<html>
<head>
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<div class="moz-cite-prefix">On 05/02/2015 14:14,
<a class="moz-txt-link-abbreviated" href="mailto:teoteoteoteo@virgilio.it">teoteoteoteo@virgilio.it</a> wrote:<br>
</div>
<blockquote cite="mid:14b5a174998.teoteoteoteo@virgilio.it"
type="cite">Considering the <span class="bz_comment_user"><span
class="vcard vcard_159069"><a moz-do-not-send="true"
class="email" href="mailto:gijskruitbosch+bugs@gmail.com"
title=":Gijs Kruitbosch
<gijskruitbosch+bugs@gmail.com>"><span class="fn">Gijs
Kruitbosch</span></a></span></span><span
class="bz_comment_user"><span class="vcard vcard_159069">
suggestion given in this post:<br>
<br>
<a class="moz-txt-link-freetext" href="https://bugzilla.mozilla.org/show_bug.cgi?id=1124176">https://bugzilla.mozilla.org/show_bug.cgi?id=1124176</a><br>
<br>
I'm writing to you to explain why the task '</span></span><span
id="summary_alias_container"><span
id="short_desc_nonedit_display">Add an item in config:about to
stop autoplay of videos inserted by <video> tag' should
be priorer than other Mozillian activities (I know it is not
absolutely right, but I think that a problem feeled since 3
years ago should be priorer than others)!<br>
<br>
1. First of all we can say that almost all sites are starting
to insert HTML5 advertisement videos in their pages that start
as soon as the page is opened.<br>
</span></span></blockquote>
"Almost all sites" is clearly an overstatement. But this in
particular is something where data would be really useful. How many
sites (not in your personal experience, but compared with something
like the Alexa top N sites, etc.) do this? Do they use the autoplay
attribute, and/or how many use JS instead and trigger .play()
themselves ? Can we get data from advertising networks on how many
of these types of video ads they run and what the trends there are?<br>
<blockquote cite="mid:14b5a174998.teoteoteoteo@virgilio.it"
type="cite"><span id="summary_alias_container"><span
id="short_desc_nonedit_display">Considering that we usually
open more than 1 tab at a time the performances of firefox
could decrease drammatically</span></span></blockquote>
We shouldn't spend CPU/GPU cycles rendering video in the background
(even while rendering audio, which is usually muted by default for
these ad things inasmuch as I've seen them). If we do, we should fix
that instead. Doesn't matter if it autoplays or not.<br>
<br>
<blockquote cite="mid:14b5a174998.teoteoteoteo@virgilio.it"
type="cite"><span id="summary_alias_container"><span
id="short_desc_nonedit_display">2. The problem is present
since 3 years ago ... so I think it could be better to fix it
before other tasks<br>
</span></span></blockquote>
There are tens (if not hundreds) of thousands of bugs filed which
are older than 3 years. A bug's age simply isn't correlated to how
important it is in the way you suggest.<br>
<blockquote cite="mid:14b5a174998.teoteoteoteo@virgilio.it"
type="cite"><span id="summary_alias_container"><span
id="short_desc_nonedit_display">3. This is a problem that only
non-</span></span>technicians have, because technicians
usually have an Ad-block that usually stops every ad-video. Thus,
considering that non-technicians don't know that there is
bugzilla, you should consider to fix it as soon as possible even
if there are not a lot of votes<br>
</blockquote>
This applies to everything in bugzilla ("there are many people who
are affected by this who won't be filing bugs or voting for them")
and doesn't in and of itself really indicate an issue's priority,
because the size of the non-bugzilla population is pure speculation
from bug to bug.<br>
<blockquote cite="mid:14b5a174998.teoteoteoteo@virgilio.it"
type="cite">4. Even though I think the suggestion given by Gijs
(write the request from a data-driven point of view) is
meaningless,</blockquote>
In fact, you're proceeding to give an implementation-focused
response. By data-driven, I meant, something based on data about how
many users this affects, and how serious a problem it is. See my
response to (1).<br>
<blockquote cite="mid:14b5a174998.teoteoteoteo@virgilio.it"
type="cite"> I will try to explain my point of view to let you
know how I would solve the problem. <br>
I've never seen firefox source code, but I think there is a point
in which you check for javascript code ... and so you could try to
check, though, for 'video.play()' ... if there is and a user has
declared that he doesn't want to start it at startup you can
simply ignore that instruction.<br>
</blockquote>
This by itself will break every <video> that doesn't use the
default controls (e.g. it would break the youtube html5 player). We
would need to start tracking whether calling .play() happens inside
a user-triggered event, and build a list of "allowed" events and
then fight over that list the way we do over popups. It's then still
more complicated than that, because in some cases the page will need
to wait for more data to arrive before calling play() will work, and
obviously, ads that use "JS-based autoplay" will do the same, and
those two cases will be hard/impossible to distinguish.<br>
<br>
Unfortunately, you've not given much of a justification as to why
this is important in the general sense. That doesn't mean it isn't
important, but we do need a better idea of how widespread it is, and
how widespread the annoyance with it is, backed up in some objective
way, rather than "I am personally really annoyed by this".<br>
<br>
~ Gijs<br>
</body>
</html>