11 ms·

The right thing to do would have been to work on those extensions BEFORE breaking users' experience.

digi_owl 9y ago
Sadly not how the FOSS world works any longer. All that matters is the "shiny" (often wrapped in some kind of "social consciousness" claptrap), and those of us that has come to rely on existing behavior has to either suck it up, move on, or fork (And even forks struggle)...
- amaranth 9y ago
  I don't remember the FOSS world _ever_ working like you seem to think it should. The CADT Model[1] isn't exactly new and it wasn't really new when jwz coined the term either. Only the kernel, POSIX-y things, and enterprise software provides what you're asking for. And forks of Firefox struggle because writing a browser is really really hard and they can't keep up with the changes needed. [1] https://www.jwz.org/doc/cadt.html https://www.jwz.org/doc/cadt.html
nicoburns 9y ago
Totally agree with this. I think it was a mistake on Mozilla's part not to put more effort into extension parity before the switch-over. But I'm relatively optimistic for the future.
- throwanem 9y ago
  Have you seen the Firefox backlog? If they'd gone for 100% API parity before the switch, we'd have had to wait another three years for e10s, and I don't know if that would have been survivable. I'm not happy about the situation as it stands, and I don't expect anyone else to be. I would be a lot more not happy about having to switch to Chrome because Firefox had become unusably slow - which it had already more or less done, before e10s started to land this year. Absent that I'd probably have had to switch already, just to be reliably able to get work done. And then there would be no one on the teams I've worked on who cared at all about maintaining Firefox compatibility, because everyone else already switched years ago. Less than ideal though the status quo be, I have a hard time seeing how any of that would be preferable.
  - MadWombat 9y ago
    > I have a hard time seeing how any of that would be preferable. Not to you. But not everyone cares about the same things. I have used Firefox as my main browser ever since it split off from Mozilla Suite. Personally, I have never, in the whole history of Firefox had a problem with its speed. At various points I might have had some problems with stability, some problems with compatibility and some problems with memory usage. But speed is just not an issue for me. Even in the "slowest" browsers most of the time is spent on network latency anyway. But I have, over the years, customized my Firefox experience to some degree. I do not do anything crazy, I don't have a zillion of extensions, but I am really used to the few I do have. And this release broke those with no sufficient replacement. And the release was distributed as automatic upgrade, so I was not even asked if I want to upgrade. A warning of some sort would be nice before breaking so much functionality. And now that the upgrade happened, there is no clear way on how to revert it. A quick google search tells me that downgrading runs a chance of corrupting my profile. I might have to risk it anyway, since after using FF57 for a day at work, I feel that my productivity and workflow are seriously impaired. At this point I am not sure whether I should stay with the LTR version of FF, switch to an FF derivative like Waterfox or change browsers altogether. The FF57 experience is so vastly different from my pre-57 workflow that I might as well be using a different browser already. There were important features to FF that kept me a loyal user through many years. Those features are gone now.
    - throwanem 9y ago
      I've been using Firefox for almost as long. Perf has been an issue for half a decade. Anecdotal evidence and telemetry analysis both suggest it's much more common to encounter the issue than not. Don't get me wrong! I'd love to have both. But if we can't - and it seems right now that indeed we cannot - then I think perf has to be the one to pick. Otherwise the browser dies and we're all SOL. Try backing up your profile and downgrading. I'd be surprised if anything breaks. Although I had also assumed anyone bringing legacy addons into 56 would be warned before the 57 update, so maybe I'm overly optimistic here.
    - kelnos 9y ago
      > Personally, I have never, in the whole history of Firefox had a problem with its speed. That's... weird. To put it nicely. I've been using Mozilla since early in its milestone phases, and Firefox since it came out. Firefox was originally a pretty nimble browser, but fairly quickly became super slow. When Chrome came out I couldn't believe the difference in speed. Everything from UI responsiveness to rendering speed to reliability was better. At some point I moved back to Firefox for a single reason: I had an 8GB laptop and Chrome's memory usage would balloon with the number of tabs I usually keep around, to the point that it made my laptop unusable. I lived with Firefox being much slower in general because its worst-case performance was much better than Chrome's on a memory-constrained laptop. Since I force-enabled e10s a year or so ago things have gotten much better (despite the bugs). I still use Chrome every now and then for the odd website that chokes on some combination of Firefox plus the extensions I have installed, not to mention Chromecast support. But Firefox 57 is finally now faster than Chrome by default, and the difference is night and day. It's completely inconceivable to me that you suggest that Firefox hasn't had severe performance problems. The fact that Mozilla has been focusing so hard on perf for nearly a decade now is more than enough evidence to me that it's been a huge problem.
- developer2 9y ago
  Mozilla absolutely made the right decision to dump a decade's worth of legacy code to improve performance, and make the future happen now. Let's be real: had they supported legacy extensions, extension developers would ignore porting until Mozilla finally said "that's been long enough, we're removing legacy support". The result is simply that the rough transition happens now instead of a couple of years down the line. Every popular extension will be ported within a matter of weeks, worst case scenario 2-3 months for those that are unabanonded, but less popular. If you are missing a deal-breaker extension, then don't upgrade yet... how is that a bad thing? "But I want the latest and greatest, super fast Firefox now - with all my extensions!". Talk about wanting to have one's cake and eat it too. The new Firefox wouldn't be new and improved if it was backwards-compatible with the stone age.
  - HelloNurse 9y ago
    You forget the big technical issue: many web extension API features that are needed to port old Firefox extensions do not exist yet (e.g. filesystem access) or are conservatively crippled (e.g. adding buttons, menu items, other GUI elements). Your 2-3 months of frantic effort will start not immediately but in a vague future when the Firefox team feels like improving the web extension API. It is fairly obvious from API documentation and bug discussion threads that they only care for feature parity with Chrome web extensions, not with old Firefox extensions. The transition could have been managed gradually, by piece by piece replacement of the old extension API with the new web extension API (made available to old style extensions). Extension developers would have made the small changes needed to port from a deprecated old API to a new API that does the same thing, up to the final step of reorganizing without significant code changes the old extension, now relying exclusively on web extension APIs, into a web extension. Abandoned extensions would have fallen by the wayside very fast, deficient APIs would have been fixed before actual usage, and Firefox would have moved forward without betraying users.
    - Yoric 9y ago
      You're pretty much describing what Firefox has been doing for the past few years. At some point, however, the rest of the world keeps moving and, no matter how painful for everyone involved, it becomes unfeasible to wait for all add-ons to be ported (or even portable).
- kelnos 9y ago
  They put 8 years of effort into extension parity. At some point you end up deciding that the breakage is less important than fixing foundational problems that cause people to leave your platform in droves due to the lack of ability to fix performance and security problems. There are a few extensions I use that have stopped working that don't have a replacement (yet?), which sucks, but the tradeoff is well worth it.
DiThi 9y ago
Most of these were done by users scratching their itch. And I'm confident that's exactly why we'll get most of those experiences back.
- CaptSpify 9y ago
  Not necessarily. WebExtensions by design don't work on firefox pages, nor do they have full access to the browser like the old addons used to have. There are a few requests our there to get the needed access added, but Mozilla is basically going to let those rot: https://bugzilla.mozilla.org/show_bug.cgi?id=1215061 https://bugzilla.mozilla.org/show_bug.cgi?id=1215061
Tomte 9y ago
Mozilla doesn't create all those extensions. And the extensions' authors had, what, two years advance warning or so? Criticize the real culpits, not Mozilla.
- wlesieutre 9y ago
  If the required WebExtension APIs don't exist (per nicoburns' comment) I'm not sure what you want the extension authors to do about it. "Complain loudly" was the strategy that got Mozilla to implement Tree Style Tabs as a WebExtension feature so that the addon could be rewritten for FF57, but there are presumably still legacy extension capabilities that haven't gotten that treatment.
- samastur 9y ago
  If required APIs don't exist any more, how is that extension author's fault? (Tab Groups user here)
  - Tomte 9y ago
    If they don't exist, sure, but most of the bickering is "I want XUL and nothing else". And how many of those lamenting the absence of some API have actually told Mozilla what they need? By all accounts Mozilla has been very responsive and helpful. Not everything was possible, but the problem lies mostly not on their side.
    - throwanem 9y ago
      > most of the bickering is "I want XUL and nothing else" I don't know that that's fair. The discussions I've seen and participated in have revolved around equivalency, not identity. > how many of those lamenting the absence of some API have actually told Mozilla what they need? The Vimperator devs and some users have been very clear on that point. > Mozilla has been very responsive and helpful. Not everything was possible, but the problem lies mostly not on their side. Mozilla has indeed been responsive and as helpful as they can be given the constraints under which they operate. But extension developers aren't really to blame here either, because if WebEx doesn't include an API you can use to do what you need to do, then what option do you have? There's frustration on all sides - Mozilla devs, extension devs, and users. But we're all pulling together toward the same ultimate goal. I don't know what improvement anyone expects to elicit by trying to assign blame.
dralley 9y ago
https://news.ycombinator.com/item?id=15696184 https://news.ycombinator.com/item?id=15696184

Sadly not how the FOSS world works any longer. All that matters is the "shiny" (often wrapped in some kind of "social consciousness" claptrap), and those of us that has come to rely on existing behavior has to either suck it up, move on, or fork (And even forks struggle)...

amaranth 9y ago
I don't remember the FOSS world _ever_ working like you seem to think it should. The CADT Model[1] isn't exactly new and it wasn't really new when jwz coined the term either. Only the kernel, POSIX-y things, and enterprise software provides what you're asking for. And forks of Firefox struggle because writing a browser is really really hard and they can't keep up with the changes needed. [1] https://www.jwz.org/doc/cadt.html https://www.jwz.org/doc/cadt.html

nicoburns 9y ago

Totally agree with this. I think it was a mistake on Mozilla's part not to put more effort into extension parity before the switch-over. But I'm relatively optimistic for the future.

throwanem 9y ago
Have you seen the Firefox backlog? If they'd gone for 100% API parity before the switch, we'd have had to wait another three years for e10s, and I don't know if that would have been survivable. I'm not happy about the situation as it stands, and I don't expect anyone else to be. I would be a lot more not happy about having to switch to Chrome because Firefox had become unusably slow - which it had already more or less done, before e10s started to land this year. Absent that I'd probably have had to switch already, just to be reliably able to get work done. And then there would be no one on the teams I've worked on who cared at all about maintaining Firefox compatibility, because everyone else already switched years ago. Less than ideal though the status quo be, I have a hard time seeing how any of that would be preferable.
- MadWombat 9y ago
  > I have a hard time seeing how any of that would be preferable. Not to you. But not everyone cares about the same things. I have used Firefox as my main browser ever since it split off from Mozilla Suite. Personally, I have never, in the whole history of Firefox had a problem with its speed. At various points I might have had some problems with stability, some problems with compatibility and some problems with memory usage. But speed is just not an issue for me. Even in the "slowest" browsers most of the time is spent on network latency anyway. But I have, over the years, customized my Firefox experience to some degree. I do not do anything crazy, I don't have a zillion of extensions, but I am really used to the few I do have. And this release broke those with no sufficient replacement. And the release was distributed as automatic upgrade, so I was not even asked if I want to upgrade. A warning of some sort would be nice before breaking so much functionality. And now that the upgrade happened, there is no clear way on how to revert it. A quick google search tells me that downgrading runs a chance of corrupting my profile. I might have to risk it anyway, since after using FF57 for a day at work, I feel that my productivity and workflow are seriously impaired. At this point I am not sure whether I should stay with the LTR version of FF, switch to an FF derivative like Waterfox or change browsers altogether. The FF57 experience is so vastly different from my pre-57 workflow that I might as well be using a different browser already. There were important features to FF that kept me a loyal user through many years. Those features are gone now.
  - throwanem 9y ago
    I've been using Firefox for almost as long. Perf has been an issue for half a decade. Anecdotal evidence and telemetry analysis both suggest it's much more common to encounter the issue than not. Don't get me wrong! I'd love to have both. But if we can't - and it seems right now that indeed we cannot - then I think perf has to be the one to pick. Otherwise the browser dies and we're all SOL. Try backing up your profile and downgrading. I'd be surprised if anything breaks. Although I had also assumed anyone bringing legacy addons into 56 would be warned before the 57 update, so maybe I'm overly optimistic here.
  - kelnos 9y ago
    > Personally, I have never, in the whole history of Firefox had a problem with its speed. That's... weird. To put it nicely. I've been using Mozilla since early in its milestone phases, and Firefox since it came out. Firefox was originally a pretty nimble browser, but fairly quickly became super slow. When Chrome came out I couldn't believe the difference in speed. Everything from UI responsiveness to rendering speed to reliability was better. At some point I moved back to Firefox for a single reason: I had an 8GB laptop and Chrome's memory usage would balloon with the number of tabs I usually keep around, to the point that it made my laptop unusable. I lived with Firefox being much slower in general because its worst-case performance was much better than Chrome's on a memory-constrained laptop. Since I force-enabled e10s a year or so ago things have gotten much better (despite the bugs). I still use Chrome every now and then for the odd website that chokes on some combination of Firefox plus the extensions I have installed, not to mention Chromecast support. But Firefox 57 is finally now faster than Chrome by default, and the difference is night and day. It's completely inconceivable to me that you suggest that Firefox hasn't had severe performance problems. The fact that Mozilla has been focusing so hard on perf for nearly a decade now is more than enough evidence to me that it's been a huge problem.
developer2 9y ago
Mozilla absolutely made the right decision to dump a decade's worth of legacy code to improve performance, and make the future happen now. Let's be real: had they supported legacy extensions, extension developers would ignore porting until Mozilla finally said "that's been long enough, we're removing legacy support". The result is simply that the rough transition happens now instead of a couple of years down the line. Every popular extension will be ported within a matter of weeks, worst case scenario 2-3 months for those that are unabanonded, but less popular. If you are missing a deal-breaker extension, then don't upgrade yet... how is that a bad thing? "But I want the latest and greatest, super fast Firefox now - with all my extensions!". Talk about wanting to have one's cake and eat it too. The new Firefox wouldn't be new and improved if it was backwards-compatible with the stone age.
- HelloNurse 9y ago
  You forget the big technical issue: many web extension API features that are needed to port old Firefox extensions do not exist yet (e.g. filesystem access) or are conservatively crippled (e.g. adding buttons, menu items, other GUI elements). Your 2-3 months of frantic effort will start not immediately but in a vague future when the Firefox team feels like improving the web extension API. It is fairly obvious from API documentation and bug discussion threads that they only care for feature parity with Chrome web extensions, not with old Firefox extensions. The transition could have been managed gradually, by piece by piece replacement of the old extension API with the new web extension API (made available to old style extensions). Extension developers would have made the small changes needed to port from a deprecated old API to a new API that does the same thing, up to the final step of reorganizing without significant code changes the old extension, now relying exclusively on web extension APIs, into a web extension. Abandoned extensions would have fallen by the wayside very fast, deficient APIs would have been fixed before actual usage, and Firefox would have moved forward without betraying users.
  - Yoric 9y ago
    You're pretty much describing what Firefox has been doing for the past few years. At some point, however, the rest of the world keeps moving and, no matter how painful for everyone involved, it becomes unfeasible to wait for all add-ons to be ported (or even portable).
kelnos 9y ago
They put 8 years of effort into extension parity. At some point you end up deciding that the breakage is less important than fixing foundational problems that cause people to leave your platform in droves due to the lack of ability to fix performance and security problems. There are a few extensions I use that have stopped working that don't have a replacement (yet?), which sucks, but the tradeoff is well worth it.

DiThi 9y ago

Most of these were done by users scratching their itch. And I'm confident that's exactly why we'll get most of those experiences back.

CaptSpify 9y ago
Not necessarily. WebExtensions by design don't work on firefox pages, nor do they have full access to the browser like the old addons used to have. There are a few requests our there to get the needed access added, but Mozilla is basically going to let those rot: https://bugzilla.mozilla.org/show_bug.cgi?id=1215061 https://bugzilla.mozilla.org/show_bug.cgi?id=1215061

Tomte 9y ago

Mozilla doesn't create all those extensions. And the extensions' authors had, what, two years advance warning or so? Criticize the real culpits, not Mozilla.

wlesieutre 9y ago
If the required WebExtension APIs don't exist (per nicoburns' comment) I'm not sure what you want the extension authors to do about it. "Complain loudly" was the strategy that got Mozilla to implement Tree Style Tabs as a WebExtension feature so that the addon could be rewritten for FF57, but there are presumably still legacy extension capabilities that haven't gotten that treatment.
samastur 9y ago
If required APIs don't exist any more, how is that extension author's fault? (Tab Groups user here)
- Tomte 9y ago
  If they don't exist, sure, but most of the bickering is "I want XUL and nothing else". And how many of those lamenting the absence of some API have actually told Mozilla what they need? By all accounts Mozilla has been very responsive and helpful. Not everything was possible, but the problem lies mostly not on their side.
  - throwanem 9y ago
    > most of the bickering is "I want XUL and nothing else" I don't know that that's fair. The discussions I've seen and participated in have revolved around equivalency, not identity. > how many of those lamenting the absence of some API have actually told Mozilla what they need? The Vimperator devs and some users have been very clear on that point. > Mozilla has been very responsive and helpful. Not everything was possible, but the problem lies mostly not on their side. Mozilla has indeed been responsive and as helpful as they can be given the constraints under which they operate. But extension developers aren't really to blame here either, because if WebEx doesn't include an API you can use to do what you need to do, then what option do you have? There's frustration on all sides - Mozilla devs, extension devs, and users. But we're all pulling together toward the same ultimate goal. I don't know what improvement anyone expects to elicit by trying to assign blame.

dralley 9y ago

https://news.ycombinator.com/item?id=15696184 https://news.ycombinator.com/item?id=15696184