Maybe “forced labor” is a better term for developers who don’t get paid?
“One Day in the life of Ivan Denisovich” applies. (Maybe we’ll get a noodle in our soup to chew on.)
maybe in either case they might prefer to do a paying job that deals with the low hanging fruit you mentioned in order to have the freedom to code whatever they are interested in (not being tied to any corporate overlord’s heel)
Give me an example of what you’re talking about, because it’s still not clicking. I no longer know what you mean by “low hanging fruit,” because “food and electricity” no longer make sense in the context you’re presenting.
Dear Gavudan, “Low hanging fruit” is an idiom about that which is easily obtained or achieved. Food and electricity may be considered such in an industrialized society, not so much in underdeveloped countries where you may be fighting for food.
Interesting. You patronized me and at the same time took rec’s opportunity to defend himself against my suggestion that he was speaking nonsense. It’s a rare thing to see someone backhand two people at the same time.
My response was a straight up definition. It sounded like you needed assistance. Don’t take it personally and yes reC, can be hard to decipher at times.
1 Like
re, sorry for late, however i did some research too besides a lotsa other duties…
tldr: feel free to move on, i was thinking out loud about void and guix, and im sure it became more messy than i could be actually proud about, sorry! 
(actually i had hard times with this topic! 
)
@mms:
the name… whatever… it could be called to be doors or pear… and the average Joe will call gnu+linux to linux, cuz of simplicity, as i would call scheme to lisp, but whatever, and also, as others pointed out, that gnu+linux is just a matter of advertising gnu (cuz its not only gnu, but they wanted their name to stand out) and lets say its fine and more good than not (imho licenses (that is the bedrock of gnu) are a sickness of humanity that renders knowledge incompatible, so public domain everything would be the best, and its still not that easy, thats why unlicense cc0 wt*pl and what else exist, while without the viral gpl, the copyleft stuffs and the like, we would have a much worse state of the industry than what we have now… but stilll… waaaay tooo much issues!!! never mind, it would be too long for now, but i may come back to this point in case of need…)
however their swag, that matters. actually what u r talking about, and what guix is about, those does makes sense, and have strong points, even if i feel like its a scheme-fanboy distro and i dont really like that lispy taste… if they would do the same with lua, then probably we would be on the same page! :V (yea, im a fanboy too X’D ) however i should admit that it actually has really nice features! btw gobo just came into my mind, but i didnt like its sauce, while what i have read about guix made me think about it, and also, when i talk about lua, then, for me, its not just the language, but i already have a mini-empire there…
the bootstrapping problem is an important thing, but theres no simple solution, and probably doesnt take an entire distribution, but something like a compiler that can compile itself and that can compile the rest of the system or other compilers/interpreters, a kernel (if the compiler cant work on baremetal) a disassembler/decompiler/hexeditor? or whatever way to be able to decide that the funds arent infected, and one needs to do it manually, cuz no matter what automatic tools could do the trick, i could return 0 for saying all good (i can even print that with all caps! (not the 0)), while its already running somewhere… maybe the size could be the final factor for saying that whatever system i used to investigate, no component has anything hidden, cuz of no space for it, that moves away from the path of my curious eyes whatever… u need a trusted system 1st, kinda much no matter what… the point is to have a bunch of different tools that can reproduce and analyze each others and at one point any hidden heuristics would reveal itself on something that went off of its happy path… a general purpose ai that lives inside the walls or the thin air? (oh wait, there are compression algorithms, and there are better compression algorithms, we r doomed 
maybe some bulky memory that looks like a digital watch display? X’D ) never mind, maybe a compiler that can reproduce itself could be enough to live with it… uhm, i think maybe others know this better how to solve this in theory and practice
reproducibility is a more interesting stuff, like from a devops perspective, or to validate those binaries that we download blindly, but lets say we stick to a commit with the void repo, and there we are, i think… it can have more factors than that, but i dunno how guix or nix can solve anything beyond picking a well-defined version… if u know, then how? sure, the whole toolchain should be identical too, and on void, this would take a well-defined checkout, and ive seen patching too for openresty/luajit2 in the name of reproducibility, so i guess these reproducible distributions patch everything that is necessary, but still, is there anything else? btw probably i would compile everything myself for resolving the trust issue…
coexistence of the different versions of everything is also nice, but stuffs requiring that have always negative reasons, if something died out, then it had its reasons, if something is an old closed source stuff, then i hope i can live without it, or it can still get an image with its dedicated environment like those various app images or even a plain chroot maybe even by checking out the void repo at the right point… btw not even guix solve the problem when something goes away, everything is pulled in, validated and oh wait, it disappeared by the time … however it exists in the repos, but… thats a binary. (ok, this is rare, not important, and i just thinking out loud! ) for current and colliding stuffs, hmm, thats more interesting, patching, holding, whatever may solve it, but thats rare enough… actually i couldnt figure out so far now how void handles the libs, ive seen a big file where all of them are listed with their name and what provides them, and im afraid thats a real undertake there, while i didnt see issues around it so far now… and yep, backward compatibility is a real and long standing issue on linux, and this is just as sad as is, while hopefully i have never stepped in that… btw i have read something that package maintainers shouldnt care about dependencies, versions whatever, how that works? one sets a minimum/maximum/exact dependency version kinda like on any other system and what i read was only about collisions, that those are not the thing to be afraid of? or is there anything more interesting than this?
rollback is easy, it can be done on a lotsa different ways… ppl on void can keep old packages like on guix, but lvm btrfs or whatever backup tool can solve the same problem.
guix and xbps are similar in the fashion that they are everything that is needed to set up or fix a whole system, and i think guix is also portable on the same level…
homogeneity of the system (shepherd, mcrond, guix, emacs what not) is good, my wet dreams are about the same just with lua and at this point, yep, void … is a bash fanboy distro X’D … :’(
being libre, hmm, i just saw the ymir-linux, that aims to be the libre void, its current state is actually sad! (btw its relatively fresh, and i hope it will find a way back to its upstream instead of fragmenting it…) otherwise its just a matter of not using stuffs i dont use and compiling a libre kernel, or something not really much different from this, and i can have my own templates, as it is intended to be on void, it provides a one size fits all repo and what somebody doesnt like can be modified at will, and it is good at this, and this is intentional. it is a nice, simple and minimalistic fund and things can be modified easily at will, while actually the removal of systemd and the utilization of musl is the lion part of the things i wish for, while i slowly want to replace everything with lua (right, not everything, there are some masterpieces that i like as is and there are things i want to reshape soooo badly, just like those that i could only deoptimize with lua, but hey, lua(jit) can play well with c and asm ) i believe void wont block me on this road with its simplicity and minimalism, and in the meantime i can have a system that i shouldnt care much about and that is very much hackable, and as it is described only by a few repos, i could even make a system that can have all of its sources, and when the zombie apocalypse will hit, and all the infrastructure will die out, i will have a full system that maybe wont turn into untouchable legacy trash on that very day, but it will be a seed for the future to install it onto a half radioactive potato… (never mind.) and actually yes, i will need to learn from guix! but not only, and also, i hope that scheme wont be the rule them all language, so my path should be different, while this is a matter of taste, and we can coexist in peace! but in the meantime, i will keep an eye on guix and nix
@amosbatto
yep, checked out the kernel, and u r right.
We have such a bizarre relationship with Google
this summarizes all X’D btw it all boils down to the point that they had their very own needs that happened to meet with the common good, and to that they have a different sudo than we, the average Joes… ($$$)
low hanging fruit in this context - food and electricity - doesn’t mean easy to obtain. it just means that COMPARED to other things they are easier to obtain. that’s what it means to me at least …
the struggle is real anywhere you are even if you’re in the top 1%. the more you have to lose the worse it gets i imagine but let’s not get into that cuz’ we’re going to derail into even worse off-topic than we already have.
suffice it to say that in ANY society if you want to have a fighting chance then i guess low hanging fruits are the minimum on the priority list no matter the means one employs to secure that … but not everybody thinks the same way and not everybody reaches those goals in the manner you have described … again there’s a different story to everybody so let’s not act like we know everything there is about everybody and define rules of engagement based on our limited perception … hope nobody takes offense here it’s not meant as survival advice 
Oh God. I’ve been brought to that page far too often.
I’m an old Skipper’s Yeoman. I did memos for a living, 30 years. I get that.
1 Like
I have read GNU Guix kernel is Linux-libre which is operational. Also has microkernel: GNU Hurd still in development.
GNU Hurd would be the kernel in GNU.
Blame Linus
GNU Hurd is way behind the Linux-Kernel i’m afraid but in a pinch with the right hw it might be plausible to invest in it further … without heavy contributions to at least one mainstream open-hw there is little chance it will get anywhere near Linux …
1 Like
Linux is the kernel in GNU at the moment. The resulting operating system is not something else but GNU. That’s the only GNU variant that’s ever been popular so far. You have to understand that it was never the plan that the few dozens of GNU project supporters do all the work themselves. Why should they?
Free software does not mean that the developers don’t get payed. The difference between free software and proprietary software is the license. The license ethers grants the user
- the freedom to run the programme for any purpose
- the freedom to study and change the programme
- the freedom to distribute or sell the programme at any price to anyone
- the freedom to distribute or sell the modified programme at any price to anyone
This permissive license is worth more than a proprietary license and should therefore be sold for a higher price.
In the proprietary software business model you sell the user’s data and/or extort to user as he is dependent on you and your software. Therefore the initial costs are usually cheaper for the user than with free software. When you sell free software you will only sell it once to each customer and then they might want you to change it in the coming years or have you to instruct their employees.
Somebody will always be willing to do the tasks which are demanded, if the incentive is high enough. I will clean your toilet for 1 million USD. And I will code and release it to you under the GNU GPL v3 for 1 million USD. And I’m not the only one.
Like Richard Stallman said:
Writing non-free software is not an ethically legitimate activity, so if people who do this run into trouble, that’s good! All businesses based on non-free software ought to fail, and the sooner the better.
4 Likes
How does one make money off of open source software, though?
it’s kinda’ like cleaning the million dollah’ ‘toilet’ … you must be really good at it if it’s that bad …
Since the Internet you cannot charge for an operating system including general apps like a browser and an office suite. But businesses still need a special setup and training and many even need modifications or additional software. I suggest that the companies supplying this service charge more than the proprietary competition, pay the developers well and sell the software under a free license.
Before the Internet rms selled copies of free software which resulted enough revenue so he could live off this business. His general advice is to charge as much as possible for free software.
See also:
GNU project: Selling free software
You mistook me. What I meant was: when the payment is high enough you will find plenty of people who are willing to do the job.
In Germany the farmers claim that they cannot find German workers. Since many decades the farmers are used to get their harvest hands from eastern Europe and now the borders are closed. I’m profoundly convinced that the German population is willing and able to do the job if they are getting paid enough — like all the thousands of years before.
Back to software development. We don’t have to care too much about the developers. As a society we need to care about the software users. Since there will be a need for software development in the foreseeable future, there will be volunteers for the job. Even when this means the developers aren’t allowed to oppress the users. The wage just has to be high enough.
Thats my issue, the wage has to be high enough. You cited a very specific example where it could work (a contract with a company to make a particular piece of software) but then said focus should be on users (which include regular people who want to listen to music and play games and post on social media). You also pointed out that Stallman made enough money off his free software before the internet. I hazard to guess that that’s no longer the case now that the internet is around because open source code can be shared and sold more cheaply or given away for free. Continuing down that path, when that happens developers can’t get paid, so they’ve lost their incentive to work, and now the users are screwed.
Not caring about the developers is wrong for multiple reasons, but that’s one that directly affects your suggestion for how things should be done.
1 Like
Ah, like like serfdom! You get to keep some of the crops in return for protection by the local lord, swear fealty, and get pressed into the army during summer campaign season.
The question is how do you pay software developers to work on FOSS. There are a few companies that have managed to figure it out with an almost 100% FOSS model like Purism, but the vast majority of FOSS code that is contributed come from companies that also produce non-free code. Look at the percentage of the Linux kernel code that comes from each company: Intel (7.65%), IBM/Red Hat (5.23%), Linux Foundation (3.02%), Linaro (2.27%), SUSE (2.01%), AMD (1.90%), TI (1.56%).
Every single one of these companies/organizations is involved in non-free software in some way. Red Hat had the closest to a 100% FOSS model, but it is now part of IBM which sells a lot of non-free code. SUSE has been promoting the sale of Microsoft and SAP software for the last 15 years and SLED has always contained quite a bit of proprietary code. Almost all the major companies that finance the Linux Foundation and Linaro are producers of proprietary code. For example, Linaro is financed by ARM, Comcast, Cypress, Fujitsu, Google, HiSilicon, Marvell, NXP, Qualcomm, IBM/Red Hat, ST, Samsung, Sandia National Labs, and Xilinx.
I worked 10 years in a company that offered both a FOSS “Community Edition” and a proprietary “Enterprise Edition” of its software. The development of the Community Edition was paid for by charging licensing fees for the Enterprise Edition. MySQL was developed under the same model. Maybe you think these companies shouldn’t exist, but I can tell you that we would have less FOSS in the real world without it.
The modern web browser is one of the most complex programs ever created, yet every single one of the web browsers is problematic. Chromium and 89% of FireFox are financed by search royalties that are based on monetizing our personal data. Opera is proprietary. Both Chrome and Safari are proprietary, although their underlying engines are FOSS.
I basically agree with RMS that proprietary software is not good for society and we should depend upon it as little as possible, but I’m also realistic that the development of a lot of FOSS depends on firms which make some of their profits from proprietary software, which is one of the reasons why I said previously that we need both the “open source” and “free software” ideologies if we want our movement to be successful in the real world.
There are many ways to chip away at the problem. Red Hat has shown the most successful business model of how to finance FOSS development, but many Linux companies tried and failed at desktop and mobile Linux, so we haven’t had the kind of paid development that we need for a lot of desktop software, like Inkscape, GIMP, Scribus, Kdenlive, Audacity, VLC, etc. I think we now need companies like Purism and System76 that charge high enough prices for hardware to be able to pay for FOSS development. We also need more companies like Purism and Mozilla that are moving to paid web services to pay for FOSS development.
We do better to focus on building up our own ecosystem and demonstrating that it is a viable alternative with a sustainable model, rather than making needlessly provocative statements where we alienate many of the companies that contribute FOSS code. I don’t disagree with RMS about the end goal, but the question is what is the strategy to get there in the real world?
5 Likes