Working on FOSS doesn't mean we work for free, right? Right?!?
-
@error said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@dcon said in Working on FOSS doesn't mean we work for free, right? Right?!?:
DAMN YOU PEOPLE, CLICK THE 'DELETE ON MERGE' OPTION IN YOUR PR!!!!
This disabled one?
Your git server admin is a moron.
There are several ways to set up the workflow. IMNSHO some of them are dumb, but some people insist on them:
- Keeping merged branches around is dumb, but some people think they need them, not realizing that they can dig the changes relevant to a feature up by the merge commit, the merge request itself or the link in the task.
- Squash merges are dumb, because now if you dare to merge one feature branch in another, e.g. because the second feature needs some refactoring done in the first one, you'll end up in merge hell as the final history no longer contains the commits from the feature branches. But some people think there is some benefit to the ‘simpler’ history.
-
@Arantor said in Working on FOSS doesn't mean we work for free, right? Right?!?:
Guarantee that installing the @8.2 formula will update every fucking thing (not just the relevant dependencies) and break my dev environment just like the time I installed 8.1…
That's why VSCode introduced the devcontainers – when you have multiple projects and they need different dependencies, you can simply have a completely separate environment in a container for each. Though … from mentioning homebrew I suppose you are on MacOS and containers have Linux inside and for some reason they tend to be quite a bit slower on MacOS.
-
@dcon said in Working on FOSS doesn't mean we work for free, right? Right?!?:
because who the fuck knows how networking in VIrtualBox works on a ubuntu host
Weird, it has always just worked for me. There is the userland masquerade option that does not require anything from the OS, not even elevated privileges, so that should be always available everywhere.
-
@Gern_Blaanston said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@topspin said in Working on FOSS doesn't mean we work for free, right? Right?!?:
Yeah, Explorer is really one of the better parts of Windows.
Prior to Windows 11, yes. Beginning with Win11 Microsoft is trying their hardest to make Explorer useless and retarded. They haven't completely succeeded ... yet ... but I'm sure they won't give up.
Finder has a few very nice features
Several months ago I bought an iPad which I hoped would be a replacement for my 10 year old laptop that finally died.
But I was never able to figure out how to do the things that I did in Windows using Explorer, and the people at the store couldn't seem to even understand what I was trying to do. Apparently the idea of connecting to a shared folder on my home network is some sort of bizarre, unknown thing in the Apple world.
Command k and then spend 15 minutes wondering why it keeps mangling the url until you give up and scp the files to a linux vm to copy them. I wouldn’t bother to be honest.
-
@Bulb said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@dcon said in Working on FOSS doesn't mean we work for free, right? Right?!?:
because who the fuck knows how networking in VIrtualBox works on a ubuntu host
Weird, it has always just worked for me. There is the userland masquerade option that does not require anything from the OS, not even elevated privileges, so that should be always available everywhere.
I'm pretty sure it's because of other weird things that are setup on our machines. (Because I've had no issues with VMs on my personal machines)
-
@Gern_Blaanston 16 years later, Jobs’ claim that iPhones (or iPads in your case) run OS X is still a lie. “Finder has some nice features” absolutely doesn’t apply to the extremely basic files app on iOS / IPadOS.
-
@dcon said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@Bulb said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@dcon said in Working on FOSS doesn't mean we work for free, right? Right?!?:
because who the fuck knows how networking in VIrtualBox works on a ubuntu host
Weird, it has always just worked for me. There is the userland masquerade option that does not require anything from the OS, not even elevated privileges, so that should be always available everywhere.
I'm pretty sure it's because of other weird things that are setup on our machines. (Because I've had no issues with VMs on my personal machines)
Whew. Fixed it. Had to change the complete contents of the /etc/netplan/... file. (The OS image was designed for an appliance on a defined network - finally after someone suggested removing that file (didn't work) I had the brillant idea of replacing it with the contents of the host's file. Would have been nice if devops could have suggested that about 6 hours ago...
-
@topspin said in Working on FOSS doesn't mean we work for free, right? Right?!?:
16 years later, Jobs’ claim that iPhones (or iPads in your case) run OS X is still a lie.
He got it backwards. What it meant is that Apple computers run iOS
-
@dcon said in Working on FOSS doesn't mean we work for free, right? Right?!?:
VIrtualBox
Btw, what's the state of the art with such virtualization solutions? Haven't played with them for ... "a few" years already. Previously, I used VMWare Player, and that was a good experience, networking always used to work, USB devices, audio, etc., just worked fine, while VirtualBox was a PITA for such requirements back then.
-
@Arantor said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@LaoC I know, I was more getting at hoe this is very much not a “mastered” situation on Windows. It’s much closer to a shitshow.
Windows is a master of shitshows.
-
@BernieTheBernie said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@dcon said in Working on FOSS doesn't mean we work for free, right? Right?!?:
VIrtualBox
Btw, what's the state of the art with such virtualization solutions? Haven't played with them for ... "a few" years already. Previously, I used VMWare Player, and that was a good experience, networking always used to work, USB devices, audio, etc., just worked fine, while VirtualBox was a PITA for such requirements back then.
I stopped using VirtualBox on my Win10 work machine a few years ago due to crashes. Been using VM Workstation since and haven't had any problems.
-
@boomzilla said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@BernieTheBernie said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@dcon said in Working on FOSS doesn't mean we work for free, right? Right?!?:
VIrtualBox
Btw, what's the state of the art with such virtualization solutions? Haven't played with them for ... "a few" years already. Previously, I used VMWare Player, and that was a good experience, networking always used to work, USB devices, audio, etc., just worked fine, while VirtualBox was a PITA for such requirements back then.
I stopped using VirtualBox on my Win10 work machine a few years ago due to crashes. Been using VM Workstation since and haven't had any problems.
I've had no issues with it on my Win10 machine. I don't use it heavily there tho. I usually use HyperV, but the machine with VB is a -Home variant, so it can't. I used to use VMware (hell, I use to work there), but stopped because
.
-
@Gern_Blaanston said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@loopback0 said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@Gern_Blaanston said in Working on FOSS doesn't mean we work for free, right? Right?!?:
Apparently the idea of connecting to a shared folder on my home network is some sort of bizarre, unknown thing in the Apple world.
It's in the Files app. Tap the icon top right, Connect to Server.
And yet, people who sell iPads don't seem to know this.
Can''t expect a genius to know these things.
-
@BernieTheBernie said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@dcon said in Working on FOSS doesn't mean we work for free, right? Right?!?:
VIrtualBox
Btw, what's the state of the art with such virtualization solutions? Haven't played with them for ... "a few" years already. Previously, I used VMWare Player, and that was a good experience, networking always used to work, USB devices, audio, etc., just worked fine, while VirtualBox was a PITA for such requirements back then.
I get paid to support people using Hyper-V as our primary recommended solution. For very rare cases where they have some
requirement (usually networking or audio) that isn't worth the lift to get working via Hyper-V, we license & use VMWare Workstation.
-
@izzion said in Working on FOSS doesn't mean we work for free, right? Right?!?:
audio
Yeah, I don't care about that in my VMs! (I just need some controlled spaces for testing - REVERT ALL! And my build machine with cert signing.)
-
@dcon said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@izzion said in Working on FOSS doesn't mean we work for free, right? Right?!?:
audio
Yeah, I don't care about that in my VMs! (I just need some controlled spaces for testing - REVERT ALL! And my build machine with cert signing.)
I don't mean to imply that getting any audio in a Hyper-V VM working is a pain in the nuts, that works pretty well out of the box for Windows guests (Linux guests are more of a bodge but Linux guests in Hyper-V is kinda
anyway). More we've had some problems with video conferencing with Teams (yeah, yeah, 
) or Zoom or similar if they try to run that from within the Hyper-V guest.But since most of our guest usages are along the same line - dev environment, join to a client's network without effecting the management policies on the hardware - they're usually doing the more driver intensive stuff directly on the host machine anyway.
-
-
@topspin said in Working on FOSS doesn't mean we work for free, right? Right?!?:
FOSS means “free open source software”.
For the end user, it is both “gratis” in that it doesn’t cost anything, as well as “libre”, in that if your printer is being retarded and full of bugs, you (or someone you hire) can fix it. (Literally what caused RMS to create the FOSS movement)RMS is an experienced and knowledgeable programmer. Most people are not. And that is the biggest problem with FOSS software.
It is nice to have the source code available so that you can fix bugs or make changes that you want -- if you possess the necessary skill. But, if you have to pay someone to do it, because you do not have the expertise to do it yourself, then the software is not really free (as in "no cost").
People often use the phrase "free as in beer" meaning that there is no cost. But a better one would be "free as in puppy". A free puppy comes with a certain amount of cost.
-
@Gern_Blaanston said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@topspin said in Working on FOSS doesn't mean we work for free, right? Right?!?:
FOSS means “free open source software”.
For the end user, it is both “gratis” in that it doesn’t cost anything, as well as “libre”, in that if your printer is being retarded and full of bugs, you (or someone you hire) can fix it. (Literally what caused RMS to create the FOSS movement)RMS is an experienced and knowledgeable programmer. Most people are not. And that is the biggest problem with FOSS software.
It is nice to have the source code available so that you can fix bugs or make changes that you want -- if you possess the necessary skill. But, if you have to pay someone to do it, because you do not have the expertise to do it yourself, then the software is not really free (as in "no cost").
People often use the phrase "free as in beer" meaning that there is no cost. But a better one would be "free as in puppy". A free puppy comes with a certain amount of cost.
The "free" in RMS's concept of "free software" is "free as in liberty", not anything about costs. That's why the GPL imposes so many restrictions on what you can do with GPL-licensed software. (Yes, that's sarcasm, why do you ask?)
-
@Steve_The_Cynic In RMS’s mind this is a valid notion, and I’m not sure he’s wrong about it.
Consider the workflow: you are a business reliant on a piece of software that is copyleft, the vendor goes out of business but the software has a critical bug. (Maybe that’s why the vendor went out of business.) You could, at this point, move to another piece of software or pay someone to fix that bug.
And there are cases where this is a very different equation (putting aside intellectual property protectionism for a minute), e.g. hardware drivers for very expensive equipment that won’t be updated past XP - if that control machine fails, you’re going to have to track down older parts that may not be as readily available, but if the driver were open source (maybe after the fact), other options for updating it exist.
Sure, that also gets you into a whole realm of hypotheticals but in RMS’s mind at least, copyleft creates alternatives that closed source explicitly forbids and that is his argument.
There’s also the socialist in me that loves the idea of open source and would happily throw both money and time at it in less trivial ways than I do today if there were a way to make it more properly sustainable than, often, relying on volunteer labour.
-
@Arantor said in Working on FOSS doesn't mean we work for free, right? Right?!?:
There’s also the socialist in me that loves the idea of open source and would happily throw both money and time at it in less trivial ways than I do today if there were a way to make it more properly sustainable than, often, relying on volunteer labour.
Some people are paid to work on these sorts of things. For them it is not a volunteer effort matter. The organisations paying are getting something for their money too: the changes they want on the timescale they are willing to pay for. Because COTS doesn't always cut it.
-
@dkf said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@Arantor said in Working on FOSS doesn't mean we work for free, right? Right?!?:
There’s also the socialist in me that loves the idea of open source and would happily throw both money and time at it in less trivial ways than I do today if there were a way to make it more properly sustainable than, often, relying on volunteer labour.
Some people are paid to work on these sorts of things. For them it is not a volunteer effort matter. The organisations paying are getting something for their money too: the changes they want on the timescale they are willing to pay for. Because COTS doesn't always cut it.
I’ve actually been in a company that did that - the uni sector in the UK mostly runs Moodle, which is GPL and I spent some time building tools for them (the unis) to push ahead with how they wanted to assess people in their courses.
I also spent vast amounts of other peoples’ money propping up their internal bureaucracies. This is what paid the bills, and I hated it.
This is the crunch point though: I want to spend more time building things with meaning (and be paid for it) but without the utter wasteland that my experience so far produced - I do not want to pour 1000 hours into an integration mashup that is a) not GDPR compliant (don’t get me started but not the UK uni sector), b) full of bollocks because all the data sources are all equally bad so there’s no one source of truth and c) fragile AF because third party vendors cannot be trusted to not break things that work perfectly well, e.g. a daily CSV whose column headings change without notice.
-
@Arantor said in Working on FOSS doesn't mean we work for free, right? Right?!?:
Consider the workflow: you are a business reliant on a piece of software that is copyleft, the vendor goes out of business but the software has a critical bug. (Maybe that’s why the vendor went out of business.) You could, at this point, move to another piece of software or pay someone to fix that bug
Sure, but it doesn't need to be copyleft (GPL(1) or close equivalent) for that to be possible. BSD-licensed software is just as easy to modify, but doesn't have the same level of prohibitions and obligations that "free" (libre) software has.
So, therefore,
Sure, that also gets you into a whole realm of hypotheticals but in RMS’s mind at least, copyleft creates alternatives that closed source explicitly forbids and that is his argument.
is true, but it is not true that copyleft is the only way to achieve that.
And don't forget that in RMS's world, mere "open source" is not free software (in the libre sense of the term "free"), and vice versa. It's only "free" if its licensing is GPL-compatible.
EDIT: belateder than that addition of missing footnote
(1) Every time I write "GPL" meaning the licence, I think of a petrol (gas(oline)) station near the Lille exam centre for the DELF (Diplôme d'éducation en Langue Française or something like that) exams required for, among other things, requesting French nationality. So, it's a petrol station, so what? Well, it had an above-ground tank for LPG, Liquified Petroleum Gas. In French, that's Gaz de Pétrole Liquifié, so the petrol station had a big ol' tank of GPL.
-
@Steve_The_Cynic yes, but BSD/MIT have one gotcha the GPL tries very hard not to have?
What is to stop someone taking BSD/MIT code, building on it and then releasing the product under a closed licence?
That’s the kicker for all the lesser copyleft licences - they don’t adequately prevent downstream taking advantage of it.
No one would have been concerned about the longer term survival of ElasticSearch, Mongo, Redis et al if Amazon had come along, built modified versions and contributed back, especially financially to these products that Amazon itself is improving and selling, but they chose to do it in a way that was detrimental to the ongoing development of these projects. Because these projects had weaker licences, Amazon could just take a dump on them, see also Hashicorp.
I see more projects working sustainably under the GPL than I do under BSD/MIT.
-
@Arantor said in Working on FOSS doesn't mean we work for free, right? Right?!?:
What is to stop someone taking BSD/MIT code, building on it and then releasing the product under a closed licence?
As far as I'm concerned, so what? That doesn't prevent anyone else from doing exactly the same thing. And doing it better!
-
@Arantor said in Working on FOSS doesn't mean we work for free, right? Right?!?:
I see more projects working sustainably under the GPL than I do under BSD/MIT.
But is that because of the license, or because of which part of industry you see most well? The thing I've seen over and over is different parts doing their own things and not really interacting with each other at all. Different tech stacks, different languages, different help sites, different approaches, different licensing norms; all these things vary from one part of the IT industry to another..
-
Releasing something under a closed licence that was previously open is no guarantee of being better - as Elastic, Docker, Hashicorp, Redis, Mongo et al demonstrate it can be the on,y way to be sustainable when someone else comes in, hosts the things you’ve built and doesn’t bother paying you.
That’s really the biggest issue here, but I’ve seen it that open source projects have been forked, closed source, a few minor improvements added, and then sold at profit with the vast bulk of that effort undertaken by people for free and a small group profited off that without contributing back.
GPL, for good and bad, does strongly discourage the profiting off others’ work without contributing back.
As for projects working sustainably… it’s not specific to particular industries, but I’ve been in the open source sandpit for a while, and it seems to me that there are not so many self-sustaining projects that are MIT/BSD (as in, can sustain individuals working on it part/full time without it being solely a volunteer effort).
I can think of many MIT/BSD projects but the number with paid developers I can think of is in single figures - while the number of GPL backed projects with paid developers is much higher.
I’d love to see evidence that MIT/BSD or similarly liberal licensing can be financially viable but there really aren’t many projects I’ve seen that sustained it.
-
@Arantor The GPL encodes the “tit-for-tat”¹ strategy of cooperation, while BSD/MIT encode the “naive”² strategy. Long ago I read a game-theoretical argument that with actors that make mistakes, a mix of those strategies works best. So a mix of those strategies is what we are getting.
¹ Try to cooperate with anybody, but never again with those who fail to return in kind.
² Try to cooperate with everybody and hope the ones who will return in kind outnumber those who will take advantage of you enough for it to work fine.
-
@Bulb the reality with 2 seems to be that the number of people who contribute back is dwarfed by the number of those who would take advantage.
If only being cooperative were better wired into our DNA.
-
@Gern_Blaanston said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@topspin said in Working on FOSS doesn't mean we work for free, right? Right?!?:
FOSS means “free open source software”.
For the end user, it is both “gratis” in that it doesn’t cost anything, as well as “libre”, in that if your printer is being retarded and full of bugs, you (or someone you hire) can fix it. (Literally what caused RMS to create the FOSS movement)RMS is an experienced and knowledgeable programmer. Most people are not. And that is the biggest problem with FOSS software.
It is nice to have the source code available so that you can fix bugs or make changes that you want -- if you possess the necessary skill. But, if you have to pay someone to do it, because you do not have the expertise to do it yourself, then the software is not really free (as in "no cost").
People often use the phrase "free as in beer" meaning that there is no cost. But a better one would be "free as in puppy". A free puppy comes with a certain amount of cost.
Here's a bicycle I made, you can have it for free.
But I need a rack! And I want it painted blue!
All the geometry charts and stuff is included so you can just have it changed any way you like.
So it's not free after all!!!1 
-
@Bulb said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@error said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@dcon said in Working on FOSS doesn't mean we work for free, right? Right?!?:
DAMN YOU PEOPLE, CLICK THE 'DELETE ON MERGE' OPTION IN YOUR PR!!!!
This disabled one?
Your git server admin is a moron.
There are several ways to set up the workflow. IMNSHO some of them are dumb, but some people insist on them:
- Keeping merged branches around is dumb, but some people think they need them, not realizing that they can dig the changes relevant to a feature up by the merge commit, the merge request itself or the link in the task.
Having the branches still available makes it much easier. Especially around people who regularly use
git pullwithout--rebase. And people who merge other people's branches.But in general, I also delete branches in Merge Requests.
- Squash merges are dumb, because now if you dare to merge one feature branch in another, e.g. because the second feature needs some refactoring done in the first one, you'll end up in merge hell as the final history no longer contains the commits from the feature branches. But some people think there is some benefit to the ‘simpler’ history.
Yes, and cherry-picks are no longer possible, nor is bisect... but all this pales in the comparison with the fact that the one squashed commit might contain work of several different people on several different tickets and all that information is completely lost.
We had a short time period when the "squash merge" was officially adopted method and it led to very
moment when I tried to get some background about a database engine change and the official "author" was FE developer that just changed some CSS styles (at the very end of the Feature branch). Basically the classic "who's on first" sketch, but on Slack.
-
@Kamil-Podlesak said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@Bulb said in Working on FOSS doesn't mean we work for free, right? Right?!?:
- Keeping merged branches around is dumb, but some people think they need them, not realizing that they can dig the changes relevant to a feature up by the merge commit, the merge request itself or the link in the task.
Having the branches still available makes it much easier. Especially around people who regularly use
git pullwithout--rebase. And people who merge other people's branches.Are you sure?
- People who pull, whichever way anyway, don't care about any other branch than the one they pull from.
- People who for some reason set their upstream to a feature branch will, if the branch is still there, silently start falling behind as new changes don't get added to the closed branch, but if it is deleted, they'll get an error that the remote branch does not exist and hopefully realize they should switch back to master.
-
@Bulb said in Working on FOSS doesn't mean we work for free, right? Right?!?:
People who pull, whichever way anyway, don't care about any other branch than the one they pull from.
-
They may care at some other time, but when doing
git pullthey don't, because only the branch specified (on command-line or as upstream) affects the outcome of that command.
-
Make them all spend time with Gerrit, where the default operating mode is to review each commit to get perfect commits so rebases are easier later.
You learn how to use Git much more betterer with branches then.
-
@Arantor said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@Bulb the reality with 2 seems to be that the number of people who contribute back is dwarfed by the number of those who would take advantage.
If only being cooperative were better wired into our DNA.
The “take advantage” in this case only means those who start to add proprietary extensions and incompatible changes and try to lock the users into their fork instead. That's not that common.
-
@Bulb I’ve seen it happen a few times on projects I’ve worked with/on. It sucks.
-
@Arantor said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@Steve_The_Cynic yes, but BSD/MIT have one gotcha the GPL tries very hard not to have?
What is to stop someone taking BSD/MIT code, building on it and then releasing the product under a closed licence?
Nothing. It's almost written into the licence itself that you are allowed to do that (subject to the requirement to acknowledge where it came from).
That’s the kicker for all the lesser copyleft licences - they don’t adequately prevent downstream taking advantage of it.
Pedantry: BSD/MIT aren't, as I understand the term, copyleft licences. Copyleft (in my understanding) refers to GPL and similar, where they exploit the copyright system to provide something that is almost the antithesis of copyright.
No one would have been concerned about the longer term survival of ElasticSearch, Mongo, Redis et al if Amazon had come along, built modified versions and contributed back, especially financially to these products that Amazon itself is improving and selling, but they chose to do it in a way that was detrimental to the ongoing development of these projects. Because these projects had weaker licences, Amazon could just take a dump on them, see also Hashicorp.
That suggests that those projects didn't have adequate gatekeeping on contributions, which is equally possible in a GPL project. In both cases, the "nuclear option" tool of last resort is to fork the project and rewind to before the questionable contributions.
I see more projects working sustainably under the GPL than I do under BSD/MIT.
That doesn't surprise me, as such, but it isn't necessarily because of the differences in the licences
-
@Steve_The_Cynic what gatekeeping? How is it gatekeeping on contributions if the offending party doesn’t contribute and doesn’t have to?
You’re right though, MIT/BSD is not strictly copyleft.
-
@Bulb said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@Arantor said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@Bulb the reality with 2 seems to be that the number of people who contribute back is dwarfed by the number of those who would take advantage.
If only being cooperative were better wired into our DNA.
The “take advantage” in this case only means those who start to add proprietary extensions and incompatible changes and try to lock the users into their fork instead. That's not that common.
How would you even know? Say your car nav/stereo/$thingy maker used a butchered BSD underneath. They wouldn't even have to tell you, nor would they have to contribute a driver for a display or GPS they wrote back to upstream, nor would you have any chance to use a mainline BSD if their version was buggy or they decided to stop supporting it.
-
@LaoC Even under the most restrictive copyleft license, you can create a fork for your own use without telling anyone, much less contributing upstream, as long as you keep it internal. (At least I don't know of any that prohibit that.) The specific FOSS license really doesn't matter until you want to redistribute your changes.
-
@HardwareGeek AGPL is the hardcore variant; you want to use it on a website as the server-side? Still gotta release it.
And the distribution angle matters; the FSF took Tivo to court over it because they were shipping GPL-based software on their boxes and were arguing that it wasn't distributing it because they weren't distributing the software separately... (this is ultimately the reason GPLv3 became a thing, to nerf that angle specifically)
-
@Arantor said in Working on FOSS doesn't mean we work for free, right? Right?!?:
AGPL
Oh, yeah. Fuck AGPL with a rusty purple cactus.
-
@HardwareGeek Like even I'm a believer in the cause but even I don't hold with Affero.
-
@LaoC said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@Bulb said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@Arantor said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@Bulb the reality with 2 seems to be that the number of people who contribute back is dwarfed by the number of those who would take advantage.
If only being cooperative were better wired into our DNA.
The “take advantage” in this case only means those who start to add proprietary extensions and incompatible changes and try to lock the users into their fork instead. That's not that common.
How would you even know? Say your car nav/stereo/$thingy maker used a butchered BSD underneath. They wouldn't even have to tell you, nor would they have to contribute a driver for a display or GPS they wrote back to upstream, nor would you have any chance to use a mainline BSD if their version was buggy or they decided to stop supporting it.
That isn't the problematic case as it's not pulling any users over from the previous project.
-
@HardwareGeek said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@LaoC Even under the most restrictive copyleft license, you can create a fork for your own use without telling anyone, much less contributing upstream, as long as you keep it internal. (At least I don't know of any that prohibit that.) The specific FOSS license really doesn't matter until you want to redistribute your changes.
That's true, but putting the binary on hardware you're distributing to customers doesn't count as "keeping it internal".
-
@Arantor said in Working on FOSS doesn't mean we work for free, right? Right?!?:
@Steve_The_Cynic what gatekeeping? How is it gatekeeping on contributions if the offending party doesn’t contribute and doesn’t have to?
I interpreted this part of what you said:
but they chose to do it in a way that was detrimental to the ongoing development of these projects.
as implying that there were contributions ("do it in a way" with "it" presumably(1) referring back to "contributed back" earlier in the sentence) that in some way made the projects worse. This presumption was then reinforced by (and reinforced) my interpretation of "Amazon could just take a dump on them", equally implying that the contributions were junk.
If the projects let such junk in, their gatekeeping was inadequate, but if, as is now clear, they didn't contribute at all, how is that detrimental? (Except in an indirect sense that what Amazon built based on the project was enough better that the mindshare thing shifts in favour of Amazon, because people who don't care about the politics of F/OSS will use Amazon's without a thought.)
(1) Yeah, bad move, I know.
-
@Steve_The_Cynic if they let junk in, that is on them.
What I was getting at is that the hypothetical model where open source works, Amazon could contribute back to these projects by accepting that they profit off them and passing back some of the revenue as a gesture of goodwill / helping to keep the lights on for them.
The problem is that Amazon doesn’t. Hence the retaliation by the likes of Hashicorp to close up their licences for their own protection from Amazon eating their lunch. And Amazon is far from the only offender, it’s just the easiest to point at.
-
@Steve_The_Cynic said in Working on FOSS doesn't mean we work for free, right? Right?!?:
if, as is now clear, they didn't contribute at all, how is that detrimental?
Embrace, extend, extinguish. Much more difficult to do if you have to contribute your extensions back.
-
@ixvedeusi Indeed, but that's what I meant by:
Except in an indirect sense that what Amazon built based on the project was enough better that the mindshare thing shifts in favour of Amazon because people who don't care about the politics of F/OSS will use Amazon's without a thought.
