Sorry to come in so late on this Kent. There are already a lot of great replies and insights in this thread, but since you goaded me into it, here's my two cents. First, I don't agree with the "counterfeit furniture is big business" notion. With the prices that good pieces command, no one in their right mind is going to invest in a high end piece without having it examined by an expert. There aren't many craftsmen out there that have what it takes to fool the top conservators and consultants. In the end, it really isn't worth the effort.
Obviously this is an interesting topic with no definitive answer(s). "Measured" drawings can be made in many ways; from photographs, from someone's reproduction, by actually measuring the original piece, a combination of these methods, etc. If the drawing is published in a book or magazine, an art department will take the drawings and redo them to fit into their publishing template. This opens up loads of opportunities to introduce errors. Almost every plan that I've seen in a magazine or book contains at least one discernible error. The trick is to find it before you start building the piece. Other errors can't be found at all. I just heard about a fellow that made a chest of drawers from plans in a book and the feet were 1" shorter than the original piece. The discrepancy would have been obvious had there been an overall height dimension given, but one wasn't. After building the piece, the fellow saw the original and it looked different than his. He finally figured out what was wrong and decided to live with his awkward-looking chest. Had he done additional research or been a more savvy woodworker, he may have noticed something didn't look right before he got to the point of no return. Who was at fault? Certainly not the poor fellow that built the piece. Neither was the craftsman that wrote the book. It was an error introduced by the publisher's art department. If you're working from plans in books or magazine articles, I recommend doing a full scale layout and doing additional research before cutting any wood.
When plans are drawn by, or in conjunction with the craftsman that actually built the piece (Al Breed, Phil Lowe, Gene Landon, Bob Whitley), I think the plans are more reliable. Usually they are laid out full scale which is very handy for checking suspect dimensions. In some cases, they are actually used in classroom situations so any errors are found quickly and corrected. These are the plans that I find most useful and reliable. I do take issue with intentional changes and embellishments that deviate from an original piece and are not called out or noted on a drawing. No one knows what the fate of any plan will be. Perhaps in the future, the plans we draw today will be taken for actual historical documents. I'd hate to see them muddy up the waters for subsequent generations of furniture scholars.
That said, even with an original at hand, I have to make judgment calls constantly when I draw plans. It wouldn't be practical to make separate drawings for each leg of a chair or table. I pick the one that I think is the best and use that as my guide. I pick one side of a case piece and use that as my model. I make a tracing of a back splat, fold it in half and split the difference. I measure drawer sides and come up with an average. I'm working on a desk interior now and each partition and divider was a different thickness (we're talking +/- 1/16"). It's not practical to draw it like that, so concessions have to be made. Usually, I can't take a piece apart to get exact joinery details so sometimes I have to make some educated guesses based on knowledge of the period, the original craftsman's methods of work, regional idiosyncrasies, etc. Often I will consult people who are wiser than me in these matters. When I draw a case piece, I dimension the openings because it makes the most sense. I've been called on the carpet because I don't include dimensions for drawer gaps but I don't know what time of year you're going to build the chest or what the moisture content of your stock is. No matter how talented a craftsman you are, you're still going to introduce your own "errors" to any piece you build, regardless of who's plans you use. In the end, plans won't build a piece for you. They just provide an easier way to create a reasonable facsimile of a piece that you like.
In conclusion, I wonder if we are misusing the word "reproduction." Maybe we should call our pieces "representations" or something else. Our modern methods of work and our contemporary thought processes really don't allow us to work in the same manner that 18th century craftsmen did. For example, it's not efficient to make drawer sides different thicknesses, customers don't want to see tree bark on the inside of case pieces or chair rails, etc. What about Phillips head screws? How many "reproductions" have you seen that incorporate drywall screws? I've seen plenty. At the end of the day, do you have time to reinvent the wheel? Plans are like a road map for a journey that we want to take. Like maps, sometimes we take a detour every now and then, but eventually we arrive at our destination.
Craig