Thanks for the breakdown on work for hire -- and I 100% agree, regarding hiring a lawyer to look at contracts.
(background: I've been freelancing for about twelve years, with several engagements that have resulted in open source contributions)
If you're freelancing, the general rule is that the customer owns everything you produce within the scope of the contract. This means you do not own the copyright, and therefore you can't open source the code. The specific phrase you look for in your contracts is "work for hire"
It's very difficult to get around that rule, or to outline specific exceptions ahead of time -- but you can change the incentives to encourage your client to contribute code to the open source community.
I offer to reduce my rates for any work we mutually agree to release as open source. I benefit by getting my name and work in broader distribution, my clients benefit by paying less for the work, and the open source world grows a little bit. It's a reasonable trade off for all parties, and even if most clients don't exercise that option, they appreciate the spirit of such an offer.
Where it gets hairy is when you're making changes to code that has been open sourced under a "viral" license, like the GPL. If that is the case, then you should inform your client that they are bound by the terms of that license -- that those changes *necessarily* become open source. Keep a copy of those emails. If your client decides that they're going to skip out on that obligation, you'll want to make sure your ass is covered if/when your client gets in trouble
By "release" I mean "to drop the claim against the video in question via the YouTube Content Management System." It's a specific action that needed to be taken with the claim tool, not a statement about releasing in the sense you're talking about. Forgive the colloquialism?
Heya -- sorry I missed your comment earlier. We released the video before we engaged here, and this morning we figured out what happened.
Plainly stated, one of our reviewers incorrectly reasserted the claim during a manual review. We're changing how we review things to address the issue in the long term. You're welcome to search this discussion for my other posts, where I provide more detail about how it happened.
I appreciate the measured response, and the thoughtful questions. Thank you.
1) The automated system isn't ours -- it's YouTube's. We don't see any of the disputes until the video creator responds to YouTube's canned responses about copyright infringement, by which time they're usually pretty pissed. The vast majority of disputes that reach us are legitimate, and released quickly. To be clear: we have no control over the initial copyright notices, we can only respond to the escalated disputes. That said, the basic concept you bring up is correct: most people don't fight the takedowns.
2) We don't automate our responses to the disputes we receive. Every single one of them is reviewed by a human, and in this case, our manual review process failed. Regarding your point about communication, it's also very difficult for us to contact the video creator; we don't get access to their personal contact information, so we either have to receive a separate email from them, or we're stuck using the canned responses from YouTube. It's a significant hurdle. I believe resolving the communication issues would fix a huge number of these escalations.
3) I can't comment on the business model, but we do believe that there is a sustainable path that doesn't screw video creators or musicians for their ability to make a living on what they create.
I'll be happy to follow up on any more questions you have.
Heya -- we do review the videos in person, and we made a mistake that we're working to fix.
Update from the OP: http://yro.slashdot.org/comments.pl?sid=2693545&cid=39170585
Update from Rumblefish: http://yro.slashdot.org/comments.pl?sid=2693545&cid=39174761
AMA on Reddit with the Rumblefish CEO: http://www.reddit.com/r/IAmA/comments/q7via/im_the_ceo_of_rumblefish_i_guess_were_the_newest/
I'm responding here because a lot of the discussions and updates are getting buried below. We're paying attention.
-Peat Bakke, Lead Architect @ Rumblefish
It looks like one of our manual reviewers did reassert the claim, and we're working out how to fix that process on our end.
There's more info here: http://yro.slashdot.org/comments.pl?sid=2693545&cid=39174761
There's a lot of threads here and on other places, hopefully this sheds some light on the claim process, and what happened with eeplox's video.
Here's the sequence, as best as I understand it:
- eeplox posted a video to YouTube that contained bird sounds.
- The automated YouTube content identification system mistakenly assigned it to one of the tracks in our catalog.
- eeplox contested the automated claim, which sent it into the manual review queue.
- One of our reviewers reinstated the claim, which triggered the response eeplox received and posted above.
When we heard about the story on Slashdot, we reviewed the video and released it.
TL;DR: We messed up during the manual review process. We've cleared eeplox's video. We're working on fixing the manual review process.
I'm here in the discussion to answer what questions we can. We're not interested in screwing over independent musicians and video creators, and we want to be as open as possible about what's going on.
If you run into content claim issues with our catalog on YouTube, you're welcome to contact us directly at YouTubeContentID@rumblefish.com. We have a FAQ about the process here: http://rumblefish.com/id/youtube-content-id.php
Paul Anthony (the CEO) has an AMA thread over at Reddit where he's answering questions. http://www.reddit.com/r/IAmA/comments/q7via/im_the_ceo_of_rumblefish_i_guess_were_the_newest/
The simple and necessary action (reviewing, confirming, and releasing the video) was done very shortly after we became aware of the issue -- several hours ago.
Right now, I'm interested in responding to reasonable questions as best I can; next, we will be reviewing what happened, and following up appropriately.
It's not a wasted evening. People are pissed, and although I can't provide the ultimate answers right now, I don't think it's appropriate to simply disappear.
I have my suspicions, but I'm going to hold off on public speculation.
Things will be much more clear tomorrow, when we can actually review the complete process.
Send me an email at firstname.lastname@example.org, and I'll be in touch.
We'll sort out exactly what happened in the next couple of days -- I really wish I could have a clear cut answer for you right now.
I completely agree on your point that it's a serious problem. I joined the company last year because they're out to fix an industry that's fundamentally broken and frequently abusive; that's why I'm spending my Sunday night trying to sort this out.
Anyhow -- I know this doesn't directly address your concerns. Sorry about that. When there's more to share, we'll figure out the right place to put it.
Forgive us for being a little short in our responses -- it is a complex issue, and we're working through it on a Sunday evening.
The YouTube content identification and dispute resolution system is not a simple one; right now we're focused on resolving the critical issue with eeplox. This is something that will unfold over the next couple of days, and we'll be doing what we should -- make sure we figure out exactly what happened, and put guards in place to prevent it in the future.
As of next Thursday, UNIX will be flushed in favor of TOPS-10. Please update your programs.