3 ms·

> The solution to today's Firefox performance problems is not Servo or Rust. It's basically engineers putting in time to fix it in the _current_ code base. If

> The solution to today's Firefox performance problems is not Servo or Rust. It's basically engineers putting in time to fix it in the _current_ code base. If you're so convinced that the correct answer is to hack more on Gecko (as if hundreds of engineers haven't been doing exactly that for a decade), then it should be easy for you to explain to me, in detail, technically, why Gecko's architecture is suitable to achieve best in class performance in the long term. > To be more constructive: Why don't you do a "Ask HN" about what stops them from using Firefox every day and you will have the answer. Don't deflect. I want you to defend the choice of continuing to hack on Gecko's C++ codebase instead of using Rust. Preferably describe specific subsystems that you think Gecko is doing better than Servo is, and what the problems you think Servo's architecture has. This should be very easy for you to do, as you seem to think it's so obviously correct that the only conclusion is that none of us even care about making the browser better.

slededit 10y ago
I'm not the original poster but I think this is dismissive of a very real problem with FireFox. As a user I've seen perceived performance regress significantly over the last few years. From the outside Servo looks like a distraction - and the results tend to confirm it. It may well be that Gecko just is that terrible and the only way to fix it is with Servo. If that's the case you guys really need to hurry up because you're loosing significant mindshare. Your organization's predecessor has also made this mistake before very publicly in the past with the Netscape rewrite.
- pcwalton 10y ago
  > From the outside Servo looks like a distraction - and the results tend to confirm it. Then show me those specific results. I'll pose it to you, too. If it's so obvious, tell me what the problems with Gecko are, specifically, and how Servo cannot solve them. > If that's the case you guys really need to hurry up because you're loosing significant mindshare. This is the first time I've been called lazy by a random HN poster. I guess there's a first time for everything. > Your organization's predecessor has also made this mistake before very publicly in the past with the Netscape rewrite. Firefox was possible because of Gecko in the first place! Do you want to be using code descended from Netscape 4, which had no dynamic restyling at all?
  - slededit 10y ago
    With respect it's not the job of the user to be an expert in the architecture of your application. The fact you won't admit there are problems identifies a real with issue with your organization. Firefox is slow, I don't know why. I was hoping the engineers working on it would know - it's scary that they do not.
    - pcwalton 10y ago
      So you don't know what the performance problems in Firefox are, but you're confident that the solution has to be a short-term fix in the existing C++ codebase. Do you realize why that does not make sense?
      - slededit 10y ago
        I suppose a better question is how did the existing code base get this bad? I work on a product 10 years older than FireFox and we've been able to make do with upgrades in place. Over time large pieces of architecture change. The difference is our customers see improvements over time not large periods of stagnation.
        pcwalton 10y ago
        That is exactly what is happening with Firefox.
        slededit 10y ago
        From the outside it doesn't look like it. That's all I'm saying - Chrome is pulling way ahead to the point I have to use it on all but my most powerful machines.

slededit 10y ago

I'm not the original poster but I think this is dismissive of a very real problem with FireFox. As a user I've seen perceived performance regress significantly over the last few years. From the outside Servo looks like a distraction - and the results tend to confirm it. It may well be that Gecko just is that terrible and the only way to fix it is with Servo. If that's the case you guys really need to hurry up because you're loosing significant mindshare. Your organization's predecessor has also made this mistake before very publicly in the past with the Netscape rewrite.

pcwalton 10y ago
> From the outside Servo looks like a distraction - and the results tend to confirm it. Then show me those specific results. I'll pose it to you, too. If it's so obvious, tell me what the problems with Gecko are, specifically, and how Servo cannot solve them. > If that's the case you guys really need to hurry up because you're loosing significant mindshare. This is the first time I've been called lazy by a random HN poster. I guess there's a first time for everything. > Your organization's predecessor has also made this mistake before very publicly in the past with the Netscape rewrite. Firefox was possible because of Gecko in the first place! Do you want to be using code descended from Netscape 4, which had no dynamic restyling at all?
- slededit 10y ago
  With respect it's not the job of the user to be an expert in the architecture of your application. The fact you won't admit there are problems identifies a real with issue with your organization. Firefox is slow, I don't know why. I was hoping the engineers working on it would know - it's scary that they do not.
  - pcwalton 10y ago
    So you don't know what the performance problems in Firefox are, but you're confident that the solution has to be a short-term fix in the existing C++ codebase. Do you realize why that does not make sense?
    - slededit 10y ago
      I suppose a better question is how did the existing code base get this bad? I work on a product 10 years older than FireFox and we've been able to make do with upgrades in place. Over time large pieces of architecture change. The difference is our customers see improvements over time not large periods of stagnation.
      - pcwalton 10y ago
        That is exactly what is happening with Firefox.
        slededit 10y ago
        From the outside it doesn't look like it. That's all I'm saying - Chrome is pulling way ahead to the point I have to use it on all but my most powerful machines.