blog.katsbits.com

Click to get the RSS master feed

xx Content ID, Sintel, Sony & Plausible Deniability

April 07, 2014, 01:02:23 pm by kat
It's important to note that whilst the basic premise of Copyright, its infringement, and its remedy, is pretty straightforward, the way persons and party's behave with respect to that is not. It can, and often is, extremely messy, complicated and convoluted simply because involved parties want to mitigate their liabilities with respect to policing content and the respective responsibilities and obligations therein. Put plainly, no-one wants to be held liable for the actions of others, nor be seen to benefit from said actions as a result.

This material is provided for informational purposes only and should NOT be construed as legal advice on Copyright, Intellectual Property Rights or DMCA. It is highly recommended legal Council be sought in matters of law.

With the above in mind, now the dust has settled on the Sintel issue we can turn our attentions to answering somewhat the “why's” and “wherefores” of YouTube being able to do what it does with apparent impunity; just how are they able to make what often turn out to be false or erroneous claims of infringement without compromising their own standing on Copyright?

By defining 'content' claims relative to “Terms of Service” infractions rather than “Copyright” infringements.

Lets be clear about this, the language contained in a typical Content ID Claim Notice is indicative of its purported purpose; it expressly questions whether the accused has the appropriate permissions to use flagged materials. That, simply put, is an issue of “Copy Right” - having the 'right' (license or permission) to 'copy' (exploit or use) someone else's property. No "if's", "but's" or "may be's".

Or so the reasonable thinking wo/man would think.

There's a Jedi mind-trick going on here you see because, whilst the aforementioned reasonable wo/man is thinking “that's a Copyright claim isn't it? And aren't Copyright holders, or their representatives, the only ones supposedly able can issue such claims?”, YouTube, with feigned shock, is saying “you said the video was yours? You lied to us so we're taking it down”. See the mind-trick? We think it's a Copyright Claim, and respond with tremendous levels of tumult, YouTube/Google see only Terms of Service violations and doesn't see what all the fuss is about.

So, the simple facts that 1) the Content ID system is in affect a de facto loophole around the obligations and responsibilities attributable to Copyright, and 2) that YouTube (ab)uses Terms of Service policy enforcement to prosecute Copyright, allows for Content ID claims to be phrased anyway YouTube see's fit, and call the process whatever it wants, just so long as neither aspects directly invokes Copyright and DMCA as commonly understood. In maintaining this faux posture, or continuing the mind-trick, it is not possible to hold them liable for losses incurred as a result of an action because, in their eyes, it has nothing to do with Copyright and everything to do with the Terms of Service Agreement users declare against when uploading videos.

If that wasn't bad enough, add “Errors & Omission Excepted” to the mix, and whatever accountability might have been left flies right out of the window; no-one can be held liable for anything, even if they were/are/could be, all because of an apparent disassociated and dispassionate machine-code error (a false-positive). And as everyone knows, machines harbor no grudges, malice or intent to cause harm so they cannot be held to account for their 'mistakes' either.

All-in-all this whole situation is possibly the best example of deliberately executed “plausible deniability” anyone, never mind YouTube/Google et al, could have asked for.

Additional Reading
2 comments | Write Comment

xx Sony blocks Sintel on YouTube

April 06, 2014, 01:05:20 am by kat
The official version of Blender Foundation Sintel movie appears to have fallen foul of YouTube's automated Content ID Matching system which checks videos for Copyright infringement - if the system finds what it thinks is a 'match' the associated video is pulled leaving the author, depending on the severity of the match, to appeal the decision. As KatsBits found out recently this whole process results in far too many 'false positives', leaving the author to contest the decision, providing YouTube/Google see's fit to provide the option. It is not always made available and when it is, the process to getting content reinstated can be prolonged and protracted.


The more insidious aspect in this however is the Content ID systems apparent operation under the auspices of third-parties that might not otherwise be aware of what's going on in their name. And this is where the whole issue of flagging content gets very murky because for an Copyright and/or content infringement claim to be legitimate it is supposed to be made by the Copyright owner, it cannot be done by a third-party unless said party is appointed an 'agent' for that purpose (which raises its own questions). Unfortunately YouTube/Google have yet to clarify their position with respect to Content ID operating in this way.

Equally troubling in this regard is that if YouTube/Google are acting as an third-party agent, it means they may very be in direct conflict with their claiming "Safe Harbor" under US Copyright Law which requires they not actively "police" their content or network. Failure in this regards tends to imply loss of Safe Harbor status and opening themselves up to all sorts of liabilities, something they've gone to great lengths to avoid previously, by hiding behind that very same "Safe Harbor" claim - "we don't 'police' or 'monitor' uploads or our network". For all intents and purposes it appears this is no longer the case.

The upshot here for Blender Foundation is not to bother, necessarily, with YouTube's appeal process, but rather to have their legal representative contact Sony PM&S directly to clarify the situation and, depending on the outcome of that, send legal notice to YouTube to reinstate the material with immediate effect (assuming the claim to be both false and Sony having nothing to do with any claims of infringement).

The broader picture is that now YouTube/Google is clearly policing content, their Safe Harbor status should be questioned. Someone needs some very large brass monkey's for that though.
8 comments | Write Comment

xx Generator, why so long to make it?

April 03, 2014, 09:31:30 am by kat
One of the main issues modders face making their projects is the amount of time they take to produce; visual quality, asset production, general construction etc. For instance Mages Skulls took about eighteen months to produce, Relic Hunt nine months and Castle Keep roughly three (largely due to it being relatively simple). They all pale in comparison to Generator being on the back-burner for about ten years! That's a long time in anyone's book! So why does this happen?

In both a general sense and specific to this single player project, the reason for such protracted develop times was due to the way the hard coded limitations of Return to Castle Wolfensteins are conflated with the use of Q3Map2 and the complexities possible using that tool to compile levels; the notorious shader overflow bug in particular causes big problems that don't usually show up until a level has had a 'final' quality compile done to check appearances, shading, shadows etc. Unfortunately it's quite typical at the point this happens that a lot of work has already been done which forces the developer to pause for thought. It should be noted that this is not to say that Q3Map2 actually causes problems but rather the combination of both RtCW and the compiler exacerbating issues that are not always easy to work around.

This was the problem encountered during Generator production; the first level had terrain related shader issues; the second and third were originally part of the same larger level that had more complex gameplay, but had to be split and simplified a great deal to fit within the limitations discussed above. These types of adjustments are never simple operations because the resulting sections need to have continuity, and when removing content, this can cause some pretty major headaches that might mean having to rebuild, entirely. It's at these points that production is often halted. When that happens other things tend take over and before you know it a lot time has passed.

Some screen shots can be found here - in-game and in-editor

0 comments | Write Comment

xx YouTube & Not Approved for Monetisation

March 03, 2014, 08:46:05 pm by kat
Tip for YouTube vLoggers. If you don't want to have Google's monetisation bot poking around your content do not use any of the suggested video "Tags" that pop up at point of upload.

Recently one of KatsBits recently uploaded videos was "not approved for monetisation" by YouTube* (note: issue has been partially resolved. Ed). One has to wonder 'why?' given the video in question is a simple time-lapse capture of yours truly mouse-fudging the production of a normal map texture using Corel Photo-Paint, nJob and Blender. It's ostensibly "informational" and/or "educational" in nature and does not "[reverse engineer, lend, sub-lease or license, materially misrepresent, etc. etc. etc.]" the applications used as per their respective EULA's.

So why does such content get flagged or come to the notice of YouTube's Content ID system?

No-one really knows, exactly, how the system works, but from observation it appears attention is triggered through the inclusion of one or more YouTube/Google suggested video "Tags" - these are collated keywords or search terms Google/YouTube thinks a person might use to find a given video on a given subject, so are suggested at the slightest prompt.

In this particular instance only one such tag was used; "Corel PHOTO-PAINT (Software)" - ironically, if a person was searching for videos on, about, or using Corel's Photo-Paint application, they would likely simply use "Corel Photo-Paint" or "Photo-Paint", not "Corel Photo-Paint (software)" as that's a little verbose and beyond the needs of most.

But that's slightly besides the point because the issue at hand here is not so much that tags trigger the attention of YouTube/Google Content ID bots, but that any video caught in this vast Content ID dragnet can be flagged, seemingly without explanation, as a Terms or Service violation whilst questioning authors having the correct "commercial use rights". Copyright in other words.

One does have to wonder just how Google/YouTube is able to justify (anonymous) content flagging in this context when it's only supposed to deal with issues re-active or responsively. In other words, they are not supposed to be proactive in their efforts (by whatever means) as that implies they police their networks contrary to what's required of them for continued safe harbor, and what they have said in the past with respect to pulling infringing materials (copyright and other contentious materials).

Google is what Google does it seems. Contradictory. */me SMH.

For content authors, Rock and a hard place as they say.


Notes:
* Monetisation is a scheme through which authors of video content uploaded to YouTube can be financially rewarded through advertisement revenue sharing - authors of videos approved to carry adverts get a (small) percentage of revenue from the advert when clicked for viewing.
0 comments | Write Comment

xx Transatlantic Trade and Investment Partnership (TTIP)

January 30, 2014, 04:37:31 pm by kat
The Transatlantic and Investment Partnership is another legislative behemoth that aims to "remov[e] trade barriers in a wide range of economic sectors to make it easier to buy and sell goods and services between the EU and the US". It affects all sectors of business, trade and investment, and even seeks to have an affect over the Internet and Copyright (why it's being mentioned here).

Additional Reading
TTIP Documents (those made available)
European Commission on the TTIP
UK Government on the TTIP
White House Fact Sheet on the TTIP
White House Press Release
0 comments | Write Comment
Connect to KatsBits on Facebook Follow KatsBits on Twitter Connect to KatsBits on LinkedIn Google+ Subscribed to KatsBits on YouTube
Search KatsBits using StartPage
KatsBits Web