6 ms·

All of this is trivially and obviously true to power users of GNU/Linux systems. My machine does precisely what I want, when I want it, in the way that I want

All of this is trivially and obviously true to power users of GNU/Linux systems. My machine does precisely what I want, when I want it, in the way that I want it. If it doesn't, within reasonable constraints like physical possibility, I can open a text editor, and make it do so. Sure, that might take me a long time. I can add almost all of this configuration to stuff like git repos; back it up; transfer it between completely different machines; I can have a very high degree of certainty that any laptop or desktop I pick up off the shelf will just work and that within 30mins - 1hour I can have it working _identically_ to the old one. By contrast, right now I am using my phone for USB tethering. There is literally no way for me to make this automatic because it's a black box. I have to enable it every time; if the cable moves a bit and disconnects; etc. Trivial, basic example? Sure. The fact that even this most basic behaviour is completely unfixable proves the point. I'm really looking forward to stuff like the Librem, pinephone, etc. The IBM-compatible PC might well be the greatest contributor to the progress of technology in the last few decades. It's heart-breaking that smartphones are this custom random locked down mess of nonsense. Hell, this article has reminded me I was meaning to set up a donation to the FSF. Off it goes. A small impact but an impact nonetheless. I really appreciate the work you guys are doing for those who can't.

lima 7y ago
Even worse: tethering uses a different APN, which allows providers to rate limit tethered traffic, and without root access, you can't bypass it.
- nolok 7y ago
  For anyone reading this and not knowing this; since your post doesn't make it explicit enough, it uses a different APN because Google specifically made it do so. They could let it be hidden. Of course carrier would then mod it in anyway, but here Google did it for them.
- mfontani 7y ago
  I'm fairly sure tethering can also be trivially detected by inspecting the TTL of packets, as a device sitting "behind" the tethering device will increase the TTL by one. https://android.stackexchange.com/questions/47819/how-can-phone-companies-detect-tethering-incl-wifi-hotspot https://android.stackexchange.com/questions/47819/how-can-ph... seems to have more on the matter.
  - ben7799 7y ago
    This can be spoofed/hidden extremely easily if both devices are open source and you want to change it though. It's not like TTL can't be changed. The tethering device can just rewrite the packets to hide the tethered device.
    - lima 7y ago
      Even better - you only need root on either device. Just set a higher initial TTL on your tethered device, or rewrite the TTL on the phone. But carriers don't use that anymore, too easy to bypass.
  - esotericn 7y ago
    Clever! I hadn't considered that at all. I suspect you mean decrement by one, though. On Linux: `sysctl net.ipv4.ip_default_ttl` (mine was 64) `sysctl net.ipv4.ip_default_ttl=65` Done ;)
ktm5j 7y ago
If you have an android phone then you probably can automate that: https://llamalab.com/automate/ https://llamalab.com/automate/ USB tethering is specifically listed under features, presumably there's some API that developers have access to. Apparently by use of the class WifiManager in android.net.wifi you can programmatically change the state of USB tethering without any kind of special privileges.
- nextos 7y ago
  Or try AOSP, while imperfect its closer to GNU.
  - decoyworker 7y ago
    Nope. AOSP runs on a mobile device which is malware https://www.gnu.org/proprietary/malware-mobiles.html https://www.gnu.org/proprietary/malware-mobiles.html
    - saagarjha 7y ago
      Those complaints have nothing to do with the device that AOSP is running on.
new_realist 7y ago
The fact that you have a beef with USB tethering does not refute the fact that OSS usability is generally shit, and that would likely also apply to smartphones. So it’s a choice between “good” and “poor, but OSS.”
- esotericn 7y ago
  "Good" is in the eye of the beholder, surely. A TV dinner is extremely usable compared to say, cooking from scratch. It has its' benefits. Not for me, though.
- yarrel 7y ago
  Smartphones are based on Free Software. This is part of what makes them so frustrating from a user freedom point of view - they lock some of the most successful Free Software out of the user's reach while sitting in their own hands.
UncleMeat 7y ago
This is true, but is it bad? I work at a major tech company. I have multiple degrees in CS. I've never recompiled anything except maybe in a class ages ago. Among even my most technical peers, I'd bet a lot of money that they percentage of them that have actually modified the software that they are using by editing its source is less than 2%. I'd wager that among the population of people using computers in some way that the percentage is less than 1 in 100,000. And if I asked people who had never done this if they'd like to be able to do it, I'd wager that the huge majority would say "not really". Its fine for power users to want something different than the rest of the world. But it becomes ridiculous to expect people to go out of their way to accommodate these tiny use cases.

lima 7y ago

Even worse: tethering uses a different APN, which allows providers to rate limit tethered traffic, and without root access, you can't bypass it.

nolok 7y ago
For anyone reading this and not knowing this; since your post doesn't make it explicit enough, it uses a different APN because Google specifically made it do so. They could let it be hidden. Of course carrier would then mod it in anyway, but here Google did it for them.
mfontani 7y ago
I'm fairly sure tethering can also be trivially detected by inspecting the TTL of packets, as a device sitting "behind" the tethering device will increase the TTL by one. https://android.stackexchange.com/questions/47819/how-can-phone-companies-detect-tethering-incl-wifi-hotspot https://android.stackexchange.com/questions/47819/how-can-ph... seems to have more on the matter.
- ben7799 7y ago
  This can be spoofed/hidden extremely easily if both devices are open source and you want to change it though. It's not like TTL can't be changed. The tethering device can just rewrite the packets to hide the tethered device.
  - lima 7y ago
    Even better - you only need root on either device. Just set a higher initial TTL on your tethered device, or rewrite the TTL on the phone. But carriers don't use that anymore, too easy to bypass.
- esotericn 7y ago
  Clever! I hadn't considered that at all. I suspect you mean decrement by one, though. On Linux: `sysctl net.ipv4.ip_default_ttl` (mine was 64) `sysctl net.ipv4.ip_default_ttl=65` Done ;)

ktm5j 7y ago

If you have an android phone then you probably can automate that: https://llamalab.com/automate/ https://llamalab.com/automate/ USB tethering is specifically listed under features, presumably there's some API that developers have access to. Apparently by use of the class WifiManager in android.net.wifi you can programmatically change the state of USB tethering without any kind of special privileges.

nextos 7y ago
Or try AOSP, while imperfect its closer to GNU.
- decoyworker 7y ago
  Nope. AOSP runs on a mobile device which is malware https://www.gnu.org/proprietary/malware-mobiles.html https://www.gnu.org/proprietary/malware-mobiles.html
  - saagarjha 7y ago
    Those complaints have nothing to do with the device that AOSP is running on.

new_realist 7y ago

The fact that you have a beef with USB tethering does not refute the fact that OSS usability is generally shit, and that would likely also apply to smartphones. So it’s a choice between “good” and “poor, but OSS.”

esotericn 7y ago
"Good" is in the eye of the beholder, surely. A TV dinner is extremely usable compared to say, cooking from scratch. It has its' benefits. Not for me, though.
yarrel 7y ago
Smartphones are based on Free Software. This is part of what makes them so frustrating from a user freedom point of view - they lock some of the most successful Free Software out of the user's reach while sitting in their own hands.

UncleMeat 7y ago

This is true, but is it bad? I work at a major tech company. I have multiple degrees in CS. I've never recompiled anything except maybe in a class ages ago. Among even my most technical peers, I'd bet a lot of money that they percentage of them that have actually modified the software that they are using by editing its source is less than 2%. I'd wager that among the population of people using computers in some way that the percentage is less than 1 in 100,000. And if I asked people who had never done this if they'd like to be able to do it, I'd wager that the huge majority would say "not really". Its fine for power users to want something different than the rest of the world. But it becomes ridiculous to expect people to go out of their way to accommodate these tiny use cases.