What's the future of Servo?
github.com
github.com
They have the kind of deep technical knowledge and ability to solve challenging problems you'd expect from a research group, coupled with the skills to make pragmatic tradeoffs and fix the complex real world problems needed to ship software. More than that, they are one of the most welcoming and friendly teams I ever worked with. The culture they created allowed them to take inexperienced new contributors and quickly ramp them up to a place where they were confident to solve challenging problems. Working in that environment and seeing what's possible has really raised my bar for workplace culture and mentorship.
I know a few people who have been victims of identity theft. After hearing their stories, it was clear to me that it would not have happened if companies weren't tracking them and selling their personal data left and right.
But it's hard to make that connection. People seem to accept identity theft as "something that just happens sometimes", even though it makes them very angry, frustrated, anxious, and eats up a lot of time and sometimes money. I think they do this because it's hard to point to one culprit as the cause, and so it feels like an intractable problem.
If people could actually understand that their lack of privacy protections on the internet (and elsewhere) is one of the main factors in enabling things like identity theft, there would be more people who would affirmatively care about privacy.
I'm not sure how to get people to make that intuitive leap on a large enough scale for it to matter, though. You're very right that people mainly care about browser performance (I'd even argue they don't care about security that much, though certainly more than they care about privacy); I know several people who definitely know better and understand privacy issues on the internet, but still use Chrome because "Firefox is slow on macOS" or something.
A little like the way people talk about "cyber warfare".
In real warfare, the enemy can attack your forces no matter what you do.
In cyber warfare, you can build impenetrable defences. The only way the enemy can attack you successfully is if you make a mistake.
This kind of thinking IMO is perpetuated by absurd tv shows where e.g McGee can hack into any computer system by tapping on his keyboard for a few seconds.
It's harmless on TV, but in the real world it leads to complacency about security. People saying"poor Equifax, they were attacked" instead of "Equifax's security was a joke, and they left their system wide open for anyone to plunder".
Now we just need to find a piece of software that doesn't have any mistakes in it.
> Nobody except tech geeks care about Servo.
This is a good point unless there actually non tech people that care about Servo.
https://madaidans-insecurities.github.io/firefox-chromium.ht...
Do. Not. Use. Chromium.
Chrome has done multiple huge projects that changed how things work at a fundamental level like Oilpan (using a C++ garbage collector), LayoutNG (a full rewrite of the layout part of the engine from the ground up), Slimming Paint (a full rewrite of the compositing and painting system), and Site Isolation.
Blink and WebKit are pretty different engines with very different behavior, bugs, and performance characteristics.
Nobody except geeks and politically minded people care about privacy that much (if they didn't they'd already use Firefox with DDG, or Brave or at least Safari).
People do care for faster, more conventient browser -- that's why we switched from Netscape to IE (when Mozilla bloated and a for a little while IE6 was faster/better), then to Firefox, and then to Chrome.
Well, lots but still insignifcant to make a browser market share raise. Which is not "lots" in the sense that matters. Most People value speed, convenience, features, and security over privacy.
>People really don’t like it if they search for and buy a bicycle and keep getting ads for bicycles for weeks.
Most (as in over 80% of the people) don't even use an ad blocker and could not care less. Even if they are annoyed but keeping getting those ads, they don't care enough to do anything about it, much less switch broswers.
Edit, to summarize: it's not that performance doesn't matter, it's just that currently it's hard enough to compete with Blink & V8. And, of course, Blink becomes better precisely because there is a competition they have to fight. (It will probably become worse when Gecko/Servo/Firefox dies.) But to be that competition without having shitload of money Google has is not fun right now. It's hardly a fight you can win. Easier just to ride on Google's back and to use Blink yourself.
Certainly losing his technical talent is likely to have contributed to the problems, but I suspect it's 100x more of Chrome's dominance and Mozilla's reliance on Google's search engine contracts for revenue.
Maybe he's just not a great guy all around.
https://nulltx.com/brave-browser-draws-criticism-over-lack-o...
> Publishers must verify ownership of their properties with Brave in order to receive contributions from Brave users. If a publisher has not verified ownership, then a user’s contributions will be held in reserve inside the browser for 90 days. The browser routinely updates an internal list of all verified publishers to determine whether a property can receive contributions. At the end of the 90 day period, any contributions marked for unverified publishers will be released back to the wallet. No funds leave the browser except to go to verified creators.
Servo had two major roles within Mozilla - as an incubator for novel browser technology we wanted to ship in Firefox, and more recently, as a lighter-weight vehicle for Mixed Reality products. The latter has been the focus for the last three years, and those products now appear to be winding down. But the former was a huge success - both Servo's parallel CSS engine and its GPU-based graphics layer are now shipping in Firefox.
While it seems unlikely that Mozilla will continue to prototype things in Servo, we're still building lots of innovative technology (and writing lots of Rust code) directly in Firefox. A few of the teams have blogged recently about the work they're doing [2] [3], and I'd encourage anyone interested to check it out.
The Servo team accomplished a ton and left its ongoing mark on the Web. These changes are tough for everyone within Mozilla, but are not indicative of any change in strategy for Firefox. Gecko is alive and well, and there are no plans to switch to Blink.
[1] https://bholley.net/blog/2017/stylo.html [2] https://mozilla-spidermonkey.github.io/blog/ [3] https://mozillagfx.wordpress.com/
And something that may be used as evidence for my skepticism, is why instead of firing those people, where we know a lot of them are pretty good ones, they did not relocate them instead?
Even if we suppose, they decided to not invest in things that are not considered pragmatic, more to the core of Mozilla... Even if some of them were working coding in Rust for instance.. why dont they just shut down the projects that they considered 'fancy' and relocate those amazing engineers to work into core parts?
For me it smells like something is rot on the top level, and that in turn, will lead to more bad decisions in the coming years, unless something really radical happens (Microsoft did it, so theres hope).
Edit: one more evidence that the things are somehow rot, is that money going to pay for ads (and this is exactly the path taken by Yahoo)
We invest a lot in Gecko though, precisely because it's important to have a world-class independent engine.
But I completely understand why you wouldn't want to "boil the ocean". That said, I hope you appreciate how much support FF has from users. Anyone of us left using are doing so by paying a price for it. Everything else mozilla does appears like distracting side-quests. I would not care much about donating for wikipedia but if Mozilla focused on their flagship product more I would be happy to contribute both financially and through bug reports and grunt work PR's.
We live on the ocean and browsers are our ship.
If those are really long term high priority goals of Mozilla, it seems that keeping the team around that has probably the most experience doing this anywhere on the planet (building innovative browser functionality in Rust) would be a good idea, even if the Servo project as an incubator was no longer a priority.
Surely, their talent and team cohesion and the team’s collective expertise could have been put to use writing Rust features directly for Firefox.
Seems to me the strategy is now to survive, more than anything else. So it did change, didn't it?
Gecko seems to be tightly bound to Firefox. Are there any plans on decoupling that and making it easily embeddable for other projects? Even MDN doesn't seem to have up to date documentation on that [0].
I purport that allowing easy embedding will allow more people to play around with Gecko, discover wildly different classes of bugs, contribute code in order to add improvements and fix bugs that they discover while embedding the project.
Given a more visible and active community around Gecko, Microsoft might've even switched to Gecko, but that's just a wild guess.
0: https://developer.mozilla.org/en-US/docs/Mozilla/Gecko/Gecko...
https://mozilla.github.io/geckoview/
It’s also pretty huge though. And Apple won’t let you embed it in an iOS app. So there are limits.
It's sad, but the reality is that the fate of a project with only one major stakeholder is tied to that stakeholder's fate.
I'm pretty sure they will all get other jobs, but we have to face the fact that the project itself will probably die.
Do you consider getting no major OEM adoption in 5 years “fairly strong?” Mobile OSes depend on network effects of both users and developers. It’s no surprise it’s a duopoly.
Personally I think OEMs should not ship a default OS and should require a choice at the point of sale.
'No major OEM adoption' is partly the USP for manufacturers like Pine Phone[1], if they are able to build sustainable business then they would be fine(From the looks of it, it seem they know what they are doing).
https://github.com/servo/servo/discussions/27575
It's a very sad situation, I hope Samsung or another organisation steps in to hire the team.
+title changed from "Servo Parallel Browser Engine Project"
Now is the critical time for a another company to step in. Microsoft has better alignment with free web than Facebook or Google. This could be a chance for them to get a competitive independent browser again.
It's interesting to compare what we know of "Midori C#" to Rust.
The point would be the technology and independence from Google, not the market share.
There's a lot that a browser can do for a company. Google has built GSuite and ChromeOS around Chrome - the integration with the browser is absurd and right up Microsoft's alley.
Microsoft could build Azure Active Directory, O365, Windows, etc, around a next gen browser and I think they'd be smart to do so.
It is. Chrome is so dominant at this point they're the de facto web standard. Other browsers have to put in significant effort just to play catch up and can't really offer any new renderer/js feature without Chrome's blessing.
Plus, we have a decade's worth of evidence that Chromium is just flat out better than the competition. Even if we set aside the problems with market share the Chromium team has consistently put out a faster engine than Firefox or IE.
Adopting Chromium allows you to (mostly) avoid putting effort into the rendering engine allowing you to focus on other features to differentiate yourself.
Microsoft still runs the dominant operating system! They have a huge opportunity here to complete. The same way Chrome advertises on Google.com, Microsoft can provide the default fucking browser that is HUGE.
> Plus, we have a decade's worth of evidence that Chromium
Chrome is 12 years old. It's been super popular for less than a decade.
Its multiprocess architecture, by far its most long term scalable feature, was behind Internet Explorer's (they started multiproc trend with, iirc, IE8).
> Even if we set aside the problems with market share the Chromium team has consistently put out a faster engine than Firefox or IE.
Acquire Servo, do better? That's my entire point.
And they already do, with a Chromium based browser.
> They have a huge opportunity here to complete
With what? Maintaining the other browser engine is not that monetizable.
Google absolutely monetizes the hell out of Chrome, in areas they compete with MS (O365? GSuite?).
Google doesn't monetize developing Blink/Chromium. Since Edge only uses that part they don't really have to care about Google's monetization and do their own. If Google stops Chromium development or goes the wrong way at some point, which is inevitable, MS will have to step in, but until then it's free real estate.
They can build all that proprietary stuff with Chromium just as well, so why should they choose Firefox?
I disagree here. Of course they do. They could not monetize Chrome successfully without:
a) Having huge amounts of control over the direction of the web
b) Having a best-in-class browser that people want to use
If IE had every feature of Chrome with regards to management and monetization it would still not matter, people do not want IE. If IE were a drastically better browser than Chrome, they would.
Choosing Servo gives them an opportunity to leap ahead of Chrome in areas they care about.
They've had that advantage since Chrome came out. Didn't stop them from losing almost all their market share.
>Its multiprocess architecture, by far its most long term scalable feature, was behind Internet Explorer's (they started multiproc trend with, iirc, IE8).
Chrome has been technologically ahead of IE since it came out.
>Acquire Servo, do better? That's my entire point.
Why do we think Servo will be better?
Absolutely. But it certainly is a huge benefit to them.
> Chrome has been technologically ahead of IE since it came out.
No doubt, in a number of ways that's certainly true. IE needs to compete on tech.
> Why do we think Servo will be better?
Because it solves real problems for browsers - security and performance, in a more than incremental way.
Why? They’re more supportive of open source than they were in 2001, yes. But Google and Facebook are also big in this area.
And MS business still relays a lot on vendor lock in to a great extent
(Amazon has this incentive too, but they're just kind of incompetent at the execution, so people don't worry as much about them. See: the Kindle Fire tablet.)
Meanwhile, Microsoft doesn't currently own any property that operates by trying to replace "eyeballs on the web generally" with "eyeballs on that property specifically" to then sell those eyeballs to advertisers. They have Bing, Cortana, GitHub, Skype, maybe TikTok soon. None of those things will really change that stance. They're not "against" the web.
Apple is in the middle on this spectrum. Apple doesn't make money off web eyeballs, but they do make money off native app eyeballs (iAds), and also off native-app sales generally. So they're motivated-enough in providing a decent web experience to be the main sponsors of the WebKit FOSS project; but they're also motivated to create APIs that allow their third-party devs to replace web experiences with native app experiences.
Apple shutdown iAd in 2016 https://developer.apple.com/support/iad/
Of course google could be forced to by governments' actions.
Today, in Chromium, Microsoft are a very small percentage of commits. Google still accounts for over 90%. If Microsoft wanted to fork Chromium, they'd need to vastly increase their investment if they wanted to keep up with Chromium.
Never underestimate Microsoft's ability to constantly change its mind.
On a predictable schedule, too! Cf. their annual reorg, which leaves everyone confused for a good 3-4 months per year.
That would be precious: Microsoft taking Mozilla browser developers onboard, creating a browser, shipping it by default bundled with their OS, and being able to claim it's to prevent monopolies.
It would also be nice for those highly skilled developers who actually believed in the open web to be able to give the middle finger to those in middle and upper management for whom Mozilla is merely a stepping stone to some cozier job at a proprietary software house whilst running Mozilla into the ground.
I don't see Microsoft ever adopting a strategy like this. It'd be nice, but given the disaster that Edge turned out to be — it's highly unlikely.
The only problem with EdgeHTML was google abusing their monopoly, read this: https://news.ycombinator.com/item?id=18697824
But thanks, will do on a laptop where I have less stuff to migrate, slower CPU with less memory, and sometimes care about power consumption.
To anyone interested, here’s the manual: https://docs.microsoft.com/en-us/deployedge/microsoft-edge-s...
There's a pandemic, probably a global recession, massive tech layoffs everywhere, and tech interviewing is an absolute nightmare no matter how good you are. I'm not sure I'd be as confident unfortunately.
> Microsoft has better alignment with free web than Facebook or Google.
Microsoft gave up on browsers a long time ago.
If someone would want to use Servo outside of a hobby project, it would be Google to integrate it into Chrome. The chances of that happening is almost zero.
Not that I expect him to have any actual motivation at this point given Servo is basically toast either way.
They had detected the most profitable improvements and moved them over to Firefox.
So I'm not really surprised about them not continuing servo, even though it had been grate.
I'm more worried about some of the other layoff.
Further it feels a bit like they got in a place where they financially couldn't go on like before and fanned out to get more revenue streams but if that fails we might be down to chrome + safari, which would be horrible, tbh.
If it really comes to that we might even want to try to push the EU to get financially involved or something.
In the end you are 80% of the way to the goal and the last 20% is politically insurmountable. This is, IME, a primary driver for the Lava Flow Antipattern.
Safari isn't available on Windows, and Google doesn't want a situation where they are the only significant browser on the most popular desktop operating system. Google is scared of antitrust action, getting another monopoly is the last thing they want.
Those don't sound like incompatible approaches. The monolith Gecko would be difficult to replace whole hog, a modular surgical approach seems more realistic.
If they don't invest in a new engine, and given that the old engine is not much to compare with Blink or even Webkit, what's left for them? Pocket and some BS VPN service?
"our source told us Moz will likely pocket $400m to $450m a year between now and 2023 from the arrangement, citing internal discussions held earlier this year."
[1]https://www.theregister.com/2020/08/14/mozilla_google_search...
What the hell are they doing with the rest of the money? It should be "paid back" to the shareholder (the foundation) as tax free dividends that are kept in a fund for the purposes of that foundation.
Do you have a citation supporting that? The old engine compares quite favorably with Chrome and Mobile Safari.
https://caniuse.com/#compare=firefox+79,chrome+84,safari+13....
The provided links only compare API coverage, while other domains (performance, security...) are important. Even so, Firefox is clearly trailing Chrome, its main desktop competitor. It's not surprising since I've heard that the latter has much more man-power. And the gap will increase.
I've looked at CanIUse, ignoring the features which are implemented but disabled by default. Chromes lacks 5 features the Firefox has. The biggest two are MathML and VirtualReality, which were both available as experimental features in past Chrome versions but were removed for lack of interest in publishing or maintaining them.
On the other side, Firefox lacks 25-30 features that Chrome has. A few of them are minor, but most of them seem complex.
Chromium's folder with build and .git takes 40GB.
Why does it matter if there is one source tree for all browsers? [0]
Is it really a bad thing if Chromium is the base, and then Google, Microsoft, etc push out forks of it?
Isn't that what we do in the linux world already with the kernel? Redhat, Ubuntu, Mint, etc all take a stock kernel from the git tree and then tweak it for their uses, yet we don't complain about that.
[0] - Yes, I know people lost their jobs, and that sucks, especially in this economy. But my question isn't about the jobs, it's about having multiple browsers and why some people vehemently believe that multiple browsers without shared source code is critical to our future. THAT is what I don't understand.
Google has a strong interest in how people end up using the web and they have a particular vision for it, a vision that many on here disagree with.
Imagine if IBM ran the Linux kernel. We would be hearing many of the same concerns.
Why is that a bad thing?
And why would Firefox being a contributor make it worse? It’s open source.
That should be a good thing.
Given a single implementation, what incentive is there to develop open standards?
How do you innovate (in standards, JavaScript performance, rendering, etc.) if there is only one group working on them?
Case in point: Firefox is the only major browser to provide MathML support. Where is the incentive to get that implemented in the other browsers, especially given MathJax?
Where is the competition to innovate in things like web tools (layout analysis -- esp. for flexbox, etc., accessibility property (WCAG, WAI-ARIA) navigation/investigation) given a single implementation?
Also, it's never a bad idea to implement the same thing twice, from scratch. Working with an existing codebase necessarily puts you into a certain mindset with its own trade-offs and constraints where moving in a certain direction might be much harder than if you are starting from scratch. Would we ever see innovations that Servo brought forth if all we had were Chromium forks? It's hard to say, but it's definitely not obvious.
Would W3C apply the same rule if there was only Chromium?
If Firefox were as well, would that be bad?
Seems like it’s a community project now.
That said, Servo did seem to be one of the more lively places, so perhaps it will continue.
Rust seems to have a strong community of voluntary contributors, and could probably be run without company backing. Similar stories if you look at major FOSS projects such as KDE and Gnome - primarily dependent on a wide base of volunteers.
Then there are projects with a mix: plenty of volunteers, along with many commercial contributors - the commercial contributors tend to be more significant (certainly they add more code - but then they're the ones pushing new features for their customers) - but then there are enough volunteer contributions that the projects aren't dependent on the commercial entities. Linux Kernel, LibreOffice, Kubernetes, VSCode, come to mind. And with multiple commercial entities, it's not a tragedy if one drops out.
Then there are those projects with a single commercial backer, and fewer volunteers. Those are the ones that die off when the company drops out. The question is - is Servo in this category?
I am not too up to date on KDE, but Gnome is mainly pushed forward by Red Hat, and recently by Canonical. Unpaid volunteers certainly make a decent amount of contributions but the vast amount of progress comes from corporations, not volunteers.
Even the Linux kernel would never had taken off beyond an hobby project had not been for contributions from Oracle, Compaq, IBM, Intel engineers during the early days.
So yeah will Servo find such a benefactor, very difficult to say, but Samsung used to contribute, if I remember correctly.
But I had no idea they'd kill something so core to their future browser development. I strongly disagree with this decision. They might as well kill gecko too, if there's no exciting future anymore.
Samsung had invested in the past, so hopefully they or someone else will pick up where Mozilla left off.
I do think that Mozilla should slowly try to reduce the amount of unique processes that they have. Part of what made servo so successful is that they did all the development on GitHub.
Mozilla should have a top level corporate policy to transition away from all their nonstandard internal tools. Get rid of Mach, Bugzilla, Mercurial, etc.
They should use GitLab or GitHub, switch to cmake, and some standard third party build service, and be done with it. Their productivity would increase, their community contributions would increase, and their overhead would decrease.
But they should not have cut their DevTools, and should have actually beefed up the developer-friendliness. Compete with Electron. Compete with PhantomJS. Improve the experience of Progressive Web Apps on mobile.
The focus on new products makes sense. I think Lockwise, Monitor, and their VPN service were all good efforts. Firefox Sync should ideally expand more into a whole suite of integrated services stop their zero-knowledge privacy-first architecture. Storage, backups, OwnCloud integration, etc.
Maybe they should even acquire OwnCloud or NextCloud.
I've just switched back to it (after using Chrome for years because of Firefox's slowness) and feel like I'm going to use it for quite a long anyway. I just hope somebody is going to keep making vulnerability patches.
I don't want to browse the web with a Blink-based browser. Total browser homogeneity is the end of the free and open web. I don't want to use a non-free non-open web.
Is the web dead to me?
That's quite a drastic statement. Maybe the devs find another company that sponsors the development of servo?
MS also seem to have a lot more interest in Rust than Google do.
(Much longer ago, they were also responsible for Tasman.)
"It's time for @mozilla to get down from their philosophical ivory tower. The web is dominated by Chromium, if they really 'cared' about the web, they would be contributing instead of building a parallel universe that's used by less than five percent?" [1]
[1] : https://www.zdnet.com/article/microsoft-guy-mozilla-should-g...
Standards bodies are powerless if there's only one implementation.
https://m.youtube.com/watch?v=62rAjp5Nr9g
https://www.forbes.com/sites/zakdoffman/2020/02/21/google-wa...
Ways to mess with Chromium Edge. It's not like it's a company full of smart, engineers, capable of skirting around the law.
What's the point of having an extremely complex standard that nobody else will ever implement anyway? Remember, it would take billions of dollars to implement a competitive browser and then you have to give it away for free.
If even a trillion dollar company like Microsoft thinks it's better to team up with a competitor rather than implement HTML all by themselves, it really should make you think.
Consider, there's only one Linux kernel, there's no "Linux standard", just one kernel - and the world runs on it. I don't see people running around complaining about how that's a threat, even though it's much harder to switch operating systems than browsers. Sure, there's POSIX, but that's a tiny subset of Linux features.
What's the deal anyway, is Mozzila a non profit for the development of the Firefox browser and an open FOSS web standards based web engine, or a sandbox for business types to play with and implement various BS ideas (ads, VPN, mobile OS, etc)?
An actual community project should rise from this state of affairs, with tanginble, browser-oriented goals, not something dependent on Google's money...
Here's the person who announced the firings... Nothing to do with development or FOSS, some laywer ex-Netscape employee:
Winifred Mitchell Baker (born 1959) is the Executive Chairwoman and CEO of the Mozilla Foundation and of Mozilla Corporation, a subsidiary of the Mozilla Foundation that coordinates development of the open source Mozilla Internet applications, including the Mozilla Firefox web browser. Baker was trained as a lawyer. She coordinates business and policy issues and sits on both the Mozilla Foundation Board of Directors and the Mozilla Corporation Board of Directors.
It's like the leeches that hang on to NGOs for a salaried office career, and don't do activism themselves...
For all his faults, at least Brendan Eich was an actual developer, and with important contributions to the web and Netscape/Mozilla/Firefox.
Not exactly nothing to do with FOSS. She wrote the MPL. https://en.wikipedia.org/wiki/Mitchell_Baker
Having said that, I do think these recent layoffs reflect badly on her if she is indeed responsible for them.
This is just so naive view ... it's always the abstract "community" which should step in and save the day. Recently it was LibreOffice too where people got all mad due to proposed branding and again "community" superhero was expected to step in, fork it and carry on the whole development.
In reality these projects are far too huge to be purely community projects. They require full time paid developers and not few of them. They won't be able to survive without corporate sponsorship (and thus influence).
Mozilla is far too huge to not have money. Just 200M from those 600 million per year it took from Google, put in the bank, could pay for 50 developers/graphic designers/ $200K per year to work on it for 20 years. That would have taken care of the engine for 2 decades...
Instead they burnt money in BS ventures, Cxx salaries, events, and so on...
KHTML was done by a tiny KDE team.
After Apple adopted it, Webkit/Safari only had a small team of people working on it.
A good team of 12-20 devs is better than some BS team of 250 with politics, communication overhead, lack of coherence etc that comes with scale...
Yes, 20 years ago.
Fortunately, web technology did not evolve and grow at all since so it should be doable now too, right?
So? 20 years ago Netscape still had a much larger team, Microsoft too, and Webkit with its small team still beat them in their own game.
Small teams can build whole OSes, browsers, compilers, and whatever they put their mind too. Given good devs/experts, they can do so even better/faster than bigger teams.
P.S. 1 Not that a 50-strong team is small.
P.S. 2 Plus, way of missing the point. The team-of-50, 200K, 20 years = 200M was just an example. 200M is 1/3rd of what Mozilla made in a year for decades. And even those 200M would just be the raw money spent, not invested or anything. They could afford to support a 300-strong team of 20 years with 2 years of their revenue un-invested. And that's with a $200K/year salary, which doesn't have to be...
https://gizmodo.com/yahoo-s-insanely-bad-deal-to-pay-mozilla...
I have not heard about this in a very long time (note this article is from December 2017), but also, litigation takes a very long time.
That is way too low to be anything more than maintenance mode. By my estimation [1], you'd want somewhere around 250 full-time equivalents to be able to declare the project in a healthy state.
[1] Take the number I'd estimate for Thunderbird, multiply by 10 for Firefox-scale, then add another 20% to account for the operations that Thunderbird gets "for free" (e.g., maintaining server infrastructure) and accounting for the fact that email standards are far slower in innovation than browser standards.
That is, based on what you said, it sounds like there is no meaningful difference between having a non-profit do all the work, and a for-profit doing work on behalf of the non-profit that owns it. But if that was the case, why would anyone bother with the extra layers? I'm inclined to assume there is some reason for the extra layers, and I have difficulty believing those differences are in society's best interest.
It can also pay it's Cxx's nice salaries and hefty golden parachutes, which makes it a for profit for them...
By the looks of it, Servo is a project that failed to deliver a new browser engine after years of work. If there were a chance that the project could be turned around to result in usable software, I'm sure Moz wouldn't have killed it. Is it worth the effort to pickup the project rather than starting from scratch? I don't know, and would appreciate if contributors share their opinion, or come up with a roadmap. By the reaction of some devs here on HN ("don't worry, Rust is safe; no really!") unfortunately I got the impression that the project was treated more as a showcase for Rust rather than a serious attempt at a new browser engine (please don't take it personally; I know this attitude from the "100% Javs" days and consider it a junior dev trait). Also, the "rise from ashes"/Phoenix metapher is no stranger to Mozilla, only that Moz picked the remains from Netscape; the effort to code a browser from scratch (in Java) had failed as well; it resulted in the Rhino JavaScript engine which is still heavily used.
There was plenty more work to do (other commenters have mentioned a new layout engine), and I'm disappointed to see Mozilla abandoning such a core next-gen R&D project that has already brought substantial improvements to their main product.
That's really not a true reflection of prior communications about Servo's goals.
The parallel layout engine failed. It just kept running into corner cases that didn't handle parallel layout well. If you look in the layout_2020 folder in the source repository, you can see an in-progress pivot to a new approach that would hopefully have less problems, but now we'll never know.
This isn't a surprise. It was supposed to be a research project; if parts of it didn't fail, that would be a surprise.
Just to illustrate the numbers: a fresh graduate with no experience will easily get more than 100k (even at Mozilla, who IME pay a bit less). A plain manager of a 10-person team at a big bay area company will be earning close to 500k (and most of their direct reports will also be in the 300k-500k range). Then you get your principal and distinguished engineers who can easily make 1M per year. 2.5M for someone leading a 1000 person company isn't expensive, and you do need someone to lead that company - to make tose strategic decisions.
You can quibble around whether or not a specific person made the right decisions (Mozilla aren't doing great, but they're also in a tough environment - maybe their CEO could be making better decisions that would boost usage - or maybe usage is entirely out of their control.) But you do need that person leading the company. And you need to retain them.
And her job definitely is needed, regardless of how well it's being done.
Normally, the shareholders own the company. They appoint the CEO, or a board of directors or whatever management structure, and they decide how much to pay them. It's the shareholders' company, it's the shareholders' right to decide whether to keep the C-suite and how much to pay them. Things can get a bit muddled with large public companies with many shareholders, or when there are dual class shares, which can partially insulate the management from the owners control, but it more or less works this way.
In the case of Mozilla Foundation, AFAICS, this does not hold. The board of directors is completely self managing; they coopt board members and appoint the CEO of Mozilla Corporation.
I might be very wrong about this, but since Baker is both the chairman of the foundation and the CEO of the corporation, it looks to me like she doesn't basically answer to anyone. It's a bit like she owns the company, except she didn't have to buy it. That's a pretty sweet deal.
If I can take a literal potato, stick it in a pot, put that pot in an expensive chair in an expensive well-lit CEO office, have the potato's PA water it every day, and at the end of the year can claim more growth and sustainability, then "how well it's being done" kind of starts to matter.
Citation needed.
>But you do need that person leading the company. And you need to retain them.
I don't even need / want a company. I want the non-profit organization I was promised, with community leading, and perhaps 1-2 BDFLs to make the final decisions.
Linux did well without a CEO...
I'm sure the team working on it are talented but I can't imagine any commercial company picking this up for any reason.
Servo is not the engine used in Firefox.
https://en.wikipedia.org/wiki/Servo_(software)#Relationship_...
Given that Gecko is probably not the bottleneck, I would consider the decision of halting the project not a bad one. Servo delivered tons of amazing things already, it is a clear success. Maybe Mozilla should consider using V8? After all, losing SpiderMonkey might not be that bad, there is enough "competitors".
And indeed, zipped Chromium sources are of 1.4 Gb and Linux kernel is of 0.2 Gb - 7 times less.
Volume of features that "we" (or rather "they") want to see in browsers and so complexity of browsers reached point of no return I think. From now and on only state actors will be able to finance development of another new browser.
I do not think that any private pure product company now is able to develop modern browser sustainably. Either we (users) shall pay to the company for it or the company shall sell telemetry, user's data, influence user choice, etc. to get money for that from FAANG & Co. And that is exactly what happens now with Mozilla.
What to do?
The solution could be in drastically limiting the scope of browser. Indeed, 95% of sites and users just need basic HTML/CSS/JS features. The only thing that needs to be added is what I name as WebVM - virtual machine a la JavaVM or .NET "VM" with DOM and CSSOM exposed in it as primitives. So if you want to add new CSS, HTML feature or to support new script language then you can provide corresponding loadable .class (bytecodes) file.
I mean: compact base (bootstrap) implementation + VM/runtime to extend this thing. All other features can be developed independently by teams who need them for their users and cases.
All those WebGL, WebAsm, rarely used CSS effects and HTML features ... 95% of us do not need them - why do we need to pay the price for these features? If someone will need them - fine, design and load them into WebVM.
As an example, I (+ 3more devs) can transform my Sciter (https://sciter.com, 0.02 Gb zipped sources) into such bootstrap browser in 6-9 months. And so can do other teams.
Sigh.
10 years ago we have thrown the baby out with the bathwater when we disabled ActiveX, Java, etc. Instead we should actually develop WebVM and so to move web tech development out of big companies and Central Comity of the Web.
Sure, ecosystem diversity and so on is great but the current forkers don't see a way to make money from that, otherwise they'd have done it.
Barely worth any money.
> Why would you fork Chromium if you’re interested in Firefox?
Free maintenance (worth a lot of money).
My point is that there was no profit model around Firefox so switching engines is negligible. The cost of building said engine is also going to be negligible to the profit returned from the browser. In Google's case this is true. In Mozilla's case they have no model and never had one to subsidize that cost (other than Google search money). Changing base engines turns Firefox into a hobby project managed by a FOSS rather than a serious competing project in the browser-verse.
In reality Blink diverged significantly from WebKit and it doesn't make any sense to call Blink WebKit.
It seems like if things keep going this way, Safari also would inevitably end up rebasing Safari onto Blink too (with their strategic changes).
I'm not sure about that. WebKit is lagging slightly behind Blink (and Gecko to lesser degree), but I don't think it bothers Apple. Budget to develop WebKit is for Apple hardly a problem as well.
Having independent browser engine gives them quite a lot of strategic power given their exclusive market share. Look no further than how they managed to single-handedly kill PWAs few months ago since it does not fit into their strategic goals (all apps must go through app store).
I'd expect them to maintain all strategic decisions if they did refork, including disabling PWA features; there's nothing that says they have to enable PWAs just because they fork from Blink.
Because I was scanning around to get a lay of the browser engine landscape and there seems to be tons of Safari traffic out there.
I was absolutely gutted to hear about Mozilla's "reductions" (I hate such euphemisms), perhaps even worse for the Web is abandoning Servo... Which I wanted to find out whether that's the case.
I've made the replies top-level comments since the thread scope is now at this level.
Therefore it would be much wiser for mozilla to switch to v8 like everybody else and to reallocate the money either towards resurrecting servo OR towards improving V8 and making it faster, which here would bring actual value to humans.
Can we not find a simple naming convention that is less ambiguous and confusing to the rest of the world that uses both software and hardware? E.g. ServoSW, ContainerSW, or stick with names that have no hardware corollary?
I mean, maybe someone with his background in charge would know why its important for Mozilla to keep this kind of projects even if it looks they are not going anywhere, because this is what gives Mozilla meaning in the end.
Im not a mozillian, so i might be totally wrong, but maybe this is a case of the MBA's taking over and trying to focus into what they have learned to do.. look at the profit margins and think they are spending too much in research projects.
There are a lot of examples of other companies losing their engineering culture, starting to think more in terms of cash and profit, and finally vanishing its purpose and meaning.
IBM(lost in the 80's), Microsoft almost loosing it in the Ballmer era, Sun, and the most proeminent case: Yahoo.
Just to make it clear: Tech companies as any other company must have a good financial health and this is very important, but they cannot afford to loose their soul.
Once their engineering and innovation culture is gone, they become void and suffer from a slow bleeding til the death.
We must face it, our economical systems of incentives, the economical game generally speaking, its broken.
And tech and arts are proeminent endeavors that tend to get trojan horsed by this yuppie mentality turning things that once had meaning, into meaningful cash cows that work for the few people that cash out from the corrupted source.
(Apparently, it works a little better for industrial-level enterprises).
My hope is that, with time, this will become more evident, and research find innovative ways to make a better economical game so that creative, art and engineering culture kind of companies can have a much larger life span.
By the way, they had a pretty good leverage on Rust (and Rust was a really risk bet that ended doing great). The fact that they fired the people with this background just shows that they have no clue of what they are really doing.
* Edit: just being clear about people with conservative views.. I dont like and even despise what he did politically speaking, but as long as he were a good fit for the job, and never forced his political and social views into the company, i dont agree with the outcome.. and the reason is starting to show up now.. Much more important things are starting to fall out, and in the end even people with progressive views and good engineering background ended fired in the long term.
There have been quite a few instances over the past few years of Mozilla demonstrating its left leaning politics:
* The Brendan Eich situation where his private contribution to a conservative cause impacted his employment
* Baker calling a former conservative employee's views "traumatic and damaging" in their obituary - https://blog.lizardwrangler.com/2018/08/07/in-memoriam-gerva...
* The removal of "meritocracy" from the governance document in the name of D&I - https://blog.mozilla.org/careers/words-matter-moving-beyond-...
* Promoting BLM and anti-racism content via Pocket and their homepage
* Replacing "master password" with "primary password" - https://support.mozilla.org/en-US/kb/primary-password-replac...
Agree with their positions or not, they are definitely alienating to some. Even if just delivering a great browser and getting people to use it is not their primary mission (and it's not directly if you've ever read the Mozilla Manifesto), it is the means by which they retain the ability pursue those goals.
Recall Michael Jordan's reason for staying away from politics, "Republicans buy sneakers too".
It doesn't matter which way their politics lean, does it? The point is that a software company can't put politics over technology.
Unless it's specifically software politics and relevant to the organization's purpose, like the FSF. But if Mozilla had a political stance expressed through software (e.g. openness, user empowerment?) it has been somewhat muddled for quite some time now.
She went a little far in what she said publically, but she wasn't wrong. I knew Gerv from very early on. He was a convert to christianity and as most converts, was vehement in it's absolute correctness to the detriment of his relationships. But, when you think you have an absolute truth of life after death on your side, people become a secondary concern.
> * Promoting BLM and anti-racism content via Pocket and their homepage
So, I will admit that I don't see there being any valid stance besides being against racism, and for treating all human life with value.
I'm going to repeat what I said about Eich in another thread:
Mozilla built a brand around openness and the idea of "putting people first" and making them feel "empowered, safe and independent." (Those quotes are from at least one version of their mission statement.) Eich's backing of the anti-gay-marriage initiative was a PR problem for them in a way that it might not have been for many other companies. Also, he violated the first rule of holes (i.e., when you're in one, stop digging); it's possible they might have been able to do effective damage control without booting him if his initial response hadn't been, in so many words, "it's my money and I can do what I want with it."
And, sure: it is, and he can. The CEO of a nonprofit Catholic hospital chain could also use their money to donate to Planned Parenthood. But, if they did, we can be reasonably sure the hospital's boards of directors would have words with them about it.
As for the others, well:
> Baker calling a former conservative employee's views "traumatic and damaging" in their obituary
It sounds like he was not merely conservative, but extremely strident and pushy about certain things. Also, you're giving that obituary a super reductive reading, given all the nice things Baker also says about the guy despite their obvious disagreements over those views.
> The removal of "meritocracy" from the governance document in the name of D&I
Some might argue that this kind of nitpicky language monitoring is an awfully unimportant tiny hill to die on. Maybe so, but it's hard not to see "I must take a stand in defense of phrasing changes I probably wouldn't have noticed if this blog post hadn't made a point of calling them out" as, well, an even smaller and less important hill, isn't it?
> Promoting BLM and anti-racism content
If someone is truly offended by "anti-racism content," I'm gonna have questions about them.
> Recall Michael Jordan's reason for staying away from politics, "Republicans buy sneakers too".
Just for the record: he's donating $100 million to Black Lives Matter.
Zero, and with all respect, "could it be... SOCIAL JUSTICE?" was an awfully silly way to open an otherwise reasonable comment. (One that I'm about to go on to disagree with, granted, but still.)
Mozilla is not being run by "the MBAs"; it's being run by Mitchell Baker, who was actually CEO before Eich and has been there since before Mozilla was a separate corporation. She wrote the Mozilla Public License. She may not be the technologist Eich is, but it's awfully hard to fault her historical commitment to open source, the open web, and Mozilla. There's a lot of conspiratorial thinking about how Mozilla's moves over the past week or so are being driven by some kind of ruthless corporate greed, but a less nefarious explanation seems more likely.
90% of Mozilla's revenue comes from Google's contract to be Firefox's default search provider, that contract is up this year, and according to news reports, the contract hasn't been renewed. these moves are being driven by a desperate desire not to go under when their income stream drops by orders of magnitude. They aren't becoming more commercial because of a "yuppie mentality"; they're becoming more commercial because they see no alternative.
As you can see reading further the core of the point i was trying to make is; we are all imperfect, thats a fact of life.
Than we have fitness, and as more sophisticated things get, fitness is harder to achieve.
To be fair, i dont know how much Eich was a good fit as a CEO for Mozilla, all i can deal with is that he was booted not because of performance, but a political stuff that got uncovered and that was against the image Mozilla was known to defend. Fair enough.. is it serious? Yes, because it damages the image of Mozilla even if its not the fault of the corporation, because it sounds like an endorsement..
Ok, but its all a matter of weighting, but not only looking for the short term, but the long term.. maybe there was someone in Mozilla that had the technical baggage and also the socio-political views
Again, i dont know how well the person picked to be the next CEO can evaluate and weight decisions in the deepest level not only of policy but also engineering.
But i´ve seen the external movements of the chess board, and this last one, made it more clear where Mozilla is heading and how they are evaluating their decisions.
Mozilla did bold movements that did not worked out, but they were not tottally misguided, as they could actually succeed, so they were not maded by a totally unaware person. Its just that its super hard to do everything right and have luck at the same time.
But i think, the real problems were not in those bold movements (and right now when you look at them in the rear window, they look just like cash burned) but in the innability to follow and keep their core projects moving.
For instance, the browser took too much time to keep up with the multi-process architecture in Chrome. It took like 4 years if im not mistaken..
Meanwhile there were a lot heat being dispersed in moonshots.. again, is not that the moonshots were wrong per-se, but as long as the most of the energy were being expended in keeping up in the core of what make mozilla matter today.
But hey, the moonshots maybe were misguided somehow, but here, there this one or two that we invested in the past that is starting to bear fruits.. we were lucky.. so why now shutdown the projects that might actually be the ones that can save mozilla in the future?
If it was just a matter of closing the offices elsewhere in the world(like the one in Berlin), but we know that part of the engineering not only now, but in the beggining of the year were booted of the company, and some of them in things that should be strategic for Mozilla in the comming years.
The pressure is hard, but the good CEO´s will make a stand and defend the things that might not look very good now, but that will be key in the future..
And i wonder, how much a CEO or boarding that could go deeper in the engineering rabbit hole could have helped into making the right decisions.
Again, i dont know very well the CEO or the board, all i see is the chess pieces moving from here.. and from what i can see, joining the pieces of the pass movements, i have a sour taste in my mouth that what was once one of the greateast, shiny companies of high technology, a role model to me and to many people, are starting to repeat the misguided steps of companies like Yahoo.
Sorry for "the MBA´s", as i dont want to offend anyone, but what i meant is the pattern in the moves, it looks a lot like things Ballmer would do, not Gates. And the difference lies in how much they are informed to take the good decisions they need to make.
And tech people often lack the philosophical, social and political baggage to make the good decisions on those kind of matters, so in my opinion they must be helped not chased. Because in the end this might be important to the survival of the entity, that might look good on the paper and in its image, but might be lost in its direction.
It sounded like a rant, and i think it was harsh on my part, but its because i care and really like Mozilla.
Its when the times are rough that the good decisions are really needed. so i just wonderered.. and i would love to be proved wrong, really.