3 ms·

I'm trying to figure out what someone's macbook being stuck in a bad state and needing an SPC and NVRAM reset has to do with this conversation. That happens all

I'm trying to figure out what someone's macbook being stuck in a bad state and needing an SPC and NVRAM reset has to do with this conversation. That happens all the time on macbooks. It also has no effect on whether FF's tabs come up next time. And if FF closes suddenly, you have a chance to discard your open tabs on start. Explain to me how my statement is shortsighted again?

sisu2019 8y ago
sorry, I meant to post this link: https://news.ycombinator.com/item?id=19268201 https://news.ycombinator.com/item?id=19268201
- kakarot 8y ago
  Sounds like an issue with Chrome, then. The OS handles its own state. That should be invisible to applications. Applications handle their own state. That should be invisible to the OS. As I said, Firefox offers a choice to restore or discard tabs after a bad start or shutdown of the program. If Chrome doesn't have good fail-safe mechanisms, I'm confused about what that has to do with Microsoft and what it has to do with me being shortsighted.
  - sisu2019 8y ago
    In this case it's an issue with Chrome, yes, very good. In other cases it could be an issue with firefox. You won't get into save mode when your OS simply locks up on boot because the bug that causes the lock up is in a tab that auto-restored firefox restores for you. /I am/ a bit confused how it's so hard for you to understand the larger issue, especially since you know about two examples that already illustrated the point.
    - kakarot 8y ago
      > You won't get into save mode when your OS simply locks up on boot because the bug that causes the lock up is in a tab that auto-restored firefox restores for you. I'm a bit confused about why you don't perceive the above scenario as a software bug that needs to be fixed, specifically a programmer error dealing with compartmentalization. Compounded further by another error when entering safe mode, as safe mode by design doesn't load any non-essential programs on start. Neither would Linux single-user mode without X. You should always be able to enter safe mode if a particular non-system program is messing things up. That's the point. Anything else is a bug and deviates from spec. If a programmer tried to argue with me that the spec was wrong in this case, they would be fired on the spot. Firefox also doesn't auto restore tabs if it was shutdown improperly, so what you describe shouldn't happen. Has it ever happened to you? Or are you just arguing against a strawman?
      - sisu2019 8y ago
        Yes of course it's a bug. Maybe read what I've written about this in my OP. Everything you mentioned is predicated on things working as intended when I am talking about working around bugs users have no control over this whole time. For those cases, like in the examples, we need to keep the real hard reset that's been there the entire history of personal computing. > Firefox also doesn't auto restore tabs if it was shutdown improperly Okay how do you think that is implemented? There is no magic fairy dust that settles in the hard drive, it's code. Code with bugs, code with limitations. Safe mode is code, code with bugs and limitations. With your restore-reboot you now rely on two components to not fail in every single program that could cause something like the examples mentioned. Amazing design, let's make airplanes like that. > they would be fired on the spot spare me the power tripping garbage, instead try to show some comprehension > strawman You don't seem to know what a strawman is.
        kakarot 8y ago
        You design software according to intended spec, and then provide safeguards for each module when it makes sense. You don't design software on the premise that nothing works as intended and spec is thrown out of the window. If your software is full of bugs then you've failed on multiple levels and that's just one of them. And in this case, the spec required by me and other consumers is that application and OS state remain completely independent.

sisu2019 8y ago

sorry, I meant to post this link: https://news.ycombinator.com/item?id=19268201 https://news.ycombinator.com/item?id=19268201

kakarot 8y ago
Sounds like an issue with Chrome, then. The OS handles its own state. That should be invisible to applications. Applications handle their own state. That should be invisible to the OS. As I said, Firefox offers a choice to restore or discard tabs after a bad start or shutdown of the program. If Chrome doesn't have good fail-safe mechanisms, I'm confused about what that has to do with Microsoft and what it has to do with me being shortsighted.
- sisu2019 8y ago
  In this case it's an issue with Chrome, yes, very good. In other cases it could be an issue with firefox. You won't get into save mode when your OS simply locks up on boot because the bug that causes the lock up is in a tab that auto-restored firefox restores for you. /I am/ a bit confused how it's so hard for you to understand the larger issue, especially since you know about two examples that already illustrated the point.
  - kakarot 8y ago
    > You won't get into save mode when your OS simply locks up on boot because the bug that causes the lock up is in a tab that auto-restored firefox restores for you. I'm a bit confused about why you don't perceive the above scenario as a software bug that needs to be fixed, specifically a programmer error dealing with compartmentalization. Compounded further by another error when entering safe mode, as safe mode by design doesn't load any non-essential programs on start. Neither would Linux single-user mode without X. You should always be able to enter safe mode if a particular non-system program is messing things up. That's the point. Anything else is a bug and deviates from spec. If a programmer tried to argue with me that the spec was wrong in this case, they would be fired on the spot. Firefox also doesn't auto restore tabs if it was shutdown improperly, so what you describe shouldn't happen. Has it ever happened to you? Or are you just arguing against a strawman?
    - sisu2019 8y ago
      Yes of course it's a bug. Maybe read what I've written about this in my OP. Everything you mentioned is predicated on things working as intended when I am talking about working around bugs users have no control over this whole time. For those cases, like in the examples, we need to keep the real hard reset that's been there the entire history of personal computing. > Firefox also doesn't auto restore tabs if it was shutdown improperly Okay how do you think that is implemented? There is no magic fairy dust that settles in the hard drive, it's code. Code with bugs, code with limitations. Safe mode is code, code with bugs and limitations. With your restore-reboot you now rely on two components to not fail in every single program that could cause something like the examples mentioned. Amazing design, let's make airplanes like that. > they would be fired on the spot spare me the power tripping garbage, instead try to show some comprehension > strawman You don't seem to know what a strawman is.
      - kakarot 8y ago
        You design software according to intended spec, and then provide safeguards for each module when it makes sense. You don't design software on the premise that nothing works as intended and spec is thrown out of the window. If your software is full of bugs then you've failed on multiple levels and that's just one of them. And in this case, the spec required by me and other consumers is that application and OS state remain completely independent.