Path: utzoo!utgpu!news-server.csri.toronto.edu!clyde.concordia.ca!uunet!cs.utexas.edu!yale!venus!leichter
From: Leichter-Jerry@CS.YALE.EDU@venus.ycc.yale.edu
Newsgroups: comp.text.tex
Subject: Re: expensive TeX book
Message-ID: <414.26d69db9@venus.ycc.yale.edu>
Date: 25 Aug 90 20:48:09 GMT
References: <8164@jarthur.Claremont.EDU> <1990Aug22.222115.936@nmt.edu> <44813@cornell.UUCP> <1990Aug23.101127.22458@ioe.lon.ac.uk> <584@array.UUCP>
Organization: Yale Computer Center (YCC)
Lines: 86

In article <584@array.UUCP>, colin@array.UUCP (Colin Plumb) writes:
[This response is not only to him, but to many messages on this topic.]
> Does anyone know what makes publishers so expensive and slow?
> I was thinking about how long it should take for a book to go
> from complete edited and formatted manuscript to bookstores,
> and it seemed like it should be doable is 6 weeks.  I budgeted
> 2 weeks for cover art & phototypesetter mastering, 2 weeks for
> printing and binding, and 2 weeks for shipping.  It all seems
> like plenty, although I could understand being a little slower
> if you're running close to capacity.
A printer is faced with a large number of fixed, or very slowly variable,
costs, including maintaining equipment and paying employees.  Any successful
printer will ALWAYS be near capacity at at least some bottleneck point in the
process.  It's a fundamental thing about queues:  As you approach and exceed
the capacity of the system, queue lengths go up rapidly.

> Yet an economics professor told me he's been handed absolute minima of
> 6 months.  He and some friends started publishing themselves so they
> could make annual revisions to their texts during each summer and have
> it ready for the students that autumn.
What they've undoubtedly done is get hold of enough resources to meet their
needs, as exactly as they can measure them.  Partly, I'm sure they are using
such excess capacity as printers they deal with have.  If you are in a posi-
tion to pick and choose printers, you can find one that happens to have capa-
city available just when you need it.  A publisher who deals with a printer
on an on-going basis probably doesn't have this option.

Besides, the larger the organization the slower it is likely to respond.
There are layers of approval, people who HAVE to get their hands on it, pro-
cedures that MUST be followed.  It's easy to say "who needs that", and to
a certain extent that's true.  On the other hand, once an organization gets
large, the only way to maintain any control over it is to have all those
procedures and committees and sign-offs.

> And what's with the price?  I have oceans of photocopies of books that
> are worth $20 to me or so, but not the $60 that's being charged.  I
> believe I can assume that printing a book is cheaper, per copy, than
> photocopying it.  So where does the extra $40 go?  Is editing *that*
> expensive?

It's not cheap.  But you've missed an important point:  While the marginal
cost of a printed book is much lower, you have to first pay large startup
costs.  On a small run, they dominate the total cost of production.  That's
one reason there are still many different printing technologies in use:  They
differ along at least two dimensions, one being quality, the other being their
trade-offs of fixed versus per-unit costs.  Unfortunately, the highest quality
remains associated with either high fixed costs (traditional quality printing
technologies) or high per-unit costs (small-press hand-run operations).

A mainstream publisher also has, as I said, various procedures for dealing
with books.  They all have to be put in the catalogs, stored in the whare-
house, etc.  These impose additional fixed costs.

As a result of this high fixed cost, book prices have an inverted supply/de-
mand curve:  If the demand is low (various specialty books), the only way to
recover the fixed cost is to price the small number of copies that will be
sold very high.

The economics of the publishing industry has for many years been built upon
the assumption that the vast majority of books will make little or no money.
A small fraction become big sellers and carry the costs of the rest.  In
recent years, with mass merchandising techniques and such, this dichotomy has
become more and more pronounced.  Traditional publishers have always viewed it
as a kind of social obligation to publish all those books that didn't make
very much.  Of course, it helped them maintain a presence in the bookstores,
was good public relations, and every once in a while a "nothing" book turned
out to make a fortune - but the old-line publishers really did believe in
this.  As the publishing business has turned into more and more of a business
rather than almost a calling - what with consolidation and the rise of the
mega-publishers - the pressures to make books "pay their own freight" have
gotten heavier.  (Publishing is hardly the only business that has seen this
kind of transition - it's a symptom of the modern age.)

>	      I realize the latency is longer, but does a typical textbook
> where the author does the illustrations in pic or whatever take more than
> a month of an editor's time?
Considering the quality of the typical self-typeset, self-illustrated book I
see out there, I wish editors would spend MORE time on these books.  (Then
again, I see all-too-many poorly-editted books that did NOT come through the
self-publishing route.  And of course there are always the high-quality self-
published books.)

A month is not a very long time for a good editor - who, after all, has a lot
of other things in his queue, too - to turn something around.

							-- Jerry