already.  I'll verify this later with a test profile, but I'm sure that's what will happen here ;)

Comment 6

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

Comment 5 accurately describes what happens (and if someone already has the pref enabled, those entries are removed from their prefs.js, since their setting becomes the default).

There's now a testcase for move and resize in the URL field.

Comment 7

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

Attachment: Like so

Comment 8

[Chris Lawson (gone)]

Comment on attachment 300811 [details] [diff] [review]
Like so

Looks good to me.

Comment 9

[Mark Mentovai]

Comment on attachment 300811 [details] [diff] [review]
Like so

I'm on a trip and haven't been on irc for a couple of days, so OK, as long as there's been an in-channel discussion of the above concerns.

Comment 10

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

[1:01pm] pinkerton: i say let's just do it
[...]
[1:40pm] ardissone|away: hmm
[1:40pm] ardissone|away: will flipping that pref affect any tinderbox tests?
[1:40pm] ardissone|away: as in, make them fail?
[1:41pm] pinkerton: good q
[1:42pm] ardissone|away: we can flip it off for the tinderboxen somewhere if so

I'll land this later when I can watch closely and back out if we need to modify the tinderbox test config.

Comment 11

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

Checked in on branch and trunk; I'll monitor boxset for bizarre T* results or orangeness and back out if we'll need to twiddle things in the tinderbox.

Comment 12

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

No orange on boxset.

We may or may not have experienced a *very* slight Tdhtml increase:

Before:
833,829,833,833,824 = 830.4 avg
After:
837,827,834,837,830 = 833.0 avg

That works out to approx 0.31% regression, or noise ;)