Okay; so it took me a while to type this...with a errand break in the middle. But since I did...
And we also get back to the weakly-worded conditional language in the proposal. Mind you, I'm no attorney, though I've had a fair amount of involvement over the years in legislation at the U.S. state level. The outcome can sometimes seem quite cumbersome--largely because it's less difficult politically to add clarification to an existing law than it is to rewrite it completely, so some chapters and sections can come to look like an accretive mass of bloviation--but the goal is always to be as unambiguous as possible. If the definition of a thing, an act, or an actor is not clear and unambiguous, the bill will never pass into a calendars hearing, much less onto the governor's desk. If unclear, judicial application of a law is open to interpretation, something we try to avoid.
Perhaps the ambiguity seen in this proposal as well as the GDPR is by the very nature of the European Union. It isn't operating under a constitution as is the U.S., and has no analogy that I'm aware of that delineates state and federal roles as does the Tenth Amendment.
Doug is absolutely correct. WikiTree is not incorporated as a nonprofit entity. It operates that way--probably even has fewer expenses precisely because it isn't incorporated that way; no annual Form 990 and other IRS filings--but there is no definition in the EU proposal as to what is and is not a nonprofit entity, what they refer to as a "non-for-profit." The ambiguity in the proposal would make some state legislators I know go completely speechless...and that's normally unheard of; trust me.
Looking at that part of Article 2, Item 5 that Michael quoted, what exactly is an "information society service"? Have you ever heard that term before? Does "give the public access" mean free, unpaid access; for-fee access; or both or either? Who or what determines "main or one of the main purposes"? Ancestry.com allows me to upload files and text to the profiles in my tree--I probably wouldn't use it if I couldn't do so--but is that Ancestry's "main or one of the main purposes"? If the "information society service" doesn't store the data/information on its own servers and infrastructure, does that make a difference? Since the Internet Wayback Machine has captured over 332 billion webpages that may no longer appear on a Google search, does it qualify as an "information society service"? What constitutes "a large amount of works or other subject-matter"; large in relation to the available archives at my one-name study, or large relative to YouTube? What constitutes "profit-making"? We know WikiTree is not organized as a 501(c)(3) nonprofit corporation but it does not charge users for its services; it does, though, have a revenue stream adequate to keep it operational. What is and isn't profit is defined by accountants and lawyers, not by Webster's...and it differs considerably among various nations.
The term "uploaded content" is critical to the entire proposal. Search the document for "upload"; you'll find 17 instances. Search for "content" and you'll find it used 85 times. Not once is either one of of those words defined, either in context of the proposal or otherwise. To paraphrase Diego Montoya in The Princess Bride, the word "upload" may not mean what you think it means. I'm typing directly into the G2G forum right now, but what I type doesn't just hang out there as text. When I click the button to send this comment on its way, the text is actually uploaded into a database, probably a flavor of SQL, where it will live. Every page we look at in G2G is not a static "page" at all: the content (that word again) is dynamically rendered on-the-fly, extracted from what we previously uploaded to the database and then displayed in a web-browser-friendly format.
And that second paragraph Michael quoted: "Providers of services such as non-for-profit online encyclopaedias, non-for-profit educational and scientific repositories, non-for-profit open source software developing platforms, as well as internet access service providers, online marketplaces and providers of cloud services which allow users, including businesses for their internal purposes, to upload content for their own use shall not be considered online content sharing service providers within the meaning of this Directive."
Where, how, and by whom are organizations classified/qualified by that paragraph? Nothing about the law is ever, "Oh, you know what we mean; everybody knows what that means." WikiTree, for example, is probably not classifiable as an "online encyclopaedia"; probably not as an "educational and scientific repository"; certainly not as an "open source software developing platform" or an "internet access service provider" or an "online marketplace and provider of cloud services."
Doug mentioned the doctrine of "fair use." It isn't as clean as it should be in Section 107 of the U.S. Copyright Act, but at least Copyright.gov provides information helping to define and understand fair use, and an index of searchable case law pertaining to fair use. The EU proposal states that, "Member States shall be free to determine the insubstantial nature of parts of press publications taking into account whether these parts are the expression of the intellectual creation of their authors, or whether these parts are individual words or very short excerpts, or both criteria." Say what?
Too, we have to think about the entire supply chain. If, for example, Ancestry decides it's forced to have a firm, contractual agreement in place for all contributors to Find-a-Grave that allows it to use the original photographs, and it decides it's too difficult to obtain, manage, and maintain the necessary contracts, might it be easiest for Find-a-Grave to just go away? WikiTree lost about a half-million DNA connections in the tree because it couldn't obtain adequate "contracts" under the GDPR for those profiles. WorldFamilies.net shut down completely as of May 25 leaving about 700 DNA projects in limbo because, as a mom-and-pop shop, it decided it couldn't meet the compliance criteria under the GDPR. I posted earlier that large companies spent, on average, about $13 million each chasing GDPR compliance. If this becomes another cash outlay, trust me, there will be cost/benefit analyses of affected services.
Even bigger supply-chain fish, at least one of the articles I linked to indicated that YouTube, that has already spent bundles on its "Content ID" system would not be able to comply with the EU copyright directive. It would have to spend bundles more...or take the EU to court, or stop providing or accepting content in or from the EU. It certainly won't shut down. Through Google Images I can find photos and digital artwork that I can download. Today, legal copyright compliance over what is done with those images is on me, as the person who downloaded them. Under this proposal, the liability shifts to Google to make certain that all rights and permissions have been accounted for.
My rights and permissions experience is admittedly dated, and it pertained almost exclusively to books. But if I were still in publishing in the U.S. and wanted to buy a book, article, or story from you, you and I would have a number different flavors and combinations of rights to negotiate:
- Made for hire
- First North American serial rights
- First serial rights
- Electronic rights
- Reprint rights
- Subsidiary rights
- Exclusive rights
- Worldwide rights
- All rights
And, typically, with some of those you and I would negotiate an applicable timeframe. For example, I could buy exclusive rights, but we could agree to limit that to, say, 36 months. "Made for hire" or "work for hire" is really the only instance that transfers the actual copyright from you to me (P.S., try never to agree to that; if you're a stringer for a news publication, you usually have to accept it).
This level of permissions granularity plays into the EU proposal not at all; in fact, it isn't even acknowledged. It even states outright that the actual copyright law is on a country-by-country basis. My had hurts thinking about how an "information society service" can be expected to monitor and manage a gamut of content rights and permissions adequately enough to ever put anything on the Web again.
Speaking of the Web, one last thought that's been bugging me, and then I'll shut up. What about the Deep Web? (And no, not the "Dark Web," a la The Silk Road; although a subset of the Deep Web, the Dark Web is generally thought of as being for illegal purposes only.) The Worldwide Web as most of us know it--and that is searchable by Google, Bing, and others--represents only about 20% to 25% of the total number of sites and content online. There are self-indexing options sites can use to allow themselves to be found on the Deep Web via the Tor network, but there are no search engines per se.
If they can attempt to legislate only the tip of the iceberg, what exactly is the EU trying to accomplish? Pareto's Law and logic would say that most of what's happening in terms of piracy is happening on the Deep Web. And it's even spelled out at the end of 37a: "In order to ensure the high level of copyright protection and to avoid the possible application of the liability exemption mechanism provided for in this Directive, this Directive should not apply to services the main purpose of which is to engage in or to facilitate copyright piracy."
Hm. "In order to ensure the high level of copyright protection...this Directive should not apply to services the main purpose of which is to engage in or to facilitate copyright piracy."
Okay. I'm going to go do something fun now. Like go clean cobwebs out of my attic....