4 ms·
This may not be a direct comment rms's computing habits, but it's related and as good a place as any bring it up. I've been trying to reconcile two views on fre
by chestervonwinch 11y ago
This may not be a direct comment rms's computing habits, but it's related and as good a place as any bring it up. I've been trying to reconcile two views on free software. I like to think about it in terms of car ownership because its more tangible to me:
As a car owner, I should have the freedom to open my car, inspect its inner workings, change/fix things as desired, and share my knowledge others. The ability to openly inspect the inner workings also allows me to assure that the car is only performing operations I purchased it to perform.
However, as a car developer, I cannot possibly hope to make money by giving the car away for free. There have been significant hours invested into the design of the car. Sure, there's money to made in maintenance, but realistically most people don't care about routine maintenance. They'd rather buy the car and ride it 'til the wheels fall off - maybe an oil change here and there.
Are these two views fundamentally incompatible? Does anyone pay an upfront cost for free software in recognition of the time spent building/designing the product as they might with proprietary?
- bane 11y agoOne of the arguments of the movement is that by participating in it, by giving away your labor to all mankind, others can benefit and this works vis-a-versa, others give away their labor so you can benefit as well. The problem of course is that this also says that labor is worthless and nobody gets to eat and find shelter unless they have some significant patronage. So sure I can build my Sprocket-Cog Software more easily because many of the software components that go into it are available to me for free (beer) so long as I free the entire thing at the end (freedom). The follow-on argument is that you can always try to build a services model around Sprocket-Cog Software and live from that. But there's a subtle implication there that building turnkey software is not an option if you want to live, but making broken, user-hostile, hard-to-use software is the only morally "good" option. (there's another argument that can be dismissed that you aren't prohibited from selling your software commercially under the right license, but that's a stupid argument, nobody will buy software from you if you just make it available for free (beer) anyways, so your only valid option (and the only one that has so far sorta "worked" in the marketplace is to provide services) RMS of course dismisses all of this with a bunch of handwaivyness, like he does issues with hardware, food, etc. because he simply doesn't have to worry about them (because he has money) or because he can't find a path he can personally follow. It's hard to take claims about morality seriously when they only apply to software and not anything else that's far more embedded in life.
- vinceguidry 11y agoIt is not because he has money that he can find a personal path to follow, he really doesn't. His speaking fee is modest, his position at MIT is unpaid. He lives like a student, if you've ever read that long document he asks others to read if they book him for a speaking gig, can't find a link to it at the moment, it's mostly about avoiding unnecessary expense and distraction.
- bane 11y agoYou do realize that he is actually in possession of a reasonably large amount of money and can live independently (if frugally) without payment for the rest of his life.
- c22 11y agoBut he wasn't in possession of it when he initially developed his views, was he?
- vinceguidry 11y agoAnd it's not like his lifestyle changed after he won those prizes, either.
- hibikir 11y agoRMS doesn't say that work is worthless, or that you have to give your labor away to all mankind. It's purely an issue of what people can do with the fruits of said labor. I've been programming professionally for 15 years, and I've gotten well paid for it. But I never worked for a company that sold its software without handing out the code right with it. Many times it has not met RMS' standards, because some of the code used proprietary libraries, but using said libraries was always the customer choice. Most of the time, the customer that uses the code keeps it to themselves, just like they keep the actual executables they compile to themselves. A few times, however, my code was just released with an open license. Either way, I'd have still gotten paid, every single time, if copyrights or patents did not cover software at all. Remember that the original complaints that made this movement happen had little to do with software being for pay, but with the chains that come along with not allowing modification: Printer driver has a bug? Tough. Game doesn't work in a newer computer? Buy a remake. Bought a new computer because the old one broke? Call MS and see if they will let you transfer the OS from machine to machine. It's issues like that free software wanted to change. Making it hard to, say, sell videogames in that kind of model, is not the real objective, but collateral damage.
- Scarbutt 11y agoHardware is a whole new angle, RMS has talk about it too I think.
- throwawayaway 11y agohttps://www.gnu.org/philosophy/selling.html https://www.gnu.org/philosophy/selling.html your misconception of the position is addressed in the above essay. RMS: Many people believe that the spirit of the GNU Project is that you should not charge money for distributing copies of software, or that you should charge as little as possible—just enough to cover the cost. This is a misunderstanding.
- chestervonwinch 11y agoBut isn't the distribution charge nullified by the first person who distributes the software free of charge (and they are free to do so because the software is libre)?
- throwawayaway 11y agoNot far from the top of the article: Since free software is not a matter of price, a low price doesn't make the software free, or even closer to free. So if you are redistributing copies of free software, you might as well charge a substantial fee and make some money. Redistributing free software is a good and legitimate activity; if you do it, you might as well make a profit from it. also: making closed and open versions is OK by rms. https://www.gnu.org/philosophy/selling-exceptions.html https://www.gnu.org/philosophy/selling-exceptions.html
- discordianfish 11y agoI think the point is that once you sold a copy of your free software once, the buyer may give it away for free. Realistically I'd say it's not really possible to make money by selling free software. That's okay if you can make money off from services and integration but selling isn't really possible and some free software advocates tend to ignore that fact. Especially if it's about other digital content like music. I don't think that Wu-Tang Clan's model of just selling one copy will very be sustainable.
- throwawayaway 11y ago
- userbinator 11y agoI think the biggest difference is that cars are physical objects which require nontrivial resources to produce, even just to copy. Even if you had all the knowledge about a car to the point of being able to make a copy, you still have to acquire the raw materials. Software is different; copying and distribution essentially costs only as much as the power it takes to move bits around. As long as physical materials are not free, I don't think it contradicts supporting free software - even paying developers for their effort - and also paying for a car. There's no incompatibility here. ... but today, cars do contain software, which makes the situation a lot more complex...
- greggman 11y agoIf you follow physical materials back to their source they are free. Dirt, gravel, minerals, oil, water, coal, rubber, etc, they are freely sitting around waiting for labor to collect them, labor to process them, labor to deliver them. When you buy a car for $30k you aren't paying for $30k of materials you're paying for $30k of labor and/or perceived value. Obviously software can be duplicated for free but it often takes significant labor. Often more than making a few thousand cars.
- boomlinde 11y agoDirt, gravel, minerals, oil, water, coal, rubber etc. aren't free. Land ownership is a thing and natural resources are usually controlled by someone. Simply controlling the access to natural resources can make you rich without lifting a finger. More fundamentally, the production of goods comes at a price that can't necessarily be easily expressed in terms of money. Extracting and using natural resources often damages the environment and more obviously deprives other people of the use of those resources. You aren't paying $30k just for materials, but you aren't paying $30k for just labor and perceived value either. Resources used in the production of cars are sometimes naturally scarce, and if you own such resources, people will pay you to extract them.
- chestervonwinch 11y ago> I think the biggest difference is that cars are physical objects which require nontrivial resources to produce I disagree. Software, too, requires nontrivial resources to produce. I guess what I'm arguing is that inherent value for both is in production / design. When you buy a vehicle, you're paying for the design, the function, etc..., not the physical materials. However, the object being purchased in the case of a vehicle must manifest itself through physical materials, and so the physical materials are the barrier for reproduction / distribution. This barrier forces payment towards the production / design. In the software case, the barrier is artificially introduced by closing the source. This seems to be the way many force payment to the production / design of software. But naturally this cripples the users freedom.
- quadrangle 11y agoRMS' view is simple: Work is valuable, people should be paid, but putting proprietary software into the world is harmful, so you should be shamed for that, not paid for it. Thus, he will readily acknowledge that making a living writing freely-licensed software may be difficult, but since it is the only ethical way to write software, we just have to figure out the best ways we can to do it. An extreme absurd version of the logic (NOT to say the moral implications are the same, just to understand the logic!!) is this: it may be easier to get rich owning slaves, but you should not own slaves. I don't care how hard life is for you or how hard you have to work or how impossible your business would be without slaves. No slavery. http://mimiandeunice.com/2010/08/25/for-the-children/ http://mimiandeunice.com/2010/08/25/for-the-children/ The fact that an unethical business model can be successful doesn't matter. We should reject it and figure out whatever else we have to do to get by.
- dllthomas 11y agoAnd of course, hopefully Snowdrift will get us to a place where it's easier to be paid developing free software! :)