5 ms·

They could at least have included Pocket and Hello as normal addons.

by dagurp 11y ago

sp332 11y ago
Parsing and loading an add-on would increase startup time. If you haven't put in your Pocket credentials, it's a single "if" statement that has to be evaluated. Having Pocket installed is the common case and the one that should be optimized for. Are you worried that data will go to Pocket even if you don't log in?
- kuschku 11y ago
  No, I just want 0 code of it on my system. Even if it’s never executed.
  - pjc50 11y ago
    That's a fairly high level of outrage, why?
    - stonogo 11y ago
      How is "not installing unused software" considered to be "outrage"?
      - pjc50 11y ago
        It's a minor browser feature. Most people will never use e.g. the mozilla developer tools.
        pdkl95 11y ago
        I find it hard to believe that someone with significant karma on HN doesn't understand surveillance-as-a-business-model. Just being a yet another Service As A Software Substitute[1] is bad enough, but in this case these rent-seekers are exploiting user ignorance (it's a "dark pattern"). Firefox has always been a local app, with remote features requiring the user to opt-in to an extension. The distinction between Mozilla "only providing a button" and the actual feature that loads from the remote SaaSS only exists for people that understand these technologies. Mozilla is being especially hypocritical with the integration of these features. During their previous projects (e.g. Australis) Mozilla pushed a lot of previously-integrated features into extensions. This caused problems for a LOT of people, but I reluctantly supported it because a minimal core with most features as plugins is generally a good design. For them to turn around an integrate a plugin that baits people into using spyware is outrageous - and somewhat suspicious. That button needs to be removed because it's an attractive nuisance[2]. [1] http://www.gnu.org/philosophy/who-does-that-server-really-serve.en.html http://www.gnu.org/philosophy/who-does-that-server-really-se... [2] https://en.wikipedia.org/wiki/Attractive_nuisance_doctrine https://en.wikipedia.org/wiki/Attractive_nuisance_doctrine
        Dylan16807 11y ago
        >This caused problems for a LOT of people, but I reluctantly supported it because a minimal core with most features as plugins is generally a good design. For them to turn around an integrate a plugin that baits people into using spyware is outrageous - and somewhat suspicious. Bah. It's basically a lightweight extension, and will soon get packaged as one. It's not integrated into the core, which means the main anti-bloat principles are still upheld. And it's not like it hides the process of making an account. If you want to sync things, you need a server. Not suspicious.
        kuschku 11y ago
        > If you want to sync things, you need a server. So why not use about:reader and Firefox Sync?
        Dylan16807 11y ago
        If pocket is "spyware", I'm not sure Firefox Sync is going to be considered all that much better.
        pdkl95 11y ago
        Sync was useful in it's original form that was encrypted entirely client-side. With the recent(-ish) changes, Sync should be considered spyware (or at least having the potential to be hijacked into spyware).
  - icebraining 11y ago
    FWIW, the code is still developed by Mozilla and open source; it just calls Pocket APIs.
    - kuschku 11y ago
      And I’d have to write a patchset and maintain it to get rid of it. It’s as if my browser had a copy of Wolfenstein3D integrated. Funny easteregg, but just a waste of development and testing time, and a waste of storage space. Every line of code costs time and money in testing. And here it costs me time every few days to fix new issues that were introduced when the code changed, to update my .patch, reapply it, recompile, repackage. Every few days. All the time. And when the regression with Gtk3.14 -> 3.18 regarding Drag-and-Drop is still not fixed, but they have time and money to implement, test and bugfix this, sorry, but then I am seriously out of options for running a stable, customizable no-bullshit browser.
      - scrollaway 11y ago
        > sorry, but then I am seriously out of options. Wow this is some level of entitlement I have rarely seen before. You have the option to contribute. You're not the only one who doesn't want pocket.
        kuschku 11y ago
        I can easily submit a patch to remove pocket and place about:reader more prominently. Is it going to get accepted into Firefox? No. Just like the last 5 times people tried to do this. What I currently do is constantly keeping my patch up to date and recompiling Firefox for my Desktop and Laptop (ARCH and Kubuntu) every night based on the current source from the latest trunk release. But it’s not nearly worth the effort to do this when the browser could easily accept one of the many patches people have written by now to get rid of pocket as part of the system and to move it into an addon. Same with the ad-ridden new tab page. Put that stuff into an addon and allow me to uninstall it. Expose the EME DRM feature as plugin on the plugins page, and allow me to uninstall it (I do not know if this is yet the case, I haven’t checked). I don’t want to have to maintain a huge patchset just to run my browser. I already have to hack-fix bugs like the before mentioned drag-and-drop bug myself (or downgrade to Gtk3.14).
  - calgoo 11y ago
    You can strip it from the code, its not too hard. The issue is with the release cycle, an having to modify the code each time. I wish they would just make a build without the crapware bundled. edit: to -> too
- stonogo 11y ago
  > Having Pocket installed is the common case and the one that should be optimized for. This statement is utterly wrong. Pocket has 14,000,000 users. Firefox has between 125,000,000 and 150,000,000 users. Assuming every single Pocket user is also a Firefox user, you're now optimizing for 10% of your users. This is clearly stupid. It's also incorrect to claim that parsing and loading an addon would increase startup time. It's already loading the Pocket button; moving that code into an addon would not affect startup time at all. What it would do is allow users to disable or remove the Pocket integration -- which of course Pocket is paying Mozilla to prevent. Don't pretend this is a technical decision. It is a business transaction.
  - sp332 11y ago
    The average add-on adds 10% to Firefox's startup time. https://blog.mozilla.org/addons/2011/04/01/improving-add-on-performance/ https://blog.mozilla.org/addons/2011/04/01/improving-add-on-... If this were an especially efficient one it might not be that bad, but it does take a performance hit.
    - mixmastamyk 11y ago
      Not when uninstalled/disabled.

They could at least have included Pocket and Hello as normal addons.

by dagurp 11y ago

They could at least have included Pocket and Hello as normal addons.

sp332 11y ago

Parsing and loading an add-on would increase startup time. If you haven't put in your Pocket credentials, it's a single "if" statement that has to be evaluated. Having Pocket installed is the common case and the one that should be optimized for. Are you worried that data will go to Pocket even if you don't log in?

kuschku 11y ago
No, I just want 0 code of it on my system. Even if it’s never executed.
- pjc50 11y ago
  That's a fairly high level of outrage, why?
  - stonogo 11y ago
    How is "not installing unused software" considered to be "outrage"?
    - pjc50 11y ago
      It's a minor browser feature. Most people will never use e.g. the mozilla developer tools.
      - pdkl95 11y ago
        I find it hard to believe that someone with significant karma on HN doesn't understand surveillance-as-a-business-model. Just being a yet another Service As A Software Substitute[1] is bad enough, but in this case these rent-seekers are exploiting user ignorance (it's a "dark pattern"). Firefox has always been a local app, with remote features requiring the user to opt-in to an extension. The distinction between Mozilla "only providing a button" and the actual feature that loads from the remote SaaSS only exists for people that understand these technologies. Mozilla is being especially hypocritical with the integration of these features. During their previous projects (e.g. Australis) Mozilla pushed a lot of previously-integrated features into extensions. This caused problems for a LOT of people, but I reluctantly supported it because a minimal core with most features as plugins is generally a good design. For them to turn around an integrate a plugin that baits people into using spyware is outrageous - and somewhat suspicious. That button needs to be removed because it's an attractive nuisance[2]. [1] http://www.gnu.org/philosophy/who-does-that-server-really-serve.en.html http://www.gnu.org/philosophy/who-does-that-server-really-se... [2] https://en.wikipedia.org/wiki/Attractive_nuisance_doctrine https://en.wikipedia.org/wiki/Attractive_nuisance_doctrine
        Dylan16807 11y ago
        >This caused problems for a LOT of people, but I reluctantly supported it because a minimal core with most features as plugins is generally a good design. For them to turn around an integrate a plugin that baits people into using spyware is outrageous - and somewhat suspicious. Bah. It's basically a lightweight extension, and will soon get packaged as one. It's not integrated into the core, which means the main anti-bloat principles are still upheld. And it's not like it hides the process of making an account. If you want to sync things, you need a server. Not suspicious.
        kuschku 11y ago
        > If you want to sync things, you need a server. So why not use about:reader and Firefox Sync?
        Dylan16807 11y ago
        If pocket is "spyware", I'm not sure Firefox Sync is going to be considered all that much better.
        pdkl95 11y ago
        Sync was useful in it's original form that was encrypted entirely client-side. With the recent(-ish) changes, Sync should be considered spyware (or at least having the potential to be hijacked into spyware).
- icebraining 11y ago
  FWIW, the code is still developed by Mozilla and open source; it just calls Pocket APIs.
  - kuschku 11y ago
    And I’d have to write a patchset and maintain it to get rid of it. It’s as if my browser had a copy of Wolfenstein3D integrated. Funny easteregg, but just a waste of development and testing time, and a waste of storage space. Every line of code costs time and money in testing. And here it costs me time every few days to fix new issues that were introduced when the code changed, to update my .patch, reapply it, recompile, repackage. Every few days. All the time. And when the regression with Gtk3.14 -> 3.18 regarding Drag-and-Drop is still not fixed, but they have time and money to implement, test and bugfix this, sorry, but then I am seriously out of options for running a stable, customizable no-bullshit browser.
    - scrollaway 11y ago
      > sorry, but then I am seriously out of options. Wow this is some level of entitlement I have rarely seen before. You have the option to contribute. You're not the only one who doesn't want pocket.
      - kuschku 11y ago
        I can easily submit a patch to remove pocket and place about:reader more prominently. Is it going to get accepted into Firefox? No. Just like the last 5 times people tried to do this. What I currently do is constantly keeping my patch up to date and recompiling Firefox for my Desktop and Laptop (ARCH and Kubuntu) every night based on the current source from the latest trunk release. But it’s not nearly worth the effort to do this when the browser could easily accept one of the many patches people have written by now to get rid of pocket as part of the system and to move it into an addon. Same with the ad-ridden new tab page. Put that stuff into an addon and allow me to uninstall it. Expose the EME DRM feature as plugin on the plugins page, and allow me to uninstall it (I do not know if this is yet the case, I haven’t checked). I don’t want to have to maintain a huge patchset just to run my browser. I already have to hack-fix bugs like the before mentioned drag-and-drop bug myself (or downgrade to Gtk3.14).
- calgoo 11y ago
  You can strip it from the code, its not too hard. The issue is with the release cycle, an having to modify the code each time. I wish they would just make a build without the crapware bundled. edit: to -> too
stonogo 11y ago
> Having Pocket installed is the common case and the one that should be optimized for. This statement is utterly wrong. Pocket has 14,000,000 users. Firefox has between 125,000,000 and 150,000,000 users. Assuming every single Pocket user is also a Firefox user, you're now optimizing for 10% of your users. This is clearly stupid. It's also incorrect to claim that parsing and loading an addon would increase startup time. It's already loading the Pocket button; moving that code into an addon would not affect startup time at all. What it would do is allow users to disable or remove the Pocket integration -- which of course Pocket is paying Mozilla to prevent. Don't pretend this is a technical decision. It is a business transaction.
- sp332 11y ago
  The average add-on adds 10% to Firefox's startup time. https://blog.mozilla.org/addons/2011/04/01/improving-add-on-performance/ https://blog.mozilla.org/addons/2011/04/01/improving-add-on-... If this were an especially efficient one it might not be that bad, but it does take a performance hit.
  - mixmastamyk 11y ago
    Not when uninstalled/disabled.