Besides being a humorous read, and learning how some readers felt that
Robert Thau's blurb should have taken precedence over the blurb by Michael
Dertouzos, the piece also describes how Amazon was able to save the book
from poor marketing by the publisher. While most book sellers had just a
single copy shelved under networking (a catergorization fault), because
Amazon allowed readers to submit reviews (and many submitted glowing ones),
the book was able to garner far more sales than it would have without the
existence of this electronic marketplace.
Thus, the case of Database Backed Web Sites seems to provide at least one
counterpoint to Rohit's theory that e-commerce is all about re-distributing
marketshare in existing marketplaces without substantively growing the
market. Greenspun's book sold more because of the existence of Amazon.
Amazon didn't poach existing sales of the book -- they created new sales of
the book.
Now, it's dangerous to blindy extrapolate from this single case to larger
populations (for example, perhaps sales of Greenspun's book on Amazon came
at the expense of other books on Web publishing, and thus perhaps the
overall market for books on Web publishing stayed the same). But, it seems
that when selling an information product like books, software, music, art,
and many hobbies (e.g., collectibles, like stamp and coin collecting) the
potential for categorization faults is high. As Greenspun documents nicely,
such categorization faults can lead to dramatically lower sales of the
information product than it would have had with proper categorization.
I've experienced this myself with music. As a devote of ambient music, I
find bricks-and-mortar record stores to be just about useless, since they
typically don't differentiate ambient from Pop/Rock, and when they do, they
usually lump it together with techno, acid jazz, and sometimes new age.
Unless I have done research online before hitting the record store, I'm
flying blind when picking out ambient music by bands I haven't heard before.
(Going to http://www.ambient.com doesn't help -- they make computer chips,
such as the MD567X V.90 USB chipset :-) Of course, truth be told, I haven't
spent much time trying to discover the ambient music hobbyist scene, which
undoubtedly exists.
I offer the following alternative to Rohit's e-zero-sum argument. For
information products, there currently exists a marketplace inefficiency due
to miscategorization of products. Online merchants have the potential to
reduce this inefficiency, and thereby increase the total market (while, at
the same time, stealing some marketshare away from existing vendors). Thus,
e-commerce, at least for information products, offers the promise of
expanding markets, rather than simple redistribution of marketshare.
- Jim