20 May
2005
20 May
'05
8:27 a.m.
On Fri, 2005-05-20 15:01:39 +0000, Jules Richardson <julesrichardsonuk at
yahoo.co.uk> wrote:
That then means that for TIFF to be a viable archive
option, it needs to
be one image per page - which also means duplicating metadata across all
pages (images) too.
...which I consider to be a good thing, to some extend. Keeping Part N?,
title, year, ... in each and every page just looks like a bug. But OTOH
keeping an image's keywords, printed page number, captions on a per-page
(or per-image, that is) basis is a good thing.
In some ways it'd be handy having metadata in each
image saying what
document it relates to, but duplicating a copy of the index,
description, search keywords or whatever across every image making up a
scanned document seems a bit excessive.
Full ACK. But keeping the index words that apply to this very page
inside the TIFF might make some sense :)
I think I'd rather keep the image data as seperate
TIFF files within a
common directory / archive making up a document, and a single ASCII file
that provides the metadata for that document (TIFF's a pretty darn good
format, but it's let down by some pretty bad tools out there)
A text file would work as well, but I'd place one text file per image
and one for the per-document metadata.
Open question to the list - would PNG files be a
better option than
TIFF? In terms of what I've said above, there's probably not a lot in it
(and TIFF probably has support on classic hardware where PNG doesn't).
But PNG possibly has the advantage that most (and certainly all modern)
web browsers support it, which possibly has implications for a web-
accessible archive...
Wouldn't a web browser also show TIFFs? At least the simple ones?
*check* No. My Firefox would like to display it with the 'display'
program of ImageMagick...
MfG, JBG
--
Jan-Benedict Glaw jbglaw at lug-owl.de . +49-172-7608481 _ O _
"Eine Freie Meinung in einem Freien Kopf | Gegen Zensur | Gegen Krieg _ _ O
fuer einen Freien Staat voll Freier B?rger" | im Internet! | im Irak! O O O
ret = do_actions((curr | FREE_SPEECH) & ~(NEW_COPYRIGHT_LAW | DRM | TCPA));