Tell Congress to Protect Our Personal InformationLearn More About Unauthorized Access to Data
Mark Simon wrote in Animation World Magazine last week about orphan works. I presume it was meant to be an in depth analysis of current orphan works legislation, but unfortunately he does himself and his readers a disservice because the article was factually shallow and didn't accurately portray anything ever offered up by this or previous Congresses. I was ready to post all the problems with it, but Meredith L. Patterson over at Radio Free Meredith did all the work for me--it's worth a read, so go read it first and then come back. Go ahead, I'll be waiting.
Welcome back. I will add a few notes, though. Simon's post was much like Brad Holland and Cynthia Turner's call to action on the Illustrators' Partnership website. Both pieces are used to scare their constituencies into action, loaded with untruths. From Holland and Turner's post:
We believe the way to speak with one voice is not to submit to a bill that would:
Create uncertainty in commercial markets;
Nullify the exclusive right of copyright and therefore
Reduce the value of your work;
Threaten the privacy protection afforded by current copyright law; and
Invite retaliation from abroad.
Simon's piece refers to this idea of private copyright registries. I know a little about this, because we proposed and advocated for visual registries as a way to help photographers, graphic designers, and textile manufacturers, or as I have come to call them, the people who don't want to be helped.
The idea isn't all that complicated. If you've never searched for a copyrighted work at the Copyright Office's Registry, go try it out here. If the work you're looking for was created before 1978, then you're going to have to fly to Washington, DC and head over to the Copyright Public Records Reading Room. What you'll quickly find is that there are no Google Image-like searches. You pretty much have to know exactly what you're looking for, and even then, the search results are limited to text fields of descriptions of: type of work, title, application title, description, notes, copyright claimant, keyword, registration number, document number, date of creation, authorship on application, and other titles. You don't get a snippet of the work, and if the work is an image, you don't get to see any graphical representation. It's all words.
What if the original copy you have in your hand is all you have? You have very little contextual details for the photo because you bought a shoebox of pics at a garage sale. Sure you could describe it, but you better describe it the precise way the registrant did, or your search results are going to return very little useful information. And even if you think you had a match, there's no way to visually match what you have in your hand with a search result from the Copyright Registry, unless you go there in person. Even then, you might not find anything because of what are called "group registrations." You can read about the 2001 rule change in the Federal Register Notice here, and here's a snippet from it:
The new rule permits a group of photographs taken by the same photographer and published within the same calendar year to be submitted as a group for a single registration. If the claimant does not wish to or cannot identify the specific date of publication of each photograph, a range of publication dates may be stated provided that all of the photographs in the group were first published within three months before the date the application, fee and deposit are received by the Copyright Office. The deposit for the group registration of photographs, or for photographs submitted as unpublished collections pursuant to 37 CFR 202.3(b)(3), may consist of images on CD-ROMs or DVD-ROMs, unmounted prints measuring at least 3 inches by 3 inches, contact sheets, slides with single or multiple images, the photograph in a form in which it has been published (e.g., clippings from newspapers or magazines); photocopies; or a videotape clearly depicting each photograph.
These are essentially contact sheets of bunches of thumbnails that photographers have been allowed to register with the Copyright Office to cut down on their costs. The result is that images can be so small that they can't be accurately matched to a 1:1 scale image you might be trying to identify, and not only that, the description submitted for a group registration may not match that individual photo on a contact sheet. It was a short-sighted change that even Members of Congress and the Registrar of Copyrights acknowledged last month.
So, what is a user to do when the registry fails him? Not use the work, the work remains lost to obscurity, and the creator loses the opportunity to be paid for a use of the work. Progress Clause Epic Fail.
So, we suggested that the Copyright Office make image searches part of its registry. That was a no-go because the C/O said it didn't have the resources or expertise to do it. Sad, because the US PTO does it, fairly sophisticatedly. Okay, well, then let's see if the market can solve this problem. We proposed(PDF) to have companies like Flickr.com, Google, and the Internet Archive setup searchable databases of visual images--ones where a potential user you could scan and upload a photo that he has in hand, and then compare it against their database of images. Visual recognition technology is to a place where you can pick out objects in an image and find those object for sale on the web, surely we can show what whole images match against a database.
These visual registries were proposed to supplement the Copyright Registry, so that visual artists who wanted to be found, could be, in a centralized location. No artist has to register anything to copyright his work, unless he wants to enforce the copyright with some statutory and criminal perks. But no one would be required to register, and no one's copyright would be taken away if they didn't submit their visual art into a visual registry. The art would just remain unfound. That is, unless orphan works legislation passed and a user did a reasonably diligent search and still couldn't find the owner. Then, the work could be exploited, and if the owner resurfaced, he'd be paid for the use of the work.
Orphan works legislation remains unintroduced in the 110th Congress, but we're expecting something any day now. We just had a hearing in March, and there seemed to be a lot of consensus on the critical areas of the policy. Since then, a number of new carve-outs and exceptions have been floated. Some of these exceptions would completely swallow the rule and likely discourage any potential user from bothering to use an orphan. That's unfortunate, because it means people like Simon, Holland, and Turner get to stick their heads back in the sand while orphans languish. We don't want to let that happen.