4 ms·

> I like to know what data applications are sending or trying to send Good luck with that. Unless you are running a Commodore 64 it is unlikely that a single

> I like to know what data applications are sending or trying to send Good luck with that. Unless you are running a Commodore 64 it is unlikely that a single person can understand, inspect and make decisions on modern operating systems or even individual applications. There was a small window in which you had the option to use your computer as a 'digital typewriter' and 'sometimes send a fax', but expected and supplied functionality this day and age relies on many small components being heavily interconnected, much in the same way that social circles are interconnected, social networks (the digital variant) are based on critical mass (not technical prowess, legal status or privacy) and the amount of people that have narrow/well-defined use cases for their computers are at an all-time low making them less and less significant to cater to. If you had a Apple computer with 10.3 or newer, this was the norm. If you had macOS 9 with iTools, it was the norm as well (for a bit until it got dropped in favour of MobileMe). The old times weren't better, just different (and much less feature-rich). Great for a few power users, bad for everyone else.

stoned 4y ago
Eh… it depends on what you think is good and bad. It’s not clear to me that surveillance capitalism will be a long term good. Gadgets and social media are fun, but digital feudalism will be (is?) a lot less fun and rewarding.
- oneplane 4y ago
  No, it doesn't. The case here is "buy a consumer device, get consumer results". If you were to expect different results, that doesn't mean that the world is wrong, it means that you bought something you didn't want.
MaxBarraclough 4y ago
> Good luck with that. Unless you are running a Commodore 64 it is unlikely that a single person can understand, inspect and make decisions on modern operating systems or even individual applications. This is defeatist, and ignores the second-order advantages of Free and Open Source software. You don't need a Commodore 64, you need a decent GNU/Linux distro. In practice, it's far less common for FOSS to contain code that works against the user's interests, as the vendor/developer has no veil of ambiguity and no deniability. Only one person needs to find the troublesome code, and they can make the rest of us aware. Everyone knows this, so FOSS malware is rarely released in the first place. FOSS isn't a silver bullet (see Firefox's telemetry) but it's not the case that there's nothing you can do but use user-hostile proprietary software for everything.
- oneplane 4y ago
  > This is defeatist, and ignores the second-order advantages of Free and Open Source software. That's both a gross misrepresentation of what I wrote, and completely wrong too. It doesn't matter how open your software sources are if you run it on closed hardware. And unless you go back to ancient hardware, it will be closed enough that you do not get what you think you are getting. At the same time, it's also just realistic: if you buy something that is made to cater to a market segment that expects a lot of complex systems to appear to work together in a seamless way, supporting many features requiring many thousands of top-tier engineers to build and maintain for many years, then you, a solo user, are not likely to be able to contain the same amount of knowledge and time to see every detail and component of such a system. So no, the product, the context and biology will not allow you to contain such a system in a single mind. Same goes for what you call user-hostile. If you ask a user which OS is more user-hostile out of a list that contains any Linux-distribution, but also Windows and macOS, it doesn't matter what is technically measures, what matters is that the user will call out that Linux distribution for being user-hostile. Again, it's not about wat is technically possible, what has been done, or what some individual does for themselves, it's about the giant user base that the commercial giants are targeting with their products. As for telemetry: just because alarmists alarm about it all day long doesn't mean it's universally bad. What is bad is going it for user profiling and not telling about it, but if I need to know how my software is performing, I know ahead of time that asking my user base is a waste of time. Either I'll need automated reporting of things like crashes, performance measurements and spindumps, or I can't do much to make the software better. There are too many configurations and situations out there to realistically cover using some imagination and testing. Even on Debian, popcon is super useful, and people who complain about their stuff not making it on disk 1 are usually also the ones not submitting. You get what you pay for, and if you don't supply how you use the product, chances are you're not going to get an improvement you like. That is a choice to make, but removing it completely "because telemetry bad" is kinda dumb.
  - MaxBarraclough 4y ago
    > It doesn't matter how open your software sources are if you run it on closed hardware. It matters a great deal. This thread is about the behaviour of Finder, a proprietary software application. More broadly, while it's true that most modern hardware prevents the user from having full control, it's also the case that much is won by using only Free and Open Source software. (Last year the FSF got their servers onto fully Free Software BIOS software, but this took considerable effort as well as the use of hardware from 2012. [0]) > So no, the product, the context and biology will not allow you to contain such a system in a single mind. Agreed, but while a total understanding of the system is one way to achieve confidence that it isn't built to work against your interests, it isn't the only way. > If you ask a user which OS is more user-hostile out of a list that contains any Linux-distribution, but also Windows and macOS, it doesn't matter what is technically measures, what matters is that the user will call out that Linux distribution for being user-hostile. Again this is defeatism. Free Software desktop operating systems have some usability issues, but in my experience they're much more approachable than many people seem to think. This is especially true now that a web browser is enough to enable most of what desktop computers are used for. It's not as if the non-Free operating systems are perfect either. FOSS operating systems aren't in such a good place on mobile as they are on desktop and server, unfortunately. > As for telemetry: just because alarmists alarm about it all day long doesn't mean it's universally bad. What is bad is going it for user profiling and not telling about it Yes, I don't think anyone disagrees. HackerNews folks give Firefox a hard time over its telemetry precisely because the user isn't given a choice. They aren't even told it's going to happen, as I recall. It's particularly grating considering Mozilla's self-promotion as being uniquely pro-user. > You get what you pay for In terms of spying on their users, Debian is far better behaved than Microsoft, despite that Debian's offering is available free of charge and Microsoft's is not. [0] https://www.fsf.org/blogs/sysadmin/closing-in-on-fully-free-bioses-with-the-fsf-tech-team https://www.fsf.org/blogs/sysadmin/closing-in-on-fully-free-...
user3939382 4y ago
> Good luck with that A couple years ago I was able to intercept every connection on an Android with a mirrored switch port and Wireshark+ a root certificate. We shouldn’t be discouraging people from controlling their own machines. Certificate pinning might be trickier, I haven’t played with it yet.
- oneplane 4y ago
  It's not a case of what is technically possible, it's a case of swimming against the current while you could also choose to swim somewhere else. If you choose something that is not what you want, and you also accept that the world isn't static, but is constantly moving and due to the way it is interconnected you have to keep up at least a little bit, you will be unhappy no matter how hard you try to turn the thing you didn't want into something you did want. It might make much more sense to just get the thing you wanted in the first place, right?
  - user3939382 4y ago
    > you could also choose to swim somewhere else Agreed, we have to pick our battles. Personally I pick this one. We should exercise as much control as technically possible over these devices that are intimately linked to every aspect of our lives.

stoned 4y ago

Eh… it depends on what you think is good and bad. It’s not clear to me that surveillance capitalism will be a long term good. Gadgets and social media are fun, but digital feudalism will be (is?) a lot less fun and rewarding.

oneplane 4y ago
No, it doesn't. The case here is "buy a consumer device, get consumer results". If you were to expect different results, that doesn't mean that the world is wrong, it means that you bought something you didn't want.

MaxBarraclough 4y ago

> Good luck with that. Unless you are running a Commodore 64 it is unlikely that a single person can understand, inspect and make decisions on modern operating systems or even individual applications. This is defeatist, and ignores the second-order advantages of Free and Open Source software. You don't need a Commodore 64, you need a decent GNU/Linux distro. In practice, it's far less common for FOSS to contain code that works against the user's interests, as the vendor/developer has no veil of ambiguity and no deniability. Only one person needs to find the troublesome code, and they can make the rest of us aware. Everyone knows this, so FOSS malware is rarely released in the first place. FOSS isn't a silver bullet (see Firefox's telemetry) but it's not the case that there's nothing you can do but use user-hostile proprietary software for everything.

oneplane 4y ago
> This is defeatist, and ignores the second-order advantages of Free and Open Source software. That's both a gross misrepresentation of what I wrote, and completely wrong too. It doesn't matter how open your software sources are if you run it on closed hardware. And unless you go back to ancient hardware, it will be closed enough that you do not get what you think you are getting. At the same time, it's also just realistic: if you buy something that is made to cater to a market segment that expects a lot of complex systems to appear to work together in a seamless way, supporting many features requiring many thousands of top-tier engineers to build and maintain for many years, then you, a solo user, are not likely to be able to contain the same amount of knowledge and time to see every detail and component of such a system. So no, the product, the context and biology will not allow you to contain such a system in a single mind. Same goes for what you call user-hostile. If you ask a user which OS is more user-hostile out of a list that contains any Linux-distribution, but also Windows and macOS, it doesn't matter what is technically measures, what matters is that the user will call out that Linux distribution for being user-hostile. Again, it's not about wat is technically possible, what has been done, or what some individual does for themselves, it's about the giant user base that the commercial giants are targeting with their products. As for telemetry: just because alarmists alarm about it all day long doesn't mean it's universally bad. What is bad is going it for user profiling and not telling about it, but if I need to know how my software is performing, I know ahead of time that asking my user base is a waste of time. Either I'll need automated reporting of things like crashes, performance measurements and spindumps, or I can't do much to make the software better. There are too many configurations and situations out there to realistically cover using some imagination and testing. Even on Debian, popcon is super useful, and people who complain about their stuff not making it on disk 1 are usually also the ones not submitting. You get what you pay for, and if you don't supply how you use the product, chances are you're not going to get an improvement you like. That is a choice to make, but removing it completely "because telemetry bad" is kinda dumb.
- MaxBarraclough 4y ago
  > It doesn't matter how open your software sources are if you run it on closed hardware. It matters a great deal. This thread is about the behaviour of Finder, a proprietary software application. More broadly, while it's true that most modern hardware prevents the user from having full control, it's also the case that much is won by using only Free and Open Source software. (Last year the FSF got their servers onto fully Free Software BIOS software, but this took considerable effort as well as the use of hardware from 2012. [0]) > So no, the product, the context and biology will not allow you to contain such a system in a single mind. Agreed, but while a total understanding of the system is one way to achieve confidence that it isn't built to work against your interests, it isn't the only way. > If you ask a user which OS is more user-hostile out of a list that contains any Linux-distribution, but also Windows and macOS, it doesn't matter what is technically measures, what matters is that the user will call out that Linux distribution for being user-hostile. Again this is defeatism. Free Software desktop operating systems have some usability issues, but in my experience they're much more approachable than many people seem to think. This is especially true now that a web browser is enough to enable most of what desktop computers are used for. It's not as if the non-Free operating systems are perfect either. FOSS operating systems aren't in such a good place on mobile as they are on desktop and server, unfortunately. > As for telemetry: just because alarmists alarm about it all day long doesn't mean it's universally bad. What is bad is going it for user profiling and not telling about it Yes, I don't think anyone disagrees. HackerNews folks give Firefox a hard time over its telemetry precisely because the user isn't given a choice. They aren't even told it's going to happen, as I recall. It's particularly grating considering Mozilla's self-promotion as being uniquely pro-user. > You get what you pay for In terms of spying on their users, Debian is far better behaved than Microsoft, despite that Debian's offering is available free of charge and Microsoft's is not. [0] https://www.fsf.org/blogs/sysadmin/closing-in-on-fully-free-bioses-with-the-fsf-tech-team https://www.fsf.org/blogs/sysadmin/closing-in-on-fully-free-...

user3939382 4y ago

> Good luck with that A couple years ago I was able to intercept every connection on an Android with a mirrored switch port and Wireshark+ a root certificate. We shouldn’t be discouraging people from controlling their own machines. Certificate pinning might be trickier, I haven’t played with it yet.

oneplane 4y ago
It's not a case of what is technically possible, it's a case of swimming against the current while you could also choose to swim somewhere else. If you choose something that is not what you want, and you also accept that the world isn't static, but is constantly moving and due to the way it is interconnected you have to keep up at least a little bit, you will be unhappy no matter how hard you try to turn the thing you didn't want into something you did want. It might make much more sense to just get the thing you wanted in the first place, right?
- user3939382 4y ago
  > you could also choose to swim somewhere else Agreed, we have to pick our battles. Personally I pick this one. We should exercise as much control as technically possible over these devices that are intimately linked to every aspect of our lives.