Ask HN: Is Google Secretly Undermining Firefox? | Hacker News Reader

Ask HN: Is Google Secretly Undermining Firefox?

106 pointsx005441 comments

I have noticed recently, after switching to Firefox, that a lot of Googles stuff doesn't work with Firefox. Most resent example is downloading attachments in Gmail. In FF clicking the download attachment button cause the Google servers to basically not respond for at least a few minutes before it would finally allow me to grab the file.

It strikes me that instead of the old motto of "Don't Be Evil" Google is now implementing the new motto of "Be Evil All The Time". Very disappointing.

View on HN

Yes, gmail, calendar, youtube, etc. are slow and sometimes broken for weeks in various ways in FF.

Of course there likely won't ever be hard evidence that they are doing this deliberately. There are many ways for them to achieve this outcome without explicitly instructing their developers to degrade FF experience.

They can successfully bullshit most users that way, but for me this is just more reason to use FF more and Google less. I just hope there are enough other people with the same attitude for us to matter.

I never use anything besides FF and if Youtube wouldn't have worked I would've noticed.

> There are many ways for them to achieve this outcome without explicitly instructing their developers to degrade FF experience.

Like?

Like what other people mentioned in this thread about prioritization.

It's not hard to justify focusing on the most popular browser (which "so just happens" to be Chrome) over a browser that is less popular and not even a system default on Windows or MacOS.

Devs write bad code, gets fixed only in Chrome. Or devs test only in Chrome from the start because they don't have time to test everywhere and if it doesn't even work in Chrome you get bad looks, so it only works well in Chrome.

It looks nefarious but it is 100% corporate priorities.

I doubt anyone at Google is working under a mandate that says explicitly "we must kill Firefox by making our services fail when people use it!"

Instead, every team working on these services at Google has a massive backlog of bugs. The product managers look at the bugs that start with "on Firefox, feature ABC doesn't work correctly..." And then they look at the bugs that are for Chrome. And they put the Chrome ones at a higher priority every time.

They know that their bonuses are tied to Chrome somehow inside the massive Google goal structure.

No one on the team argues about the decisions to deprioritize those bugs.

And, bugs for Firefox get fixed at a slower rate than Chrome, and here we are.

Corporate priorities can be absolutely 100% nefarious as the wage-fixing antitrust investigation and subsequent settlement shows:

https://pando.com/2014/03/25/newly-unsealed-documents-show-s...

https://pando.com/tag/techtopus

In May of this year (2019), https://blog.chriszacharias.com/a-conspiracy-to-kill-ie6 was on HN here: https://news.ycombinator.com/item?id=19798678 .

I take this last event as an example of "the right hand doesn't know what the left hand is doing": these sorts of events that further corporate priorities without corporate knowledge or corporate sanction clearly can and do happen.

---

Widely speculative thought experiment follows

1. Let's hypothetically assume there was a rogue team at Google trying to kill of Mozilla Firefox.

2. Let's further suppose like the YouTube IE6 story they flipped some switch to start the process yesterday.

3. Let's further suppose they were discovered by any one or more of their managers / their directors / Google's legal and compliance / C-level executives today.

4. In my opinion, what we should be asking ourselves is: what actions the last group of people (3.) can take at and with what would be corresponding probability of that action being carried out.

---

tl;dr Do we trust Google to not "be evil"?

A Google engineer would have no incentives to kill Firefox unless he can convince a promo committee that killing Firefox is good for Google, so I am not sure why anyone would do it. I mean, do you think someone would put "Slowed down Firefox gmail page load by 50%." on their promo packet?

So for it to be done in secret for many years you would need someone with enough clout to circumvent the promo process behind it. I am not sure how high up you need to be to do that, or if it is even possible to do for engineers.

Instead if you take the promo driven perspective, it is a lot easier to make things work well in a single browser. So the engineer makes a site twice as faster in Chrome, writes "Made the site twice as fast" in their promo packet and gets promoted. No need to mention that Firefox performance got worse after the fix, the promo committee wont have time to check that.

Did the people who killed IE6 have to convince a promotion committee that killing IE6 would be good for YouTube and Google?

Basically every web developer hated IE6 so they have personal reasons to kill it, I don't know of anyone who hate Firefox.

This. Nobody at the time liked IE6 but it’s not a stretch to assume that a lot of Googlers use Firefox on personal machines.

If you log into gmail via Edge, Google will pop-up a div [1] and email you an ad [2] to push you onto Chrome. Prioritization is not an excuse when they're actively writing code inside gmail to push users off of non-Chrome browsers.

1: https://twitter.com/ridiculous_fish/status/10810743891336806...

2: https://twitter.com/ridiculous_fish/status/10810743910169108...

However, this is policy.

E.g., as an independent vendor or developer, your priorities are probably: "Let's make sure this works on all (or most) major browsers before release." Probably, you have a list of high priority compatibility targets and a list of lowers ones, and there's some room for taste and judgement here. But, if your list of high priority targets consists just of your own products, you're much on the former path of the Microsoft, back when they arguably had some problems with their priorities.

Edit: I once had a conversation with a Chrome developer, when Chrome 32 broke the established behavior of source buffers in the Web Audio API. I was really surprised to learn that they were working in total isolation and didn't even know how other major browsers behaved and what the established behavior was. I wouldn't blame the individual developers, but I'd expect there to be a process in the corporate culture to check such things (at least) before release. (Are we breaking anything?) Having no such a process – and by this favoring your own product and vision over standards – is arguably an aspect of corporate policy. Especially, if you have all the resources to establish such a process.

(speaking only for myself here)

I work at Google and I use Firefox. Based on everything I have seen, it boils down to: browser compatibility requires work. If you don't test it in Firefox, and don't prioritize fixing any issues that appear, then things are going to break. It doesn't require any sabotage, just entropy. I wish Google prioritized Firefox compatibility more.

And that plausible deniability is reason enough that it would be a bad business decision to allocate resources to Firefox compatibility as long as Chrome is dominant.

I recently started using Fastmail in place of Gmail and the difference in the frontend responsiveness is ridiculous. Well worth the 5 dollars a month in time saved. Gmail in Firefox is like navigating a drunk man home.

Also, duckduckgo :)

This so much. My trial just expired, but I will be waiting for my first paycheck to throw money at them.

This so much. My trial just expired, but I will be waiting for my first paycheck.

I've always wondered why performing a Google image search on Firefox for Android always yields low quality images, yet Bing or other search engines don't have this problem.

Try changing the UA string and you'll see that Google is explicitly downgrading the search experience on Firefox for Android

It's not just Google, there are startups making enterprise software these days that officially don't support Firefox.

Some media and banking websites I visit too. I think devs are simply not bothering to test their sites with FF.

In some cases management will say "we don't have the resources to support FF" too.

Is this because FF doesn't follow some standards?

Chrome does seem to support more official API's than Firefox, according to [1] "Can I Use" Browser Scores.

Still, this reminds me of the article [2] "Chrome is Not the Standard"

[1] https://caniuse.com/#comparison

[2] https://v4.chriskrycho.com/2017/chrome-is-not-the-standard.h...

I don't understand why it's okay for such an influential and powerful company to also control how 70% of users accesses the internet. Chrome needs to be spun off from Google.

>I don't understand why it's okay for such an influential and powerful company to also control how 70% of users accesses the internet.

It's not. This is an anti-trust case waiting to happen.

Aside from the slow front ends, Google stuff works on FF for me. Since gmail is their service I use the most I switched it to the basic HTML view to keep my sanity.

The definitive answer can be gleaned by setting the user agent to Chrome (or anything besides Firefox). Have you tried that?

This might not produce the results you expect, because they might be doing feature detection.

Changing the User Agent might help to fool a HTTP server, but not the client side scripting, where every browser offers a different API. In web applications such as GMail, this is the relevant piece.

And I confirm OPs assumptions. I also feel this is a tricky way of Google to make Firefox users feel their browser of being "slow". It is basically power abuse. As Microsoft did it in the 90s.

Google would nearly certainly break the site in such a way that they have plausible deniability.

Browser fingerprinting these days goes FAR beyond trivial user agent values.

Could it be an extension? Have you tested with all extensions disabled?

Do you see the same issue in any other browsers?

If you want to get to the bottom of this, you need to treat it like any other bug and come up with a reproducable test case to share. Until we know more, there's no point in speculating.

Secretly?

I don't know. Hangouts never seemed to work but Meets does, and I think that's newer.

And I thought there was a problem with my DNS that prevented YouTube from loading. (I assumed that chrome was using some special settings)

This is really the shitties thing I have seen from Google, especially first hand.

I am using firefox since 2008, google services are lacking performance in firefox from last 2 years.

First it was only youtube videos, but now it has expanded to docs and other services also.

From last six months Gmail never opens on first attempt in my firefox second is required, but if I switch to chrome it not only magically opens at first attempt but with faster page load.

I run FF on Android and desktop Linux. So far, I didn't notice anything suspicious with Google services such as YouTube and Drive. If technical problems arise, my choice will be to keep using FF and stop using Google services (including YouTube) without hesitation.

It could just be a matter of test coverage and priority, rather than malice

Yes. Try using Google Maps in Firefox.

Try using Google Maps on Android with the dedicated official app. It's slow as hell. Maybe Google Maps is just very slow anyway.

Reply on news.ycombinator.com