Be at uni

>Be at uni.
>Professor "You'll shift your research work to be compliant with the confidentiality agreement. You won't disclose source code from your research. There are people out there that are just waiting for something like this to happen and make profit at our expense."
>Me: "O- Ok prof."

How to argue for opensource with "real people"?
The only arguments that came to my mind were meme stuff like "You're afraid of being spied on but you use windows as an OS on both your personal machine and our laboratory computers" or "The NSA spies on us anyway. They are likely to have a backdoor on our lab's git repository"

Other urls found in this thread:

universalrobotics.com/
twitter.com/SFWRedditImages

Well, you can release whatever you make with a GPLv2 license (or another one that promotes FLOSS), which will mean that no one can take your work and profit off of it. Like that's literally the whole point of a license.

Anything funded with public money (which probably includes your university research) should really be made public domain, or at least copylefted. I'd advocate GPLing it all after the research is published (assuming that's when the confidentiality agreement expires).

I'll likely get an answer like "companies don't care for license, they'll just use your code. You can't know it's our lab's work once its compiled"

BTW, you're supposed to share your work under the GPL if you use GPL licensed tools, right? Does that means everything out of gcc is GPL contaminated?

The thesis is also confidential. Not even released on the uni's library. Just on printed text format (and PDF on the server) for other members to read it.

No I don't believe that there are any legal issues regarding the GPL infecting compiled code. Ive at least never heard of it being a problem.

OP what is your research about, not minig data , just genuinely curious about the whole secrecy.

I can't into details of licenses...

I'm in robotics. Basically its very efficient and robust motion control (but the results can be generalized to other topics). But the underlying idea is "so simple" that a company that figures out the "secret" would be able to mass produce devices with the same quality as our cheap homemade robotic devices we student use for experiments.

I guess its like if NASA said "lol magic" when asked how did they did the guidance and navigation for the Apollo instead of mentioning they used a Kalman filter.

I mean, if it's confidential for a good reason, then yeah it shouldn't be open source.

Well, then you can tell him that violating a license can lead to legal action, so duh they won't do that.

desu that would be a good thing for everyone if everyone took your idea and made it widespread if it's actually that good. Could be a huge improvement for robotics.

>robotics
It's too late, a company already won the robotics race.
universalrobotics.com/

And who's going to enforce that? Are your going to decompile every application? Can't even just compare the binary since different compilers and compiler options will result in massive differences.

>Basically its very efficient and robust motion control (but the results can be generalized to other topics). But the underlying idea is "so simple" that a company that figures out the "secret" would be able to mass produce devices with the same quality as our cheap homemade robotic devices we student use for experiments.
In most cases, something that looks like it would be amazing in academia turns out to be garbage in real world applications.

That's true. So then I'm misunderstanding the point of a license. What does it actually do, then, if it can't really be enforced for proprietary software?

For the legal part, I was intern in a startup and they didn't give two shit about stealing code on github without giving proper reference / credits

For the idea part, we use the magic spelling the teacher proposed but he's meming himself in consulting so I guess he wants all the money for him

> in most case
Do you have examples?

>he's meming himself in consulting so I guess he wants all the money for him
I can give you an email address to give the details to. There's being an entrepreneur and there's being a greedy fuck. If it's some technology that can help the world it really should be free.

>examples
Cameras come to mind.
>just stick cameras on the machine, if it can see we can make it figure out the orientation of the part!
>oh fuck we changed the ceiling lights to new LED-based lights and now all our fucking machines are down because the brighter environment fucked the contrast calibration
>fuck, this machine is down
>oh, it's just some grease on the camera that stopped it from functioning

There no shenanigans so I'm not going to go full snowden and lose my credibility as a researcher

>lose my credibility as a researcher
You know there are methods that whistleblowers use to leak information without getting caught, right? If this magical robot tech is so simple and easy then you'd be able to leak it pretty easily without drawing attention to yourself.

pretty sure the GPL doesn't prevent that.

>take your work and profit off of it
It specifically allows people to take your work and profit off of it, but only if they share the changes they make to it with everyone that uses it.

do GPL shills even read their own fucking license

Why would anyone release a powerful idea for free
I mean, I myself publish my hobby code under the MIT license on github but I understand the appeal of keeping things for you, especially if money is related.

It's not the optimal thing to do for society but life isn't a cooperative game.
If one don't meme himself to success, nobody's gonna be there to have your back.

The FOSS guys won't help you if you can't pay your loan. You're there alone.

Learn that ur opinion is invalid because you have not established yourself

That's basically saying everyone but the inventor should be able to capitalize on the researchers idea and make money while user doesn't. Fuck them all user, keep the code to yourself. Sounds like you'll be rolling in money soon. Seriously, don't fuck it up.

The main point of the fsf is to legally enforce and protect the gpl. They will cover your ass. Tell that to your prof.
Also if you don't release the binary you don't need to release the source either.

I havsaid to a couple of professors to fuck off for forcing students to use matlab. I have always suggested opensource alternatives to them, like python and ROOT for computing and visualization. Usually they let students to use these, other times I have written letters of protests to the dean and head of the department for forcing students to use malware.

Luckily both the dean and head are actually very pro opensource, and usually make things rigth. Research should be done on 100% free software, and be fully open to the public.

Matlab is fucking based tho.
I have like open source software : I use linux as the only OS on my device since I'm out of highschool, I write my reports in pdf and make my diagrams with inkscape.

But man, matlab is so much betten than the "alternatives"
> I have written letters of protests to the dean and head of the department for forcing students to use malware.
>malware
Audible kek.
You're either a troll or stallman-tier "I live in an distorted reality"


>Research should be done on 100% free software,
Yeah, it's better if the free alternative is relevant
>and be fully open to the public.
No.

Research is backed by the public, why should it not be open to the public?

>Research is backed by the public
>Implying all the money that goes into research comes from citizens' taxes.
Ok, now you don't know what you're talking about kiddo.

Also, do you really thing that the research conducted by the army should be made public?
Every one is not naive like GNU/RMS and refuses to use a computer password.

Wait, do students in the US actually have to pay to learn? That's fucked.

GPL is pretty hard to actually enforce. But stuff to do with proper use of licensed code does come up when a company's code is being audited for a buyout or something.

Not from the US, but I'm in a private Uni.
Expensive as hell if you ask me, but the uni being known for "manufacturing CEOs", the alumni network and image in the public makes it worthwhile in the long run.

As for the money, the research money comes from either the government, the tuition fees or companies investment.

The US doesn't have a strong enough economy to support free education.

Also that stupid light powered wifi thing.

This is why the GPL is such harmful cancer - even the shills don't actually understand what it covers.
The only thing it really guarantees is that any and all source code that is part of a whole for a program is to be made open and free to copy/use/modify/distribute should said program use ANY GPL-infested code.
For a dumb example, let's say you write a program that collects shitposts off Sup Forums. Your code is your own and nobody has a right to it. You use a library that identifies shitposting because that exists but, because it is under a not-shit license like MIT, Mozilla Public License, Public Domain, etc. your code is still your own but the library you're using still adheres to its own licensing rules.
Then you decide you need a library to count because we're using JavaShit in our imaginary world right now. It's released under Glorious GPL of Freedom. Not only is your own code now required to be licensed under the GPL, the shitpost-identifying library is to be re-licensed under the GPL as well.
Even though the author of shitpost-identifier wasn't responsible for your choice to use it and maybe even purposely didn't license the code under the GPL, said author's code is now GPLd.
Your code is GPLd.
All because of an insignificant shitty GPLd library that contributed little if anything to your project.
The LGPL exists to handle inclusion of libraries so that the LGPL doesn't infect the rest of your code but this requires said library to be dynamically linked. If it is statically linked it is considered as "part of the binary" and, so, infects the rest of the code it has nothing to do with.
It's a fascist license that stomps on the Freedoms of all programmers. Use the People's License. Use Mozilla Public License.

No. GPL tools can be used to build non-GPL programs. However, if said program embeds the GPL'd tool into its source code, then you done fucked up and have a GPLd shitfest on your hands.

> the shitpost-identifying library is to be re-licensed under the GPL as well
Not true, it just needs to be GPL compatible. For example, the MIT and BSD3 licenses are ok.

See
And any gpl compatible licenses. Also learn to dual license.

Compatibility with the GPL only means that the program as a whole, including the libraries using gpl-compatibile licenses, can be licensed completely under a singular GPL.
A program using a GPL library and an MIT library would then be required to be licensed under the GPL; the MIT license allows this hence being GPL-compatible.