Strangely, I agree with the original author who sent the DMCA.
Imagine that you're working on some project and you find one of those forks with no GPLv3 attached (or worse, some other license that is more permissive). You integrate it, publish, and then find out that your project is in copyright violation.
I'd rather see a takedown induce an easily remediated repo change, than a big legal PITA down the road.
It's not like he took his repo down, nor any of the forks that kept the license. He just DMCA'd the ones that stripped the license files.
Yeah, you could maybe argue that DMCA is kind of heavy handed but for a change it doesn't seem to be an abuse, it's actually being used for what it was intended. In particular the fact that he went through the trouble of spelling out exactly what was being violated, gave a number of rather reasonable ways to address the violation, and provided links to replacement repos for the code in question I think shows that some time and consideration was spent on this. This wasn't just your run of the mill blanket "OMGZ MY IP!" type of DMCA request we're used to seeing from media companies and certain large corporations.
That was an option that he gave (towards the bottom). I have no idea if GH passes these on to the repo owner or not. GH may have just removed the repos.
If I didn't botch counting, there were 69 of them. Thats a lot of projects to deal with. Dealing with a single point of contact (Github's DMCA procedure) is a lot more efficient than dealing with 69 points of contact.
While I agree with you, if they went through the trouble to copy the code and strip out the obvious attribution and license information, I'd have a hard time believing they didn't do it with intent. I'm no DMCA fan but you'll have a hard time selling me that this was an honest mistake that is easily remedied via some friendly emails. But maybe I'm wrong.
well people here are infringing a GPL license by stripping it from the original code. I agree with that. All my projects on github are GPL so I'd do the exact same thing in the same circumstances.
Question. Is stripping the authors names OK? Obviously stripping the license isn't, but does the GPL require you to "advertise" all contributors? I thought that was considered a problem with one of the BSD variants.
Looking back through Github's DMCA history, Tecnick.com has sent a fair number of DMCA takedown notices in the last couple months. Looks like they're in crackdown mode.
used this in the past but nothing current. interesting to see but it does seem the DCMA request is valid considering what I've been reading about all the stripped gplv3 and other references to OP.
Now for somebody fork it from a recent pull and put in the required licensing and references as, from what I understand, this repo had nice improvements from the OP.
Just a matter of time now.
Actually it does which is kind of the point, it's just those conditions are fairly easy to meet. The TL;DR: version is that if you use LGPL code in your project you have to redistribute the source code of that LGPL code including copyright notices. In practice that requirement can usually be fulfilled by simply providing a link someplace convenient to wherever you original got the code from yourself so it's pretty easy to comply with. Additionally GPL and LGPL differ primarily in that GPL requires that code that uses it must itself be licensed under GPL, where as LGPL allows non-LGPL code to be linked to LGPL code (so long as the conditions of the LGPL code are still being met, primarily the distribution clause). The big difference between GPLv2 and GPLv3 was mostly about closing some loopholes that some companies (primarily TiVo) used to make it impossible to modify certain GPL code on their devices (mainly this involved using hardware DRM). This obviously violated the spirit of the GPL because the whole point is to allow for people to tinker with the code, and preventing that by coupling it to proprietary hardware as a end run around the GPL necessitated the creation of GPLv3 specifically to make doing that a violation of the license.
GPLv3 doesn't have any use conditions; it's free to use, for any purpose. You are probably thinking of the conditions on the things that aren't use - modification, distribution, etc.
If you've been using the repo for a long time and you didn't have a local copy to use in case github went down, etc, your deploy was already broken. Having local mirrors of third party repos, and pulling those into your build system is fairly trivial, and will save a lot of heartache in cases like this.
Either @dineshrabara stripped them since the first commit is based on an official release three days earlier and lacks the notices or they copied it from someone who moved really fast to do so and distributed them in some mysterious third location. The former seems more likely.
ETA: An HN account of the same name was created 235 days ago, so it's probably the same person if they choose to respond.
Imagine that you're working on some project and you find one of those forks with no GPLv3 attached (or worse, some other license that is more permissive). You integrate it, publish, and then find out that your project is in copyright violation.
I'd rather see a takedown induce an easily remediated repo change, than a big legal PITA down the road. It's not like he took his repo down, nor any of the forks that kept the license. He just DMCA'd the ones that stripped the license files.