Re: Out of Print Classic Computer Arch & Compiler Books

hbaker@netcom.com (Henry Baker) wrote:
>I have several times tried to interest publishers in some sort of a
>'classic hits' of the computer world, with _no_ interest, whatsoever.


The problem is that it isn't economical to create a huge stack of dead
trees and then sell them off a few books at a time. It costs a
significant amount of money to print up a bunch of books using
traditional methods. There are set-up costs that push you into
printing hundreds of copies or more, and paper+ink is not cheap. So,
if you don't sell the books off quickly you've got all that capital
tied up. Also, (from the publisher's point of view) why print an
"old" book for little profit when there are all those authors
clamoring at your door with new books that *might* turn out to be
hits?


On-demand printing might work, but in practice I think it's just too
expensive for a commercial enterprise. You have to spend time getting
the book ready to reproduce, and store it against hopes that more than
one person will actually order it. When was the last time *you*
ordered something from University Microfilms? and how much did it
cost? and what was the print quality? -- a valuable service, but it
has distinct limitations.


And microform distribution just doesn't do it, at least for me.


>[snip]
>I have therefore concluded that the Web is the best place for preserving
>this kind of information.


I agree, and that's why I put my book on the web. On the web it costs
very little out-of-pocket money to publish other than the author's
time (which may be in short supply). And, more importantly, the
"shipping" costs are essentially free worldwide, subject to finding a
reasonable access provider.


But there's still the problem of permanency. I wonder how long it
will be until university libraries take on providing "guaranteed
permanent" web publication sites. Perhaps there is a role for
IEEE/ACM/etc. here.


If you really do want to get the golden oldies published, I can
suggest some alternatives (assuming you have author+publisher
permission, but they don't have the time).


- Get on-line text for the originals, and html-ize it for WWW. But,
this requires on-line text (perhaps rare for books pre-1980?), and is
much easier if you have a copy of the final marked-up galleys with
corrections (I made a copy of mine to keep; did other authors do
this?) It can still be a pain to get crisp images that fit in the
limited number of pixels on a display.


- Demand-printing of photocopies from the originals. It would have to
be a break-even/hobby activity to make the price reasonable (and in
fact Mountain View Press is going to do this on my book).


- Have author grant blanket copying priviledges so anybody can get an
original via inter-library loan, go to a copy shop and get a copy made
(the copy shop should want to see documentation of the copying
permission!).


- Scan the book and put tif/gif images on-line (takes a lot of disk
space, slow to download). But, I think it is Berkeley that did this
with their old tech. report library a while back.


- Scan the book and OCR it (very labor intensive to get 100% accuracy,
correct fonts, etc.). This is, perhaps, what we'd all like to see --
but it is just such a hard job!


I think the fact is that the university library system has
historically done a decent job keeping old books available, if not
conveniently available. But, with budget cutbacks and the glut of
books on the market (resulting in university libraries each buying a
smaller percentage of books and journals that are available), this
traditional means of access preservation is threatened. It is a big
problem, and I hope it gets solved soon.


-- Phil




Phil Koopman -- koopman@cs.cmu.edu -- http://www.cs.cmu.edu/~koopman


--

Return to the comp.compilers page.
Search the comp.compilers archives again.