Book People Archive

page 123

all of this pagenumber confusion extends to several online versions too.

let's take the book "books and culture", written by hamilton wright mabie.
this was google's first public-domain example-book, so let's use it here...

at the university of michigan site, if you ask for page 123, you get page 
117:
>   http://mdp.lib.umich.edu/cgi/m/mdp/pt?seq=123&view=image&id=39015016881628

to get page 123, you need to ask for page 129.   that's nuts.

one of the things that michigan touts as a big advantage of its system
is that its urls are stable, and thus can be used for reference purposes.

but hey, what good is a "stable" url system which is as opaque as this one?

let's say you wanted to reference page 188; quick, what number do you use?
sure, you can figure it out -- if you said "182", though, you would be wrong,
since you need to _add_ 6, not _subtract_ 6 -- but why should you need to
do this little arithmetic test in your head every time you want to talk about
something as basic as a frickin' _pagenumber_?   the answer is, you 
shouldn't.

***

over at google, if you ask for page 123, you actually get page 123.   yippee!
>    http://books.google.com/books?id=0cLAAAAIAAJ&vid=0J3cgt0_qO9ngTEGqZJXft5&dq=%22books+and+culture%22&jtp=123

i wouldn't want to type in that url manually, but at least when it got to
the pagenumber part, i'd be able to instantly translate it for page 188.

but if you download the .pdf of this book, you must ask for page 130
to get to page 123.   crap!   (it looks identical to the michigan version,
but google has a page at the front of the .pdf telling us it's their baby,
which then makes all the pagenumbers get thrown off by one.   icky...)

***

at my demo site, if you ask for page 123 of mabie, you get page 123:
>    http://www.greatamericannovel.com/mabie/mabiep123.html

should be pretty easy for you to do the conversion for page 188.

***

oh, by the way, if you download jose's digital reprint of mabie,
you'll see that the .pdf pagenumbers sync with the originals...
so if you ask for page 123, you really go to page 123.   hurray!
i don't know if that was a lucky accident, or if jose planned it.
might have been just a little bit of both...          :+)

***

the good news is that, on their f.a.q. page, the umichigan folks
acknowledge they have a problem with their pagenumbers, and
invite people to make suggestions on how to solve the problem.
given this open nature, i'm sure they'll get to the solution soon...

***

finally, one more illustration.

for this, i will use jose's einstein .pdf, this time his first version.
to get to page 123 in that, we have to ask acrobat for page 139.
one of many neat things about jose's digital reprints is that if you
click a pagenumber, the scan from google loads in your browser.

then, of course, when you're there at google, you can download
_their_ .pdf of the same book.   ok, let's do that, and check it out.

to get page 123 in the google .pdf, you have to ask for page 144.
so page 123 is page 139 in one .pdf, and it's page 144 in another.

who's on first?   that's what i said.

do you see where all this is going?   one nightmare of the future
is _not_ that we'll have no electronic copies of a specific p-book,
it's that we'll have 23 different electronic versions, with all of 'em
using a varying pagenumbering scheme.   it's just plumb crazy...

all i want to do is point to page 123, for crying out loud!

i said it before, and i'll say it again, if you're replicating a p-book,
then you need to use one -- and only one -- set of pagenumbers,
and it's the set of pagenumbers that that p-book has always had...

-bowerbird

p.s.   by the way, the google o.c.r. on this page, as shown at michigan,
was pretty good.   there are 5 errors on the page, but each of those is
related to an end-line-hyphenate, and thus could be handled easily by
an automatic clean-up tool -- which, unfortunately, hasn't been used...