5 ms·

Just making a browser? It's the piece of software I'd least be willing to touch. Even Microsoft gave up and switched their core to Chromium. Considering that,

Just making a browser? It's the piece of software I'd least be willing to touch. Even Microsoft gave up and switched their core to Chromium. Considering that, it's quite amazing that Mozilla is capable of not just making a browser, but also an email client, with a budget of 500 million.

blub 7y ago
MS didn't give up, their new strategy is similar to Google's, so they teamed up with Google to make Chrome an even better spying machine. They just have to change the telemetry URL in the source and replace the Google log in with their own.
brianpgordon 7y ago
Microsoft still makes a browser; they just don't make their own rendering engine anymore. Which makes sense because it's hard and Blink works just fine. I wouldn't mind if Mozilla did the same thing. The Vivaldi folks are using that strategy and managing to make the best web browser out there with what I assume must be a microscopic fraction of $500 million/yr. And I don't share your opinion of how far a $500 million budget goes. They're not rewriting the entire thing every year.
- AsyncAwait 7y ago
  > I wouldn't mind if Mozilla did the same thing. Then there would literally be no point to Firefox existing as a counterbalance to Google deciding the web's future.
  - brianpgordon 7y ago
    Theoretically they could still focus on developing a privacy-first browser on top of the core engine, and continue to influence web standards by implementing them as plugins with some glue patches maintained as a running fork on top of the Chromium code. As it is now, the fact that Firefox uses Quantum instead of Blink and SpiderMonkey instead of V8 isn't really a differentiator for the vast majority of users. There's a vague sense that a technology monoculture is unhealthy, but I would argue that the real threat is a browser monoculture, and Mozilla apparently can't stop that from happening with their current strategy because Firefox is losing.
    - GeekyBear 7y ago
      Given the rise of devices with many cores that are not particularly powerful, technology that moves web page rendering to a multithreaded model, like Servo and WebRender will only become more important, no?
    - klingonopera 7y ago
      I disagree, it's a good thing that Mozilla develops their own engine, this way, when we give either browser some data, we can verify that both engines are doing what they're supposed to be doing by comparing the results, and they should be the same. If we switch to a single, universal engine for all browsers, bugs, faults and errors in that engine may slip through, and no one would really notice for absence of comparison.
    - yjftsjthsd-h 7y ago
      Brave is trying that approach. It feels like an uphill battle, but time will tell.
- mda 7y ago
  Chromium based Edge is definitely much more than the rendering engine. Do they use their own networking, process management, Javascript engine, extension system? It is probably still 80-90% chromium.
  - brianpgordon 7y ago
    Rendering/layout engine is an admittedly sloppy term referring to the whole stack, except for the JavaScript engine. I think my point is still valid, that Microsoft hasn't given up on their browser, just most of the technology underlying it - which their users don't care about. The fact that Edge is a pretty decent epub reader (for now... https://www.thurrott.com/cloud/web-browsers/microsoft-edge/213350/microsoft-is-killing-epub-support-in-edge-classic https://www.thurrott.com/cloud/web-browsers/microsoft-edge/2...) isn't necessarily tied to the fact that it's EdgeHTML based. If they wanted to, Microsoft could implement the same feature set on top of Chromium.
mcbits 7y ago
The most recent financial statement I could see is from 2017: https://assets.mozilla.net/annualreport/2017/mozilla-fdn-2017-fs-short-form-final-0927.pdf https://assets.mozilla.net/annualreport/2017/mozilla-fdn-201... They spent $253 million (45% of revenue) on software development, presumably the bulk going to Firefox. Other major expenses were $66 million for marketing (yay Moz://a!), $72 million for the dubious "general and administrative" category, a few million here and there for other things, and the rest went into the $500 million nest egg. So they could certainly spend a lot more on Firefox if needed, but it's not evident to me that Firefox's weaknesses are money related.
- brianpgordon 7y ago
  > They spent $253 million (45% of revenue) on software development, presumably the bulk going to Firefox. I'm curious if anyone can locate information on how much of that was spent specifically on Firefox, the browser. It seems like Mozilla was working on a bunch of peripheral stuff during that 2016-2017 period. * Work on Firefox OS was still ongoing through 1H 2016. (https://groups.google.com/forum/#!msg/mozilla.dev.fxos/FoAwifahNPY/Lppm0VHVBAAJ https://groups.google.com/forum/#!msg/mozilla.dev.fxos/FoAwi...) * They were building a free subscription service to check HIBP for your info. (https://monitor.firefox.com/ https://monitor.firefox.com/) * They were building a file-sharing service (https://github.com/mozilla/send/blob/master/CHANGELOG.md https://github.com/mozilla/send/blob/master/CHANGELOG.md) * They were building a new VR-first browser and content aggregator (https://blog.mozilla.org/blog/2018/09/18/firefox-reality-now-available/ https://blog.mozilla.org/blog/2018/09/18/firefox-reality-now...)
  - pbhjpbhj 7y ago
    >They were building a free subscription service to check HIBP for your info. (https://monitor.firefox.com/ https://monitor.firefox.com/) // Wasn't HIBP a single person. So presumably a checking system could have been churned out by, being generous, a team of 6 on a mean-average $100,000 wage. Lets say $1.2M per year (doubled to account for facilities, etc.) as a top end. Realistically surely that's a couple of months work for such a team? Looking at the Send github (https://github.com/mozilla/send/graphs/contributors https://github.com/mozilla/send/graphs/contributors) it's essentially 2 people, about 1.5 person years in the commits. The first and fourth items you list sound like absolute white elephants that management should have allowed only as "we'll give you a full time dev wage for it and spare facilities, show us what you can do and we might take it on". So, what have we got left $240M?

blub 7y ago

MS didn't give up, their new strategy is similar to Google's, so they teamed up with Google to make Chrome an even better spying machine. They just have to change the telemetry URL in the source and replace the Google log in with their own.

brianpgordon 7y ago

Microsoft still makes a browser; they just don't make their own rendering engine anymore. Which makes sense because it's hard and Blink works just fine. I wouldn't mind if Mozilla did the same thing. The Vivaldi folks are using that strategy and managing to make the best web browser out there with what I assume must be a microscopic fraction of $500 million/yr. And I don't share your opinion of how far a $500 million budget goes. They're not rewriting the entire thing every year.

AsyncAwait 7y ago
> I wouldn't mind if Mozilla did the same thing. Then there would literally be no point to Firefox existing as a counterbalance to Google deciding the web's future.
- brianpgordon 7y ago
  Theoretically they could still focus on developing a privacy-first browser on top of the core engine, and continue to influence web standards by implementing them as plugins with some glue patches maintained as a running fork on top of the Chromium code. As it is now, the fact that Firefox uses Quantum instead of Blink and SpiderMonkey instead of V8 isn't really a differentiator for the vast majority of users. There's a vague sense that a technology monoculture is unhealthy, but I would argue that the real threat is a browser monoculture, and Mozilla apparently can't stop that from happening with their current strategy because Firefox is losing.
  - GeekyBear 7y ago
    Given the rise of devices with many cores that are not particularly powerful, technology that moves web page rendering to a multithreaded model, like Servo and WebRender will only become more important, no?
  - klingonopera 7y ago
    I disagree, it's a good thing that Mozilla develops their own engine, this way, when we give either browser some data, we can verify that both engines are doing what they're supposed to be doing by comparing the results, and they should be the same. If we switch to a single, universal engine for all browsers, bugs, faults and errors in that engine may slip through, and no one would really notice for absence of comparison.
  - yjftsjthsd-h 7y ago
    Brave is trying that approach. It feels like an uphill battle, but time will tell.
mda 7y ago
Chromium based Edge is definitely much more than the rendering engine. Do they use their own networking, process management, Javascript engine, extension system? It is probably still 80-90% chromium.
- brianpgordon 7y ago
  Rendering/layout engine is an admittedly sloppy term referring to the whole stack, except for the JavaScript engine. I think my point is still valid, that Microsoft hasn't given up on their browser, just most of the technology underlying it - which their users don't care about. The fact that Edge is a pretty decent epub reader (for now... https://www.thurrott.com/cloud/web-browsers/microsoft-edge/213350/microsoft-is-killing-epub-support-in-edge-classic https://www.thurrott.com/cloud/web-browsers/microsoft-edge/2...) isn't necessarily tied to the fact that it's EdgeHTML based. If they wanted to, Microsoft could implement the same feature set on top of Chromium.

mcbits 7y ago

The most recent financial statement I could see is from 2017: https://assets.mozilla.net/annualreport/2017/mozilla-fdn-2017-fs-short-form-final-0927.pdf https://assets.mozilla.net/annualreport/2017/mozilla-fdn-201... They spent $253 million (45% of revenue) on software development, presumably the bulk going to Firefox. Other major expenses were $66 million for marketing (yay Moz://a!), $72 million for the dubious "general and administrative" category, a few million here and there for other things, and the rest went into the $500 million nest egg. So they could certainly spend a lot more on Firefox if needed, but it's not evident to me that Firefox's weaknesses are money related.

brianpgordon 7y ago
> They spent $253 million (45% of revenue) on software development, presumably the bulk going to Firefox. I'm curious if anyone can locate information on how much of that was spent specifically on Firefox, the browser. It seems like Mozilla was working on a bunch of peripheral stuff during that 2016-2017 period. * Work on Firefox OS was still ongoing through 1H 2016. (https://groups.google.com/forum/#!msg/mozilla.dev.fxos/FoAwifahNPY/Lppm0VHVBAAJ https://groups.google.com/forum/#!msg/mozilla.dev.fxos/FoAwi...) * They were building a free subscription service to check HIBP for your info. (https://monitor.firefox.com/ https://monitor.firefox.com/) * They were building a file-sharing service (https://github.com/mozilla/send/blob/master/CHANGELOG.md https://github.com/mozilla/send/blob/master/CHANGELOG.md) * They were building a new VR-first browser and content aggregator (https://blog.mozilla.org/blog/2018/09/18/firefox-reality-now-available/ https://blog.mozilla.org/blog/2018/09/18/firefox-reality-now...)
- pbhjpbhj 7y ago
  >They were building a free subscription service to check HIBP for your info. (https://monitor.firefox.com/ https://monitor.firefox.com/) // Wasn't HIBP a single person. So presumably a checking system could have been churned out by, being generous, a team of 6 on a mean-average $100,000 wage. Lets say $1.2M per year (doubled to account for facilities, etc.) as a top end. Realistically surely that's a couple of months work for such a team? Looking at the Send github (https://github.com/mozilla/send/graphs/contributors https://github.com/mozilla/send/graphs/contributors) it's essentially 2 people, about 1.5 person years in the commits. The first and fourth items you list sound like absolute white elephants that management should have allowed only as "we'll give you a full time dev wage for it and spare facilities, show us what you can do and we might take it on". So, what have we got left $240M?