As noted in The New York Times, Harper-Collins has put the text of Bruce Judson’s Go It Alone: The Secret to Building a Successful Business on Your Own online; ostensibly this is a pilot for more books to come.
Harper-Collins isn’t doing this out of the goodness of their hearts: it’s an ad-supported project. Every page of the book (it’s paginated in exactly the same way as the print edition) bears five Google ads, a banner ad, and a prominent link to buy the book at Amazon. Visiting Amazon suggests other motives for Harper-Collins’s experiment: new copies are selling for $5.95 and there are no reader reviews of the book, suggesting that, despite what the press would have you believe, Judson’s book hasn’t attracted much attention in print format. Putting it online might not be so much of a brave pilot program as an attempt to staunch the losses for a failed book.
Certainly H-C hasn’t gone to a great deal of trouble to make the project look nice. As mentioned, the pagination is exactly the same as the print version; that means that you get pages like this, which start mid-sentence and end mid-sentence. While this is exactly what print books do, it’s more of a problem on the web: with so much extraneous material around it, it’s more difficult for the reader to remember where they were. It wouldn’t have been that hard to rebreak the book: on page 8, they could have left the first line on the previous page with the paragraph it belongs too while moving the last line to the next page.
It is useful to have a book that can be searched by Google. One suspects, however, that Google would have done a better job with this.
it’s all very strange–i’d think it would be just as much work to replicate the blocks of text on each printed page as to create arbitrary new divisions based on logical breaks, so i assume that represents some kind of active choice.
not a shred of css in the html either. i wonder how all this came about. with the exception of the context-based ads, this whole project could be straight out of the mainstream web of what, 1997?
andrew said:
> It’s all very strange–i’d think it would be
> just as much work to replicate the blocks
> of text on each printed page as to create
> arbitrary new divisions based on logical breaks,
> so i assume that represents some kind of active choice.
i’m not sure what “arbitrary” is doing in that sentence,
but i’d say that the “active choice” was the obvious one
— i.e., to replicate the pagination of the paper-book.
i am of the opinion that that is a wise thing to do,
for a number of reasons, most generally to leverage
the synergy of the powerful combo of online/offline,
and most specifically to avoid the need for scrolling.
however, their design pulls the latter off poorly, since
they couldn’t resist putting a large (and meaningless)
banner for the book at the top of each and every page.
so even if we hide the extraneous menubars and such,
we still often need to scroll to see all of the text and
have access to the previous/next links. sometimes
marketers are unable to resist their worst impulses.
> not a shred of css in the html either.
who needs it?
books are inherently simple creatures. thus,
like a good soundtrack, the mechanical aspects
of an electronic-book should go unnoticed in
most cases, with only the most-sophisticated
observers capable of sensing their subtle effects
while working their magic to enhance the content.
so sure, by all means, if you can suggest some
_benefit_ that might derive from c.s.s., yes! but
a general call for c.s.s. — any c.s.s., just give me
something technoid, please! — seems misguided.
(but again, view-source reveals quickly that the
“designers” of this page — if we can honor them
with the title — were _not_ on the simplicity track.
it’s such a bunch of gobbledygook that — frankly —
i am surprised that it _doesn’t_ contain any c.s.s.!
so my “defense” of them here doesn’t really apply.
still, it’s certainly _possible_ to imagine a page with
simple code that _does_ serve the purpose, and well.)
> i wonder how all this came about.
seems pretty obvious to me. they’re trying to flog
a book that has had its day in the hard-copy world.
what do they have to lose? it is this mentality that
will eventually prove to be _the_ driver for e-books:
the possibility to squeeze dollars from the backlist.
> with the exception of the context-based ads
which, by the way, are the most backward part of this.
people who are reading books do _not_ want to be
bombarded with advertisements, thank you very much.
but as i said, marketers often have the worst impulses.
> this whole project could be straight out of
> the mainstream web of what, 1997?
some of us had conceptualized the cyberspace library by
1981, and we’re still waiting for _that_ version to appear. :+)
-bowerbird
so sure, by all means, if you can suggest some
_benefit_ that might derive from c.s.s., yes! but
a general call for c.s.s. — any c.s.s., just give me something technoid, please! — seems misguided.
well, it wasn’t so much a complaint as an observation based on the html source—no one is making me read this book on the web (yet). i’m just genuinely puzzled by the way the pages were constructed, and the chain of decisions that must have led to it. yes, the plain ascii characters are just as readable as it would have been in 1997 (or 1981), but isn’t it odd that here in 2006, harper collins, which is what? a multi billion dollar german conglomerate? (i get the various big publishers confused) but in any case doubtless containing hundreds of talented print & digital designers in various places within its various tentacles; that this huge company chooses with its first (as stressed by the issuance of a press release) to locate the web design in 1996 or 1997 instead of 2006?
which all leads me to believe that they really probably didn’t have much to do with it at all, aside from the press release.
andrew said:
> well, it wasn’t so much a complaint as an observation
ok. i didn’t mean to over-react. :+)
> to locate the web design in 1996 or 1997 instead of 2006?
i’m very much interested to see what “a 2006 design”
would look like, in your view. do you have a demo?
i’m design-challenged, so i might learn something.
-bowerbird
p.s. i like the “clean and spacious” look of many sites
that are well-regarded by the design community, and
i think user-chosen “skins” are a fantastic approach,
but i am singularly uninspired by most blog designs.
also, most e-book designers have insisted, from the
very beginning, on pagination rather than scrolling.
(and i sure hope bob stein hasn’t changed his mind!)
The lack of CSS as a design oversight has less to do with the actual presentation of the material on-screen, b/c as bowerbird points out, it looks fine except for the ads everywhere. By choosing to use HTML only, and not CSS, H-C is declaring itself way behind the times – 1996/1997 – because the use of CSS, and well structured, well formed documents are one of the major ways that the web has advanced in the last 10 years. Choosing not to make even a hint in that direction seems to be a signal of their ignorance and disregard for the medium.
good point, jesse. thanks for making it.
but i believe that that “advance” is bunk.
oh, i’m all in favor of documents that are
“well-structured and well-formed”, but
my extensive research has indicated that
it is not necessary to have heavy markup
embedded in the text to have documents
that are “well-structured and well-formed”.
i have written programs that are able to
ascertain the “structural elements” of a book
by a simple intelligent parsing of the content.
meaning the costly job of markup is unneeded.
headers, for instance, are rather easy to locate.
if styling is present, as it often is, the styling
can be used to locate headers, as they tend to
be bold, and bigger than most of the other text.
they’re often preceded by a page-break, and so on.
footnotes, on the other hand, are usually smaller
in textsize than most of the other text, and like to
“hide out” at the bottom of a page, paradoxically
making them more conspicuous than you’d think.
indeed, i’ve found that even when you _strip_ the
styling information away from a file, thus creating
something more like a project gutenberg e-text,
it’s still possible to determine the structural aspects.
sometimes the program has to work a little harder,
but it still does a very good job.
yes, i know i’m a heretic as to the prevailing religion
about separation of content from presentation, but
my research indicates that — out in the real world —
presentation is the best thing that _predicts_ content.
in summary, one main job of the typographer is
precisely to translate structure into presentation.
so, by analyzing presentation, we can infer structure.
i know, i know, you’re gonna tell me that’s impossible,
but there’s no need for a “debate” on that topic now,
as i’ll be making a series of examples available soon;
i just thought i’d take the chance to give a heads-up.
-bowerbird