RE: [NGC4LIB] Observations on ebook readers

Posting to NGC4LIB

Harvey Hahn wrote:

The concern I wish to share is that of altering free public domain text without any obvious or external clues to that effect other than comparing the text with other sources.

A blatant example I recently became aware of is a work by the famous market trader W.D. Gann entitled “The Magic Word”. The 2008 “Revised Edition” published by The Richest Man In Babylon Publishing CHANGED(!) the “magic word”, substituted a different translation for the selected Bible verses contained therein, and omitted significant amounts of original text (and probably altered other text, too). Yet the title and author attribution remain the same (the author died in 1955, so he could NOT have “revised” this work). Someone who did not know of these major alterations might think this is essentially the original work with a few updates here and there, not a whole new work! (This publisher has done similar “editing” and reformatting of other Gann works, to the detriment of the originals, which depend on the exactness of language, pagination, etc.)


You are absolutely right. But it would be a mistake to think that this does not happen with printed materials. It has since day 1. Normal library catalogers deal with these issues much less often than rare book catalogers, and especially, catalogers who work for antiquarian book dealers.

The very concept of “edition” is very different for each group. Actually, I think the discussion on Wikipedia describes it as well as anything I have seen I also suggest the excellent page at for a much more indepth terminology of “edition” in antiquarian terms.

The determination of “who really wrote what” was some of the work the scholars did at the Library of Alexandria, because often, they would have a text that would purport to be by, e.g. Aristotle and they would determine that it was not Aristotle, make up a persona and call it “Pseudo-Aristotle.” Then they had to detail the textual variants and so on, which was all absolutely necessary to determine what Aristotle *really* wrote in a manuscript world.

But of course, it all continued into the world of printing. Early printed books show lots of lots of major variants, but even modern ones do as well. I’ll bet that if you would look at the items in a library’s collection that claim to be “copies” many can look quite different. The layout on the t.p. can be different; the array of dates on the t.p. verso can be quite different, so it would be logical to assume that at least some of the text is different too, but for us, so long as it fits into LCRI 1.0 that it has the same: 245 abc, 250, 260 abc, 300 a, 4xx, it is by definition a copy, even though the text itself may be quite different because we do not compare the text. Rare book dealers’ bread and butter goes much deeper than this, though. Take a look at the points for a Harry Potter book at: to discover how they look for a true 1st edition. And while I may not care, a lot of others do and the difference can be amazing. Here is an example of how much money you can get if you find that real 1st edition! Wow! Maybe I would care after all! Signed editions go for much, much more.

Regular catalogers do not and cannot go into such detail, but we will have to find solutions to the problem you mention in this networked world. Still, there is this basic tension between, in FRBR terms, the work/expression (abstract) and the manifestation/item (physical). Normal catalogers assume that what they see in the 245, 250, 260, 300, 4xx describes the text (i.e. work/expression), and rare book catalogers do not. For example, I may catalog a book that says it is the text of “The Old Man and the Sea” but it starts out with “Call me Ishmael.” As a cataloger, I would not bet very much money at all that the text from one “item” of a book is 100% the same as another “item,” but yet they are considered to be the same “manifestation.” Still, I would bet the house that the 245, 250, 260, etc. information is the same.

In the digital world, we have different possibilities available to describe the text by using automatic word counts, file compares, and other tools which could allow for a much higher degree of exactitude than ever before. Perhaps people will be able to get lists of differences between one text and another, a kind of automatic manuscript collation. Or maybe we’ll say that a difference under 5% means it’s a copy. Maybe we’ll even bring back the idea of the autograph and archetypal copy.

It’s really an interesting time!