ExtortionLetterInfo Forums
ELI Forums => Getty Images Letter Forum => Topic started by: Ozzie1 on September 02, 2011, 04:10:34 PM
-
Reason no one has taken up a class action yet is because it "appears" Image Copyright Extortionists (ICS) have not broken any laws. However, the technical community who deals with crawlers, bots and the indexing process knows exactly what the ICS are doing. They, for example, allow the search engine(s) bot(s) to crawl and index their images. However, they don't allow it to include the copyright or watermark properties. So the unsuspecting conduct an image search on a search engine for example of a type of image like houses, babies, etc. The thumbnail images appear in the search engines for example. The unsuspecting don't see a copyright or watermark so they right click it and save the image thinking it is free and in the public domain. Then they upload and use it on their web site. Now, if you look at the thumbnail image properties, you will not see the word "ICS Company Name".."Copyright".."Watermark"... Why? Because it was deliberately omitted, or in other words blocked by the ICS. However, what you will see is a meta-based text ID string that is embedded in the image that "calls" back to an ICS when the ICS bot used to spider the web in search for its images runs across it. Then ICS compiles the list of sites and sends $1200 demand letters where if they just sold the image they generally receive between $10 and $300. ICS could include the copyright text and watermark when crawled by the search engines so those properties displayed, just like they include their meta-based text ID string used to pull a "gotcha"... The cause of action is Unfair & Deceptive Trade Practices for a start..... ;) BTW, we have found the attorneys who are very interested now that we have pointed this out to them......
-
Thanks for an interesting post.
So, what you are saying is that while the trolls don't put any copyright metadata in their images, they do put data in them that their 'bots' can recognize easily?
In addition, are you also stating that the trolls 'seed' these images intentionally in order to potentially cause infringements?
Thanks,
S.G.
-
Yes. That is exactly what they are doing. They could easily include the copyright information, name of troll company, etc. However, they do not. But they do embed the meta-based text ID string so they can find the image later. We feel this incorporates the elements of and violates the Unfair and Deceptive Trade Practice Act of several states. Very nasty for them. If filed, my guess is they would settle very quickly before discovery got too far down the road and exposed them.
-
although this may be plausible, I have my doubts.. In my scenario, the image in question was almost 12 years old (a dinosaur) I doubt there was any data added to this file, being as picscount / other image recognition software didn't even exist back then. Furthermore if an image is edited in photoshop and saved using the "save for web" option all meta-data / EXIF data is stripped out to crunch down the file size as much as possible.
Just my 2 cents worth...and I'm certainly not saying that they have not adopted these practices as of late, anything is possible with them given their tactics and track records.
-
Just a quick note here.
I don't want to 'praise' the stock image industry or anything.
But, Picscout does work pretty well. Even if an infringing image is scanned from printed materials, Picscout can match it to the original photo.
It's said that Picscout can often match even modified or cropped photos to the originals. Some forum participants have experienced this personally, if I recall correctly.
We can probably agree that the laws do not state that the author or owner of an image must put copyright info on or in the meta-data of an image.
The opposite is also true in that it's not illegal to put identifying information into an image.
But, I do think that seeding images around the web, to make images look like 'public domain' would be a cause for action.
We should watch for this, because I do feel that it's happened in the past, and will increase in the near future.
I don't know that the big companies have done this, personally. But, i've heard some things from third parties about a handful of 'photogs' doing this.
But, I can't prove it myself.
On the other side of this, it's best to err on the side of caution, as most everything is owned by somebody. Even if it says that it's "free".
Those are just my opinions.
S.G.
-
Interesting. I would be very keen in seeing a copy of the image you have, if you still have it. The PicScout Bot is crawling and indexing all the images on a site which are binary or compiled. The typical user can't read or modify these files using an application because they are compiled. The binary contains a signature tag that is text which is then compared to the libraries the troll has on file for the image using basic algorithms. This practice has been going at least since 1992 to "tag" code to prove who owns it - I know because we have been doing it since 1995. The early search engines - Lycos, JumpStation, Alta Vista, Excite, Infoseek, Web Crawler, etc. have been crawling with bots since the early 90s. The major troll put its site online in 2000 and was using this signature file along with copyright notification and later, watermarks (See archive.org). Image hijacking began in 1996-97 so it is plausible that the image you have or had contained a signature. One thing you brought up in another post is also very interesting. If trolls are "concealing" and/or "disguising" their bots as user agents, and not announcing themselves, that is computer fraud and is - criminal. We will dig into this next week. Thanks for the feedback.
-
We have confirmed that Picscout is cloaked or mimics a user. Also suspect that this site and some of its users are employees and agents of the Image Trolls. Things will be heating up shortly.
-
It was documented that picscout does conceal itself as a user, and I am glad you were able to confirm this..If this indeed computer fraud / hacking, or something along those lines I sincerely hope things heat up shortly. I have gone to great lengths to try to block picscout from my servers, as far as i am concerned they are stealing my bandwidth, and not following robots.txt which in my eyes is ignoring the no trespassing sign.... It's also been mentioned that there may be trolls among us..