5 ms·

RMS did a really poor job of explaining why closed source firmware is problematic to free software, I agree, and I'll explain better below. I also have some cri

by _ktx2 3y ago

RMS did a really poor job of explaining why closed source firmware is problematic to free software, I agree, and I'll explain better below. I also have some critique of activist language. First, activist language is pretty hard for any human to listen to. Anyone that's been tackling the "long fight" of what they view as oppression has this map in their head of things that add up to their main point in life and cannot help but continually extrapolate it for people. On some topics, people eat this kind of language up. On others, it's rather dull, especially if the person using the language can't quickly connect relatable points for people. RMS was good at long-form broad thinking that was dealing with the issues at the time. He was pretty fuck all when it came to really explaining them in a digestible, conversation style manner. As for why proprietary headphone firmware hurts free software: take some Bose noise cancelling headphones and try to connect them to something like a mainstream Linux distribution like PopOS and use them on Zoom. They'll work, but only in a low quality mode, or the audio receiver will work but not the microphone. Now do the same with Airpods, or use Airpods on an Android phone. Airpods will work and you'll get high fidelity audio, but they're constantly exhibiting weird disconnection behaviors. On Linux you'll get incredibly low fidelity audio. Proprietary drivers are at the source of all of this. RMS came from a day where the pioneers of Linux and alternative operating systems weren't building them for servers. They were building them for users because Windows was expensive and the machines to run it on were more expensive. Oppression to someone like him, and many others, were linked to this and many other circumstances at the time. The idea that the optimizations of computing were available only to a few was utterly disasterous. To connect the two ideas, product x which is proprietary works best on product y which is also proprietary. It's a system that reinforces itself and excludes or marginalizes the rest, which eroded the original mission of building optimized software that is collectively free enough for anyone in the world to use. Somewhat ironically, a lot of people in this thread are complaining about diversity when really that's what most of these initiatives were born in.

MontyCarloHall 3y ago
>take some Bose noise cancelling headphones and try to connect them to something like a mainstream Linux distribution like PopOS and use them on Zoom. They'll work, but only in a low quality mode, or the audio receiver will work but not the microphone. My Bose QC45 headphones work identically on all my hardware (Mac, Linux, and occasionally Windows). They can use both standard Bluetooth or a standard 3.5mm TRS connector, neither of which requires any special drivers. Although I've never had any Linux-specific issues, I would guess that if you're experiencing them it's due more to Linux's notoriously poor audio stack than the actual hardware, which conforms to industry standards. >Now do the same with Airpods, or use Airpods on an Android phone. Airpods will work and you'll get high fidelity audio, but they're constantly exhibiting weird disconnection behaviors. On Linux you'll get incredibly low fidelity audio. I've also done this to a limited extent (AirPods on Android), and did not notice anything unusual, aside from a few missing Apple-specific features (e.g. auto-device switching, which is not part of the Bluetooth protocol). >To connect the two ideas, product x which is proprietary works best on product y which is also proprietary. It's a system that reinforces itself and excludes or marginalizes the rest It is generally in the best interest of companies to produce goods usable by a maximal marketshare; willingly excluding customers and their money is usually not a good business strategy. This is why companies voluntarily conform to standards. The main exception to this is when standards inhibit the quality of the good. In that case, companies are willing to trade a smaller marketshare for a better user experience in that smaller market (e.g. AirPods' Apple-only features). As an extreme example, consider cars. Parts are proprietary to a given model because each car is optimized for its specific market niche (size, appearance, performance, etc.), which requires all parts for a given model to be closely integrated. The engine of a Honda is not compatible with a Nissan, because engines and cars must be so tightly integrated that it would be an impossibly bad experience otherwise.
- kdmccormick 3y ago
  How would you feel if Google started building its websites to function significantly better in Chrome than in any other browser? (I guess is already happening to some extent... but not as egregiously as it could.) Because that's the same idea. We have Web standards, which all websites should use, just like we have wireless standards which all wireless hardware should use. Efficient implementation of the standard should be all that's required. Why do we do this? Because it's better for individual large companies? No, it's not. We do it because following interoperable standards is better for everyone. Having tightly-integrated Apple products is nice, but having tightly-integrated everything is better for consumers, innovators, and small new companies. Car analogies are not apt here. Cars and engines are physical things. Software is not.
  - MontyCarloHall 3y ago
    >How would you feel if Google started building its websites to function significantly better in Chrome than in any other browser? (I guess is already happening to some extent... but not as egregiously as it could.) Because that's the same idea. We have Web standards, which all websites should use If the proprietary Chrome-only extensions to web standards provided a dramatically better user experience, I would have no issues with this, as long as users are 100% free to choose whichever browser (and online services) they please, and as long as other companies were free to independently implement/reverse engineer these extensions. I would only have a problem if Google abused a monopoly position to force users to use its products, à la Microsoft and Internet Explorer in the late 90s. >Why do we do this? Because it's better for individual large companies? No, it's not. It's almost always better for companies to maximize their marketshare, unless maximizing marketshare would compromise the quality of the product. The iPod was originally Mac only, and was a rather niche toy until Apple opened compatibility, at which point it became a global sensation. This was likely due to the fact that in 2001, when the iPod debuted, all Apple computers had a port fast enough to transfer gigabytes of music (FireWire, 400 Mb/s), while most non-Apple machines still had USB 1.0 (12 Mb/s), so a high capacity MP3 player simply would not have been a good user experience. It wasn't until USB 2.0 (480 Mb/s) became common a year or so later that a large MP3 player became a good experience on non-Apple machines, and Apple rationally decided to maximize its marketshare. >Car analogies are not apt here. Cars and engines are physical things. Software is not. You are correct that analogies to physical goods are imperfect; I was merely using them to point out how the experience of tight product integration can sometimes be more valuable to the user than interoperability. It all depends on the use case.
noirscape 3y ago
There's a place for talking about the viability of FOSS firmware for headphones and similar devices, but I'm sure we can both agree that on an airplane ride (omitted from parent response, but that's when it happened) is not the right time. The problem imo isn't the argument, it's where and when it was presented. Most people on airplanes are, at least in my experience: half-tired, somewhat sleepy and vaguely annoyed due to all the annoyances and frustrations with getting through customs. Few people want to be subjected to angry rants at that point. RMS should at least have recognized that[0] this is the sort of rant that is best put on his website after the ride rather than at the person who just was wearing some headphones[1]. With RMS having placed himself as the de facto speaker for the FSF, this is an extremely bad look for them once this story got out and by extension Free Software as a whole[2] (and why, imo, he should've been moved to some sort of "guru" position long ago instead of being the public speaker). [0]: Yes, I realize he's probably autistic. The counter I have to that is that autism isn't an excuse. It's an explanation and RMS has habitually rejected any attempts to understand that his behavior may be regarded as extremely rude according to everyone who has worked with him, which is why I still consider this to be a detriment to him as a person. Almost every autistic person I know is miles more understanding than RMS is in that their behavior might sometimes come across in ways they didn't intent to. RMS is full head-in-the-sand-I'm-always-right on this sort of thing whenever he's confronted. [1]: Which probably would have made it a more useful argument too, since more people would've been aware of it as opposed to severely irritating a random individual on an airplane. [2]: Not to dismiss the merits of FOSS on that, of course. There's plenty of other organizations that I would consider more useful advocates such as the FSFe and the SFC.
- jfengel 3y ago
  An autistic person may have difficulty reading social cues, but it's hard to imagine a clearer social cue than "this person is wearing a device for the express purpose of not listening to things".
- CaptainFever 3y ago
  > Yes, I realize he's probably autistic. The counter I have to that is that autism isn't an excuse. It's an explanation It’s a disability. > Almost every autistic person I know is miles more understanding than RMS is in that their behavior might sometimes come across in ways they didn't intent to. It’s also a spectrum. Something that might be easy for an autistic person might be hard for another. That said… > and RMS has habitually rejected any attempts to understand that his behavior may be regarded as extremely rude according to everyone who has worked with him, which is why I still consider this to be a detriment to him as a person. If this is true, then I can see your point.
kdmccormick 3y ago
> Anyone that's been tackling the "long fight" of what they view as oppression has this map in their head of things that add up to their main point in life and cannot help but continually extrapolate it for people. Well said. This is a good point for any activist to keep in mind.
mablopoule 3y ago
Good insight on "activist language", it's something I've been sensible (and irritated) more and more the last few years, but had trouble putting words on it.

by _ktx2 3y ago

MontyCarloHall 3y ago

>take some Bose noise cancelling headphones and try to connect them to something like a mainstream Linux distribution like PopOS and use them on Zoom. They'll work, but only in a low quality mode, or the audio receiver will work but not the microphone. My Bose QC45 headphones work identically on all my hardware (Mac, Linux, and occasionally Windows). They can use both standard Bluetooth or a standard 3.5mm TRS connector, neither of which requires any special drivers. Although I've never had any Linux-specific issues, I would guess that if you're experiencing them it's due more to Linux's notoriously poor audio stack than the actual hardware, which conforms to industry standards. >Now do the same with Airpods, or use Airpods on an Android phone. Airpods will work and you'll get high fidelity audio, but they're constantly exhibiting weird disconnection behaviors. On Linux you'll get incredibly low fidelity audio. I've also done this to a limited extent (AirPods on Android), and did not notice anything unusual, aside from a few missing Apple-specific features (e.g. auto-device switching, which is not part of the Bluetooth protocol). >To connect the two ideas, product x which is proprietary works best on product y which is also proprietary. It's a system that reinforces itself and excludes or marginalizes the rest It is generally in the best interest of companies to produce goods usable by a maximal marketshare; willingly excluding customers and their money is usually not a good business strategy. This is why companies voluntarily conform to standards. The main exception to this is when standards inhibit the quality of the good. In that case, companies are willing to trade a smaller marketshare for a better user experience in that smaller market (e.g. AirPods' Apple-only features). As an extreme example, consider cars. Parts are proprietary to a given model because each car is optimized for its specific market niche (size, appearance, performance, etc.), which requires all parts for a given model to be closely integrated. The engine of a Honda is not compatible with a Nissan, because engines and cars must be so tightly integrated that it would be an impossibly bad experience otherwise.

kdmccormick 3y ago
How would you feel if Google started building its websites to function significantly better in Chrome than in any other browser? (I guess is already happening to some extent... but not as egregiously as it could.) Because that's the same idea. We have Web standards, which all websites should use, just like we have wireless standards which all wireless hardware should use. Efficient implementation of the standard should be all that's required. Why do we do this? Because it's better for individual large companies? No, it's not. We do it because following interoperable standards is better for everyone. Having tightly-integrated Apple products is nice, but having tightly-integrated everything is better for consumers, innovators, and small new companies. Car analogies are not apt here. Cars and engines are physical things. Software is not.
- MontyCarloHall 3y ago
  >How would you feel if Google started building its websites to function significantly better in Chrome than in any other browser? (I guess is already happening to some extent... but not as egregiously as it could.) Because that's the same idea. We have Web standards, which all websites should use If the proprietary Chrome-only extensions to web standards provided a dramatically better user experience, I would have no issues with this, as long as users are 100% free to choose whichever browser (and online services) they please, and as long as other companies were free to independently implement/reverse engineer these extensions. I would only have a problem if Google abused a monopoly position to force users to use its products, à la Microsoft and Internet Explorer in the late 90s. >Why do we do this? Because it's better for individual large companies? No, it's not. It's almost always better for companies to maximize their marketshare, unless maximizing marketshare would compromise the quality of the product. The iPod was originally Mac only, and was a rather niche toy until Apple opened compatibility, at which point it became a global sensation. This was likely due to the fact that in 2001, when the iPod debuted, all Apple computers had a port fast enough to transfer gigabytes of music (FireWire, 400 Mb/s), while most non-Apple machines still had USB 1.0 (12 Mb/s), so a high capacity MP3 player simply would not have been a good user experience. It wasn't until USB 2.0 (480 Mb/s) became common a year or so later that a large MP3 player became a good experience on non-Apple machines, and Apple rationally decided to maximize its marketshare. >Car analogies are not apt here. Cars and engines are physical things. Software is not. You are correct that analogies to physical goods are imperfect; I was merely using them to point out how the experience of tight product integration can sometimes be more valuable to the user than interoperability. It all depends on the use case.

noirscape 3y ago

There's a place for talking about the viability of FOSS firmware for headphones and similar devices, but I'm sure we can both agree that on an airplane ride (omitted from parent response, but that's when it happened) is not the right time. The problem imo isn't the argument, it's where and when it was presented. Most people on airplanes are, at least in my experience: half-tired, somewhat sleepy and vaguely annoyed due to all the annoyances and frustrations with getting through customs. Few people want to be subjected to angry rants at that point. RMS should at least have recognized that[0] this is the sort of rant that is best put on his website after the ride rather than at the person who just was wearing some headphones[1]. With RMS having placed himself as the de facto speaker for the FSF, this is an extremely bad look for them once this story got out and by extension Free Software as a whole[2] (and why, imo, he should've been moved to some sort of "guru" position long ago instead of being the public speaker). [0]: Yes, I realize he's probably autistic. The counter I have to that is that autism isn't an excuse. It's an explanation and RMS has habitually rejected any attempts to understand that his behavior may be regarded as extremely rude according to everyone who has worked with him, which is why I still consider this to be a detriment to him as a person. Almost every autistic person I know is miles more understanding than RMS is in that their behavior might sometimes come across in ways they didn't intent to. RMS is full head-in-the-sand-I'm-always-right on this sort of thing whenever he's confronted. [1]: Which probably would have made it a more useful argument too, since more people would've been aware of it as opposed to severely irritating a random individual on an airplane. [2]: Not to dismiss the merits of FOSS on that, of course. There's plenty of other organizations that I would consider more useful advocates such as the FSFe and the SFC.

jfengel 3y ago
An autistic person may have difficulty reading social cues, but it's hard to imagine a clearer social cue than "this person is wearing a device for the express purpose of not listening to things".
CaptainFever 3y ago
> Yes, I realize he's probably autistic. The counter I have to that is that autism isn't an excuse. It's an explanation It’s a disability. > Almost every autistic person I know is miles more understanding than RMS is in that their behavior might sometimes come across in ways they didn't intent to. It’s also a spectrum. Something that might be easy for an autistic person might be hard for another. That said… > and RMS has habitually rejected any attempts to understand that his behavior may be regarded as extremely rude according to everyone who has worked with him, which is why I still consider this to be a detriment to him as a person. If this is true, then I can see your point.

kdmccormick 3y ago

> Anyone that's been tackling the "long fight" of what they view as oppression has this map in their head of things that add up to their main point in life and cannot help but continually extrapolate it for people. Well said. This is a good point for any activist to keep in mind.

mablopoule 3y ago

Good insight on "activist language", it's something I've been sensible (and irritated) more and more the last few years, but had trouble putting words on it.