The purpose of the copyright law is to protect creativity.  Facts are not created, they are discovered, which means facts cannot be copyrighted.  This is especially true for things that were in the public domain to begin with, for example, all government records, including the Federal Census, and county records, such as marriages, wills, deeds, etc.  The logic here is that the taxpayers have paid for the compilation and preservation of these records, so they belong to us, the taxpayers.  The government cannot and does not copyright them — nor can you.

The basic facts of a person's life (e.g., their name; date and place of birth, death, and marriage; and the names of their parents, siblings, spouse, and children) are not copyrightable.  These data cannot be "owned" by anyone.  Most historical documents are also in the public domain, not only because of their age, but because the copyright law was never intended to impede scholarly research — and history belongs to humanity.  (Personal letters, even if historical, are an exception, as are some historical documents under some conditions; the author and the author's heirs are provided a special period of protection under copyright law, though even it expires eventually.)

Just as copyright protection is not intended to impede scholarly research, it is expressly not bestowed as a reward for labor.  I think it bears repeating for emphasis:  the amount of labor invested in creating something has absolutely nothing to do with whether or not it is eligible for copyright protection.  If you choose to expend your time and energy in a non-creative process, such as extracting and distributing records, that's your prerogative, but don't do so with any expectation that the fruits of your labor will be granted copyright protection — in most cases, they won't be.  However, any written analysis of those data is copyrightable. 

You are, of course, free to keep the facts you discover a secret.  That, too, is your prerogative.  But what is the point of doing research if you don't share your results?  You might as well have spent your time on the sofa watching television for all the good you've accomplished.  And most genealogists who hoard their data end up dying before it's published and having their heirs throw their work away — this tragedy has happened twice, that I know of, in my own family.  One of the many empowering aspects of the Internet is that it makes it so easy and inexpensive to publish your work while you're still alive; and, unlike a printed book, you can continue to expand and correct a web site.  An error published in hard copy lives forever.

As for what is copyrightable on my web site, I do own the copyright to any original writing you find here, such as this essay on copyright or my explanations of various aspects of DNA testing.  Any creative writing, such as a family history or a discussion of a pedigree is protected by copyright, as are any photographs I've taken or drawings I've made, and I own a compilation copyright to this web site as a whole. 

Under the concept of "fair use," you may quote my writings (with attribution, of course, because otherwise it's plagiarism — and both accurately and in context because otherwise it's assumed the distortion is intentional), and you may print out individual web pages for your own personal use.  You are not prohibited from creating your own web site, even if we have the same ancestors and use the same data.  We don't "own" our ancestors or the facts associated with their lives.  Similarly, if you have learned something about copyright from this article, you do not need to cite me or quote me whenever you write about copyright.  Like facts, knowledge is not copyrightable.  Only the original writing I use to convey the knowledge is protected by copyright.

The test of whether the amount of material quoted falls under the concept of "fair use" is not hard and fast, but one way to assess the fairness is to compare the size of the quote to the whole of the work from which it is quoted relative to the size of work into which it is placed.  For instance, quoting a sentence from someone else's book into a book you are writing is bound to be safe because one sentence is a tiny part of both works (the original and your own).  But when the amount quoted becomes, for example, a paragraph from an article into a smaller article of your own, you are bound to have crossed the line.  The amount quoted should always be a small (as in tiny) part of the work quoted from and a small part of the work quoted to.

As a rule of thumb for determining what is protected writing, the following is a useful test:

As for the occasional extensive quoting of books and journals you find here at my web site, beyond the concept of fair use, the "drop dead" date on copyright is 1 Jan 1923.  If something was published on or before 31 Dec 1922, it's in the public domain and, therefore, freely usable by anyone, but consider anything published in or after 1923 to be still under copyright protection.  Yes, there are specific exceptions even after 1 Jan 1923, but do you really want to find yourself in court battling an infringement suit?  Unlike criminal law, where the defendant is considered innocent and the burden of proof is on the prosecution, in copyright infringement litigation, the plaintiff is presumed to have copyright protection, so the burden is on the defendant to prove the work was not protected.  Unless you're in the mood to buy your attorney a new yacht, don't risk it.

If you really want to quote a post-1922 work extensively, beyond the concept of fair use, you need to contact the author and get written permission.   If the author is deceased, you will need to locate and contact their heirs for permission.  If you intend publishing your book through an actual publisher — as opposed to self-publishing through a print shop — they will expect you to submit these written permissions along with your manuscript.  So, do gather them as you go, to avoid a frustrating delay in getting your book out.

Unlike many others, I do not make the pointless admonition forbidding commercial use of the data on my web site.  In the first place, the data are already in the public domain — if they weren't, just who gave me permission to use them? — so there's no legal basis to support such an admonition.  In other words, the stricture is totally unenforceable, either in principal or in practice.  In the second place, to quote the famous quip by Gypsy Rose Lee, "How can you sell it when the amateurs are giving it away?"  The way to prevent the commercial exploitation of genealogy is for all of us to freely give away what we have.  It really is just that simple.

DISCLAIMER:  I am not an attorney.  This article does not constitute legal advice.  If you want dependable information on U.S. copyright, visit the source:  the United States Copyright Office, which is the source of the following information.*

See especially these links on What Does Copyright Protect?: 

"Copyright does not protect facts, ideas, systems, or methods of operation..."

"Several categories of material are generally not eligible for federal copyright protection. These include among others...

"Works consisting entirely of information that is common property and containing no original authorship (for example: standard calendars, height and weight charts, tape measures and rulers, and lists or tables taken from public documents or other common sources)."

"§103 - The copyright in a compilation or derivative work extends only to the material contributed by the author of such work, as distinguished from the preexisting material employed in the work, and does not imply any exclusive right in the preexisting material."

"§105 - Copyright protection under this title is not available for any work of the United States Government..."