Photo metadata software/notes
by Gerald Oskoboiny
I am writing some software to help me manage my online photo collection; in
particular, to make it more accessible, searchable, navigable,
etc. I am using this page to keep track of related stuff (e.g., sample galleries/pages, software, metadata, documentation, my work, misc other stuff.)
To steal ideas from...
- photo.net:
well-evolved, feature-full galleries, forums, ratings, etc.; sample
photo; cool feature: view
folder as text
- pbase.com: another general-use
photo gallery site, very nicely done
- picpix: uses fotobilder,
written by the creator/maintainer? of livejournal
- sample
photo page from Karl, lots of metadata
- nice
gallery of NYC pics, I like the nav links on the left and small
inline pic on the right; maybe do something similar but use
short descs instead of numbers? Also, "preload" feature is
interesting.
- Time's
photo essays have a similar style of navigation, works well
- AliaTerra Camel
Festival: another nice navigation style, my current favorite, I think
- misc other photo/gallery styles to steal ideas from: diveintomark, yahoo, kungfugrippe
- using RDF:
![[Powered by RDF Icon]](http://www.w3.org/RDF/icons/rdf_powered_button.40)
- non-RDF but useful/interesting anyway:
- libexif,
gexif, exif: gexif is a general EXIF photo metadata editor, and
exif is an EXIF dumper. This stuff is much, much better than
most exif-handling software I have seen (listed below).
- metacam:
self-contained app that reads and displays exif info
- makethumbs seems
quite nice; fairly mature, but still quite simple; also has notes on exif
readers
- Eric Johnston's photoweb and exiftags
- AutoScrapbook
- Apache::Album
- album seems
fairly mature, has themes, doesn't seem to do much/anything with
metadata
- Apache::Gallery has nice output
- photopc/epinfo: photopc for
extracting photos from a camera or media card and epinfo for
playing with exif-embedded metadata
- exifdump.py:
python script that does the same thing as epinfo (slightly
different format, but identical content)
- jhead: Exif
Jpeg camera setting parser and thumbnail remover (ouch,
that sounds painful!)
- jhead-xml:
a patch to jhead to output XML (this page also has links to
interesting related stuff)
- jpegtran for
lossless rotation of jpegs
- Gallery seems
like a good software package for maintaining photo collections
(written in PHP)
- David R.
Martin's software includes an albuminfo
file similar to how captivate works (also similar to philg's stuff from
years ago)
- iGal seems nice;
easy to use, up to date, similar approach to mine, though not quite as
buzzword-compliant
- wwwis for adding
height/width attrs to HTML pages
- Linux
Journal discussion of misc gallery software
- imagemagick
tips incl faster
jpeg resizing
- commercial/windows software:
- Pixlabs has digital
photo metadata editing and HTML gallery generation software, looks
fairly good
See the changelog within the captivate script for
info on more recent updates.
- add --get and --set, e.g.
--set dc:coverage "Honolulu, Hawaii";
also accept settings on stdin
- add link elements to each page, to take advantage of prefetching
in mozilla, etc.
- fix bug: "captivate 27/*" no worky (bogus relative hrefs in HTML files)
- take care of @@'d stuff in captivate
- add an output mode where a single image is big and rolling over the
others displays them in the big image spot? (using onmouseover)
- use CSS
for captions
- add style to the RDF so galeon et al. will display it nicely
(samples: W3C RSS feed, Zakim teleconf data)
- add dc:coverage to photo page titles? and maybe yyyy-mm?
- instead of storing cameraShortName in .captivate.rdf, deref the
camera URI and get it from there (and keep a local cache of that page?
or just rely on whatever http proxy/cache is already there?)
- add option to generate indexes in a few different formats, selected
by command line and/or .captivate.rdf
- make captivate more usable (cf. hugo's
comments):
- create $HOME/.captivate.rdf if it doesn't exist
- store user prefs in .captivate.rdf (incl (pointers to?) HTML
headers, footers, templates?)
- add option to regenerate RDF and/or HTML without editing
captions (kinda done; can use "--noshow" and "--noedit" parameters)
- add feature to generate galleries with links to
individual pages (in misc formats? or several formats?)
- add a <link rel=meta> to each generated HTML page pointing to
the RDF behind it
- add a _got a better caption?_ link to each photo page
(mailto for now, form later)
- add access control metadata somewhere, make that part of
the captivating process
- add the ability to delete a photo (and variants) while captivating
- store metadata on photo variants somewhere?
- include height/width tags in generated HTML
- figure out how to manage the relationship between galleries
and individual photos (and make links between them)
- figure out how to delegate most of the work (captioning,
rating, access control?) to friends and other users of the site
- set up something that lets anyone publish photos on my site?
(a la yahoo, msn photo stuff.) Hmm... probably would be too much of a
pain due to people putting up porn sites etc. Might be fun/interesting to
figure out solutions to that though (figure out how to delegate abuse
control to end users, or set up automated monitoring, e.g. if a certain
photo exceeds a certain number of downloads/hour, shut down the whole
site)
- Yahoo photos has a "order prints" feature... should figure out how to
set up something like that
Misc unresolved issues to discuss on www-rdf-interest or somewhere
(review notes from chat with em first):
- some metadata is generic (refers to each variant of the image), but
some refers to one specific version (e.g. exif:ExifImageWidth refers only
to the original.) How to represent that in RDF?
- what's the diff between xmlns URIs ending in hashes and not?
Should other URIs used in RDF end with a hash as well? (e.g., t:camera)
Very early rough notes, probably redundant now:
necessary components of photo page:
short description (title?)
longer description
location
date
size (nn k)
camera info (linked)
link(s) to other version(s)
other optional info:
link(s) to collection(s)
exposure, iso, flash, metering mode, ... info
misc notes:
- when extracting photos, add a minimal comment to the jpeg
immediately after extraction using photopc (before adding to
cvs), including: copyright, and uri of the image?
- script to RDF->HTML: call it decapitate? doesn't really
fit, but that's a cool name. (captivate/decapitate)
@@
from http://www.w3.org/TR/2000/NOTE-photo-rdf-20000928/
The proposed DIG2000 [DIG2000] file format for the (also proposed) JPEG2000
[JPEG2000] image compression algorithm contains an XML-based metadata block
with entries for people, places, events, GPS location, camera type, etc. It
allows extensions with additional entries. The draft of October 1998 doesn't
use RDF.
[DIG2000]
Digital Imaging Group. DIG2000 file format
proposal. Oct 1998. Report (draft) ISO/IEC
JTC1/SG29/WG1 N1017. URL:
http://www.digitalimaging.org/pdf/wg1n1017.pdf
[JPEG2000]
Joint Photographers Expert Group (JPEG). Jpeg 2000 image coding system. 9 Dec 1999.
Report (final committee draft) ISO/IEC CD15444-1:1999. URL:
http://www.jpeg.org/fcd15444-1.zip
Last modified: $Date: 2007/06/14 16:41:40 $
Gerald Oskoboiny, <gerald@impressive.net>
