Hacker News Reader

Comment by mandliya | Hacker News Reader

Wow! if this is true.

"A person claiming to be a former Edge developer has today described one such action. For no obvious reason, Google changed YouTube to add a hidden, empty HTML element that overlaid each video. This element disabled Edge's fastest, most efficient hardware accelerated video decoding. It hurt Edge's battery-life performance and took it below Chrome's. The change didn't improve Chrome's performance and didn't appear to serve any real purpose; it just hurt Edge, allowing Google to claim that Chrome's battery life was actually superior to Edge's. Microsoft asked Google if the company could remove the element, to no avail."

View on HN

I was a browser dev. I do not work for google, and in general have a rather low opinion of the company and its ethics. But... c’mon, that quote is stupid. Every web dev knows there is an infinity of strange reasons to add divs in odd places. The idea that this is a stealthy tactic targeted to slow down edge is ridiculous. Edge will patch the issue in a general way, and that will be that, and now Youtubes source would be stuck with the unnecessary complexity. Even if youtube’s devs would be okay with this, there are so many better ways to force youtube users over to chrome..

My guess is this dev had to debug a reported issue “youtube is slow”, and after realizing it is difficult to patch (likely because of shitty code in edge), started pointing fingers. No. That is called a bug, plain and simple. Maybe try fixing your browser implementation so a single empty element doesn’t destroy video performance...

Edit with one more thought: I also feel youtube’s response in not removing the element seems completely justified. Youtube has features to implement and deadline to meet. Why should google have to throw dev resources at fixing problems stemming from microsoft’s bugs with no benefit to themselves? M$ should be the one footing that bill, again, by paying their own devs to fix problems of their own creation.

Although the article somewhat implies it was intentional, an unintentional degradation that severe is still a big issue.

What incentive does Chromium/Google ever have to fix an issue that only affects Microsoft users or non-Chrome users?

To get more users watching videos on YouTube.

It's not a bug; it's a reasonable design choice. You have to have some kind of heuristic, because of the limitations of scanout compositing.

I elaborated more here: https://news.ycombinator.com/item?id=18703568

I wonder if this edge developer's moral indignation also extends to the repeated aggressive "move to edge" tactics. I'm sitting on my windows 10 laptop, where we went from "hey i can switch my browser at will" to "hey we buried it", and then also "when you try switch, it tries to convince you to stay with edge in the hopes you will get tired of trying to switch the defaults".

But that's not enough, apparently, so it was also "when you browse with chrome or firefox, we give you notifications saying you are draining your battery faster and should move to edge"[1]. This predates the stuff talked about in this article.

That's also like the start of these tactics. I'm sure Google and Mozilla asked Microsoft to stop doing that, "to no avail". Meanwhile, the complaint here is basically "they made our annoying and stupid notification not true". Having dealt with these tactics repeatedly, i can't work up any indignation over this. The only one i feel even mildly sorry for is Mozilla.

[1] https://www.theverge.com/2016/8/3/12369326/microsoft-windows...

Wait a guy who works for a company that uses the world's most popular website (the google search homepage) to aggressively proselytize it's own browser is complaining about Windows prompting you to use Edge? For real dude?

Edit: I'll expand. You work for a company the bundled Chrome as a default install with many other toolbars and utility programs. You sell a line of computing products where Chrome is not only the default, it is literally the only browser than can be installed. Chrome is the default browser on the world's most popular OS (Android). Multiple Google properties have refused to work on other browsers and displayed a popup asking people to switch the Chrome (Inbox is the one I remember most vividly). Here's a link to people trying to figure out how to get Chrome to stop asking to be default every time it opens: https://www.tenforums.com/browsers-email/59843-how-stop-goog... As you can see it's not exactly an easy task. All of this, and probably more, and you're here talking about what Windows does? C'mon man.

What does that have to do with his point? Could you address his comment instead? "C'mon man."

He doesn't have a point. He has whataboutism. So I decided to address his whataboutism by pointing out that Google is honestly even worse at what he is talking about, and thus his whataboutism is in fact invalid. So in a roundabout way I did address his "point".

You certainly have an interesting perspective on what you did there.

That comment would be equally as worthwhile and useful if you had started it with some other form of bias to justify it like "Wait, a woman who is black ...."

If your comment starts with some form of "well you are an x so ...", it's probably not actually addressing anything other than your own interesting biases against people.

Or you know, just dismiss whatever you want with bias and claims of whataboutism. I guess that works too!

Despite your claim of whataboutism, i'm instead showing the author is incredibly conveniently eliding large parts of actual history in trying to make their point. You may want to argue those shouldn't matter (or make it worse). That's at least a more valid argument than you made.

I would argue if you asked most people they do in fact matter because the world is not an abstract idealist paradise.

The internet definitely never has been in the past, so saying it isn't in the future is not particular interesting.

I am so unbelievably tired of the hacker news "you work for company x and therefore your opinion is tainted and you aren't allowed to have personal opinions" crap I don't even know where to begin. It's just another form of worthless bias being used to dismiss arguments.

It's no different than any other form of bias, even if it seems more socially acceptable here.

If you want to argue with what I actually wrote, happy to have that discussion. Otherwise, not interested, sorry.

Microsoft was spamming users with messages saying Chrome or Firefox was draining their batteries and they ought to try Edge, and you think that justifies Google putting a feature in Youtube which actually drains an Edge user's batteries?

Maybe you want to walk that statement back a bit? Wow.

>Meanwhile, the complaint here is basically "they made our annoying and stupid notification not true".

What? I am so lost. The complaint is that video acceleration is being disabled, perhaps intentionally, resulting in loss of battery life. That is a bad thing for users, regardless of this whataboutism nonsense. If there was some reason the empty tag was added, Google PR is free to let us know. The fact that they did not is quite telling.

How can you support something blatantly bad as this unless you work for that company? Oh wait...

And Chrome is full of dirty tricks of it's own. For example they paid Gigabyte to add Google Chrome to show up as an 'update' to system software.

http://puu.sh/CjVYU/804038501f.jpg

So lets stop this, which company is worse thing and work towards fixing bugs and making users happier.

> How can you support something blatantly bad as this unless you work for that company?

I tend to agree with the point made in a different part of the thread that arguing against people based on the company they work for is not productive. People are more than just cogs in a machine and have a right to be regarded as such, even if it's just for the sake of argument.

The correct question for DannyBee might be How can you support something blatantly bad as this even if you work for that company?

I'm sure there are plenty of Googlers who would have a problem with deliberately adding a misfeature to a site to slow down a competitor's browser, plenty would fix a bug like that if it were reported and it was really just a bug, etc. Those guys aren't all in leadership positions...

Reminds me of how Google banned Windows Phone's YouTube App citing that the app wasn't written in HTML5 even own app wasn't written in HTML5 -- https://www.theguardian.com/technology/appsblog/2013/aug/15/...

To play devil's advocate a bit, the rule was that third party Youtube apps needed to be written in HTML5. This by definition does not apply to the official app.

This does not make the rule less dumb, but it is clear.

Huh? Why does the empty HTML element do that? YouTube has a lot of not-empty HTML elements overlaid on the video too... I can't imagine a simple <div> alone is doing this, otherwise wouldn't the rest of the UI also do that?

This is standard practice, the system compositor has a fast path for video decoding and rendering. The fast path will not be taken when software composition is needed (eg, something is overlaid or blended).

If you’re asking why they didn’t think to design the fast path to deal with Byzantine faults, it’s probably because why the hell should they?

Yes, I understand that. However, empty divs are not an edge case, and I can imagine a lot of not-insane reasons an empty div might be on top of a <video>. The entire player chrome is within the video nowadays, and it often intersected in the past (the old seek bar for example.) If there is more to this story than an empty div, I would be interested to hear. Today's YouTube seems to put display: none on almost all of the overlays when the user is not hovered, and if that isn't good enough, I think that is a bit absurd. (Actually, there are some empty divs, but they're really just empty... Some contain display: none'd divs.)

The strange part is that Edge developers were not capable of quickly fixing the problems that an empty div was causing.

It's not as simple as you make it sound. If they fixed one issue Google could put another. It's always easier to break something than to fix it, exponentially so.

Yes, just put in a quick fix and release it before taking the time to test it properly, what could go wrong?

This is sounds like something we used to be up in arms about wrt to Microsoft back in the day.

Question about browser implementation: if the Edge team knew that this element's sole purpose was to mess with their performance, couldn't they just configure Edge to ignore that HTLM element when loading a YouTube page? Just strip it from the DOM completely, as though the element never existed?

When building a web-browser, you don’t make one-off patches for each website.

Microsoft made many one-off patches for a lot of 3rd party software in Windows; if this hurts them I cannot see how it is different.

Those were not malicious though, and the software would not intentionally break it next patch. But in this case it would.

Weren't they? In any case? I'm not sure about that, but yes, could be true.

To expand upon my original question, I get that adding in functionality specific to each website is not ideal. But the technical cost here (removal of one DOM element from one website) seems to be drastically smaller than the payoff (not getting screwed over in terms of performance on one of the world's most popular websites).

Am I seriously underestimating the eng cost of such a patch, or is there some unwritten rule that browsers aren't supposed to mess with the HTML that a page serves?

Because within hours of the patch Google would modify YouTube to break it again and Microsoft’s users would have to wait until the next Patch Tuesday - repeat ad infinitum.

Oh you do, well at least you used to: https://dev.opera.com/blog/opera-s-site-patching/

Under the condition that a major website is behaving in a purposefully malicious manner, a one-off patch might be reasonable.

How many one-off patches will be needed since the malicious website will just change something else as soon as the fix is deployed? And rinse and repeat.

a very small retribution for wasting i think it was thousands of man-years on Microsoft-related bugs.

Why not consider this a bug, and patch Edge to ignore the hidden/empty div in a context in which it serves no purpose?

If the intent was to slow down Edge, then this is just a cat and mouse game. Google can always ensure that its web properties are more efficient on Chrome.

That's why it's good for Microsoft long term that they pull out of the html rendering business- they know they cant fight Google in any meaningful way because, well, Google has the monopoly on many Internet services and are able to "optimize" for their own experience.

Even though it's good for Microsoft resource and focus wise, it's bad for the consumer. Supporting Firefox will become more important, like back in early 2000s.

Always good being downvoted without comments. This only ever happens when being critical of Google... go figure. :)

Haha. Love it!

This must be an example of "do the right thing" as opposed to "don't be evil".

"Don't be evil"

Makes you wonder if that was just a really good PR campaign all along.

To their credit, they no longer claim this as their motto...

Reply on news.ycombinator.com