52 comments

[ 3.1 ms ] story [ 108 ms ] thread
what is/was source3?
> we set out to recognize, organize and analyze branded intellectual property in user-generated content

Looks like technology to fight copyright infringement

Agreed; also, turn it around and it becomes a user-generated-content tool as well. As in anything that goes viral, the brand owner can take ownership of and exploit as an unauthorized derivative work, free of any user recourse.

Well, except Fair Use, but I'd hope that's part of their model to consider, because it's important I think.

But nobody really knows what fair use means. We have a general idea but that's it.
No, we do know what Fair Use is it's just that people like to claim it as an excuse when it's not valid.

CASE IN POINT: Mike Masnick. That sonofabitch is the most ignorant loud-mouth "I ain't never made nothing creative so everything is Fair Use" sinkhole asshole since Nick Denton.

Fair Use is alive and well. Idiots who try to wear it as a defense cloak get what's coming to them. I know the right lawyers to dail up if I ever need to, but I haven't needed to, because I'm not Masnick.

Might be a smart move by Facebook: if this technology was used on Facebook, then they might have to shut down half of Facebook. Better to buy them out.
Possibly, but they could also choose to enforce copyright claims on people who can pay. Or maybe facebook just doesn't care about my meme pages and wants to purge them so I spend more time looking at branded content.
The porn industry has been doing this for a while. I remember about 4-5 years ago were services available to content producers that scanned tube sites and newsbins for copyrighted content. You had to pay for it though (SaaS, they wouldn't sell the code). It was somewhat of a racket, because the people selling the detection service were also running the tube sites that were condoning piracy.

It's amazing how the porn industry is often on the front line of technology. They rarely get any notice though.

Does it predate content ID on youtube? I suppose the difference with ContentID is that it's free to file a claim and it's limited to youtube and not multiple tube sites.
I'm guessing something like a "Let's Go Crazy Babydance" detector. Combine Shazam with image recognition seeded with brands and product images and you can cover a lot of UGC with takedowns or ad rev takeovers.
(comment deleted)
That's a promising move by Facebook. Maybe they're finally going to start cracking down on the rampant freebooting that dominates their platform.
I have the strange feeling this will really only be used to enforce the copyrights of large music and movie industry corporations and will not be effective at protecting the IP of independent creators.
It looks like it's a system similar to contact ID. I hope this helps slow down all the free booting.
Wonder how much they'll be using the technology and how. Will they try to do it like YouTube with trying to keep videos up, and having the revenue go to the copyright owner? Or maybe they won't do much. Comments seem to be divided.
I suspect that it will more likely be similar to YouTube where they try to keep images and other IP up, and generate ad revenue for both FB and the creators. Most of the team behind Source3 worked on a startup called Rightsflow, which did music licensing and eventually was acquired by YouTube. They were responsible for doing a lot (all?) of the music licensing for videos, so that instead of videos being taken down they could instead stay up and have rev-share for everyone.
A major congrats to the source3 team! I'm excited for them and for the changes they'll bring to Facebook.
Did Facebook ever consider building out their video product outside the newsfeed to keep people on the platform? YouTube seems like easy prey with FB's reach.
They are/have. Several news sites are already embedding FB videos.

If you're asking if they'll make a social network with video... They already have. It's Facebook. And they've recently begun pushing live video a lot harder.

YouTube has great discoverability but no shareability with video. Facebook has the exact opposite. I wonder why they don't tackle this. IMHO videos in the newsfeed are not enough to attract content creators. There needs to be some kind of special section/UI.
Wait what? YouTube videos can be embedded, shareable, with time ranges and what not? with relatively short URLs by default?

That's not "shareable"? That seems like exactly what the shareable web ought to be!

What Facebook has that Youtube does not is that you can share with the click of a single button (without any prior setup) to friends and family. For a not-insignificant-portion of internet users, Facebook is how we communicate with others, so shareability to Facebook matters quite a lot.

I don't disagree that youtube is just as easy to use, but that doesn't change what Facebook has that youtube doesn't.

„shareability“ not from a users POV but from a creators POV: how can the videos I create reach an audience that hasn't seen them before. Examples: Instagram hashtags, Facebook Share-Button, Facebook-Likes/Visibility, etc. Whoever solves this in combination with a great UI for discovering videos (and making content creation easy) will be the next YouTube.
It can also be difficult to share Facebook videos unless you want to share to your timeline or send an individual Facebook message.
What happens when the thieves acquire the police? Who is going to police content, and for what purpose now? We already know that Facebook has profited tremendously from the freebooters.
In a more real-world sense, I think you can look at the current state of law enforcement around drug laws in the US for a solid guide to the sort of relationships you're describing. Eventually, the police and the thieves become the same thing; the state thieves.
In the United States, more private property was lost last year to "civil asset forfeiture" than to conventional thieves. https://www.washingtonpost.com/news/wonk/wp/2015/11/23/cops-...
neither the article nor the year it refers to were last year. And even the article says it isn't true:

"...but this isn't exactly right: The FBI also tracks property losses from larceny and theft, in addition to plain ol' burglary. If you add up all the property stolen in 2014, from burglary, theft, motor vehicle theft and other means, you arrive at roughly $12.3 billion, according to the FBI. That's more than double the federal asset forfeiture haul."

Furthermore, a big chunk of that 2014 number was Bernie Madoff, per your atricle.

And finally, a big chunk actually goes back to the victims of the crimes ultimately linked to the forfeiture, again per your article.

They're both corporations. A thieves/police distinction has always been meaningless.

If you want content to be policed because policing content is a social good, make the government (the actual police) do it.

If you want content to be policed because it's profitable, the vertical integration of "thieves" and "police" lets you leverage synergies.

Can't wait to see how source3 helps Facebook be a more despicable service.
Jeez, talk about hold-outs. Lots of grandparents have already been on Facebook for a number of years.
Thats great. But why is this on hackernews ? Whats the relevance ? There is a single page linked to this.
booom! http://www.source3.io/bundle.css (open source CSS) ???? WTFFFFF!!!! what have you contributed to Open Source and internet overall?

>* Bootstrap v3.3.7 (http://getbootstrap.com) * Copyright 2011-2016 Twitter, Inc. * Licensed under MIT (https://github.com/twbs/bootstrap/blob/master/LICENSE)

I fail to see what you're trying to point to here.

Are you saying a company shouldn't be allowed to use open source if they're not "contributing to open source and the internet overall"?

I'm sorry, what do you mean by this comment? If they use MIT licensed software there's no obligation for them to contribute back anything, not that it's good or bad, but it's the terms of the license.

Also there's no way to know if they have contributed or not, and even if there's no direct contribution (pull reqs) we can't know if they haven't reported bugs or supported employees that wanted to collaborate to the projects on their free time.

(comment deleted)
(comment deleted)
Dear mom: when I grow up, I want to create the world's best content identification and DRM system, and along the way build a platform to manage IP online and establish relationships with brands!
Johnny, we could not be more proud of you. Love, Mom.
"When I grow up I want to hold an honest job, be an upstanding citizen, earn a strong wage, pay my fair share of taxes, and work on interesting problems, technical or otherwise."

If we can respect lawyers and accountants, surely we can respect people who have worked hard to grasp an economic opportunity, even if we wouldn't in a thousand years worked on the problem they chose.

Hadn't heard of Source3 before but would they potentially be the first investment Facebook has made in a Youtube-like Content ID program? Given the accusations and complaints of folks who find their YouTube content re-uploaded to FB without permission or compensation, maybe this will be a good thing (if you've more-or-less accepted the status quo of Content ID and its drawbacks).

A highly-upvoted HN discussion on a Slate article about YouTube videos being plagiarized by Facebook users: https://news.ycombinator.com/item?id=9854160

It makes complete sense too as FB makes bigger plays at pre/mid-roll ads in videos. Seems like this could be an effort to limit their liability from serving ads on top of work with copyright issues and then profiting from it (or paying out a revshare to the uploader who may not be the rights holder).
Will their system be able to discern fair use or just brand anything that contains samples of "IP" to be "illegal"?

Content owners already have far too much control over their wares.