once again, someone over at distributed proofreaders has
raised the issue of people selling hard-copy of p.g. e-texts
at amazon.com...

so a brand-new newcomer, "exilefromgroggs", posted this:
>   I would suggest an alternative strategy.
>   Set PG up as an Amazon seller (yes, seller!).
>   Charge $1/£1 per book for books that are
>   bought that way, and point out that 100%
>   of the money that PG receives will be ploughed back
>   into the project to make further books available.
>   In the seller's blurb for the book, point out that
>   the books can, in fact, be downloaded for free
>   from PG's website, and include a link. I would imagine
>   that being up-front and open, and being clearly linked
>   to a "good cause" would mean that PG would rapidly
>   become a high-profile seller in its own right through
>   Amazon, and would take the wind out of the sails of
>   people who are trying to make a fast buck for no effort.

because, you understand, that's always the exact way that
those d.p. volunteers picture these amazon resellers -- as
"people who are trying to make a fast buck for no effort."

i fully understand that most of these resellers put as little
effort as possible into churning out the "product" they sell.

but the truth of the matter is that it takes time and money
to turn an e-text into a printed book, and even more work
to handle distribution, and the profits aren't _that_ great...

but nobody bothered to tell that to the brand new newcomer.

because doing so would force the existing volunteers there
to confront the fact that their workflow doesn't have a way
to create a decent print product.  the ascii text is unstyled.
the .html product is formatted for a web-browser, not print.
the e-book products are formless blobs that barely work in
the machines for which they're intended, let alone for print.

that's why the vast majority of the resellers start with the
text-file, instead of the .html file.  so all of the work that
post-processors put into making the .html file "look good"
is just wasted energy.  every one of these html-books is a
"snowflake" -- unique unto itself, per the post-processor --
and it's far too much work to try to figure out each one, so
you could port it to print.  or, for that matter, anything else,
including e-books, which is why those turn out rather badly.
with a library this big, there has to be some standardization,
or else you'll be unable to keep the files updated over time...

of course, it's _easy_ to say that now, with all the experience
we have from having the books _not_ being _convertable_ to
the formats we want.  but some of us said it all _years_ ago,
that letting each post-processor go off in their own direction
was a sure-fire way to make sure their work'd be short-lived.

-bowerbird

p.s.  there is also _another_ newcomer over there who wants to
program an html5 web-app that interacts with the d.p. site, and
nobody is bothering to tell him not to waste his time and energy,
because even if he codes it up, the powers-that-be will ignore it.

p.p.s.  ...waving back to lucy...     :+)