david said:
>   I spend enough time sitting in front of a computer;
>   I'd love a nice looking printout that
>   I can take and read where ever I want.
>   What end users want varies quite a bit.

ain't that the truth...

anyway, in spite of the 55k monster,
i still forgot to mention some things.

one is that the .pdf contains a feature
that you can look at while you wait for
that 55k message to be "approved"...

the feature is that, when you copy out
the entire text (do select-all, then copy),
you will get text that is _much_ cleaner
than you've come to expect from a .pdf.

first, notice that blank lines were retained!
(for those who don't know, this is _rare_.)

i accomplish this trick with "dummy lines".
specifically, if you delete "::" (double-colon)
-- but not the line-end that follow them --
you'll find blank lines are as in the original.
(the "::" are at the right margin in the .pdf.)

if you then delete all lines with a bracket
-- those will be the page-number lines --
you should find that the file is very close
to what the original .zml file looked like!

(if i remember correctly, the only editing
needed at that point involves hyphenates
acrobat handles badly, but don't quote me.)

again, if you know anything about acrobat,
you'll know that clean copied text is rare.
and getting output that is almost identical
to what you put in originally is remarkable.

the goal is that you can then put this text
back in the zml-viewer from whence it came,
and output another .pdf identical to the first.

and if you then copy the text out of that one,
with some minimal editing (which could be
fully automated),  it'll be identical to the .zml.
and will therefore itself create an identical .pdf.
ad infinitum.  or until computers are outdated.
or you get bored.  or whatever...

i called this "round-tripping" in earlier posts.

try that with the .pdf produced by the .tei,
and see if you get the .tei file out...       :+)

***

you can do the same with the .html version.
copy all the text out of the browser window,
and it will be very close to the .zml version...

my goal is that it should be a 2-minute process
to go from any version of the file to any other...

.zml file to .html file -- 2 minutes, max.
.zml file to .pdf file -- 2 minutes, max.
.html file to .zml file -- 2 minutes, max.
.pdf file to .zml file -- 2 minutes, max.

if i can bring this about, every version of the file
can be considered "close enough" to the master
to _be_ the master.  that's an accomplishment.

-bowerbird