When I see folks from organisations talking about #OSS (Open Source Software) rather than #FOSS, I assume they're aligned with Microsoft and 'corporate open source': software released under weak open source (not Copyleft) licenses that is amenable to being exploited by corporations who want to incorporate it (usually contributing little or nothing back to the dev community) into their proprietary services and sell it to their oblivious customers.
@djsumdog >Are there any other licenses comparable to the GPLv3 or AGPLv3? Well there's the Lesser GPLv3, which is a lesser masterpiece than the GPLv3.
I've read a lot of software licenses and most of them are terrible.
The only licenses I've read that don't have numerous issues are the AGPLv3, GPLv3 and LGPLv3.
Those "#OSS-loving" corporations *hate* software that uses a Copyleft license (e.g. the GPL, AGPL and other licenses with inheritance clauses) because they can't treat it like gratis R&D they can exploit. I believe proprietary software is ethically tainted (https://davelane.nz/proprietary) and so I have zero respect for those corporations. I also have a substantially lower impression of institutions/orgs that align themselves with those corps by referring to OSS rather than #FOSS. Details matter.
@ocdtrekkie >I prefer permissive licensing So called "permissive licenses" are usually more restrictive than the GPLv3, since they take a massive overstep by writing that copyright covers usage of software.
MIT expat: "the rights to *******use*******, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, *******subject to the following conditions*******:"
GPLv3: "You are not required to accept this License in order to receive or run a copy of the Program. Ancillary propagation of a covered work occurring solely as a consequence of using peer-to-peer transmission to receive a copy likewise does not require acceptance. However, nothing other than this License grants you permission to propagate or modify any covered work. These actions infringe copyright if you do not accept this License. Therefore, by modifying or propagating a covered work, you indicate your acceptance of this License to do so."
> I think the failure of the OSI to accept SSPL or a similar variant thereof is a massive failure of the people we let dictate the definition of the term "open source". The "SPPL" is a proprietary software licensed designed so that MongoDB can go collect money for exceptions when they feel like that they need some more money. MongoDB tried to do this with the AGPLv3 before, but they soon realized that businesses they attempted to extort would just comply with the AGPLv3's requirements and not pay them.
Their "fix" to this was to take the AGPLv3 and modify it into a license that is almost impossible to comply with, so extorted companies would be forced to pay for a license exception.
The "OSI" is pretty good at slipping proprietary software licenses past the "open source definition" (https://opensource.org/osd - yup there's proprietary JavaScript on that page and it's clownflared), but even they couldn't get the "SSPL" past the 10 requirements.
>If I wrote something I expected to get ripped off by Amazon, I would probably copyleft it. "Amazon" happily uses trivial libraries under pushover terms just to save a buck and a few hours, you should assume that they will.
You should license all your software under the AGPLv3-or-later or the GPLv3-or-later or in very rare situations the LGPLv3-or-later.
If you really wish to surrender all your software to the proprietary overlords, only this license will do a proper job once it's ready: https://wpdd.info/
@lightweight I might worry about that if I wrote code I expected valuable derivatives of. =) If I wrote something I expected to get ripped off by Amazon, I would probably copyleft it.
@ocdtrekkie ok, fair enough. The only downside to that approach is that you might end up having to buy back a future (slight or big) derivative of your original code. Which, to me, seems a bit on the nose. It's your call, of course.
@lightweight I permissively license all of my code because I'm not interested in holding any conditions over it. I have no intention of monetizing it, and have no real desire to track usage (or misusage) of it.
@ocdtrekkie I can't understand why a developer would license their code with a permissive open source license, unless their goal is to maximise uptake, especially by corporate users, who'll incorporate their code into their proprietary products. If you're developing an end-user app, then it makes no sense to me at all. I tend to think those devs are either naive or planning to close their code in future (or use an 'open core' model). **spits** Got no time for that.
@lightweight I prefer permissive licensing but I think we need solutions to enable open source authors to charge large corporations heavily for usage. I think the failure of the OSI to accept SSPL or a similar variant thereof is a massive failure of the people we let dictate the definition of the term "open source".
@ocdtrekkie >There's so much wrong in this comment I'm just going to pass on addressing it, tbh. You're telling me that I'm so correct that you cannot think up of any counterarguments that won't fall flat?
@edavies@lightweight If the software is incidental, what harm does it make if it's on a reciprocal license?
A reciprocal license only gets in the way if someone wants to exploit the software, add anti-features, add proprietary features etc. Or if one has corporate counsels living in 1999.
@lightweight For every mega corporation which “exploits” open-source software there are probably dozens of small companies for whom much of their software is entirely incidental to their business.
Last year (and probably next week) I had a medical scan which I'm most intrigued by (it seems indistinguishable from magic). A nice aspect was that after the technician had done the scan the results were on the consultant's computer in minutes. I'd be entirely happy if that was done with open-source software and would hate to have this sort of innovation tripped up by viral copyleft licences.
Personally, I'd never use GPL for any software I wrote though I could be persuaded to consider LGPL.
@thamesynne >the LGPL only exists because there was some legal ambiguity as to whether linking to a dynamically-loaded library was derivation or not, which the LGPL resolves by explicitly saying "no, it isn't" I suggest you actually read the LGPLv2.1: https://www.gnu.org/licenses/old-licenses/lgpl-2.1.html
The LGPLv2.1 notes that a linking with a dynamically loaded library makes a derivative work, but an exception is made to allow proprietary derivative works as long as the LGPLv2.1'd parts remain LGPLv2.1 and that users are permitted to reverse engineer the proprietary parts for debugging modifications.
Otherwise the LGPLv2.1 is the GPLv2 - you can choose to not use the exception and use only the GPLv2's terms instead.
>but in practice, one could take a GPL'd library, write a server around it that responds to incoming RPC requests by calling the exact functions and sending out their replies, release the code to that server, and use it to serve one's proprietary program with exactly the same results as an LGPL'd equivalent. You'll end up with a chain of derivative works no matter how long you make that chain - a decision if doing such is copyright and contract infringement is really up to a court, but courts tend to see through circumvention attempts.
@edavies@clacke@lightweight as i understand it (and i've just spent the last hour checking), the LGPL only exists because there was some legal ambiguity as to whether linking to a dynamically-loaded library was derivation or not, which the LGPL resolves by explicitly saying "no, it isn't". but other than that, it *is* basically the GPL, and anything that does more than link to it is under the same obligations as a GPL'd program would be.
but in practice, one could take a GPL'd library, write a server around it that responds to incoming RPC requests by calling the exact functions and sending out their replies, release the code to that server, and use it to serve one's proprietary program with exactly the same results as an LGPL'd equivalent.
if only the industry had focused on making RPC/IPC fast enough that dynamic linking was seen as an unnecessary complication, the question wouldn't have arisen in the first place...
@thamesynne the word 'viral', with very negative connotations, was adopted by corporate interests. It's not viral, it simply applies to derivative or dependent works. It's inherited. @edavies@clacke