Notes on downloading and preparing photographs
(For gphoto2)

gphoto2 does not want dc2xx to be loaded.  Comment out, rename module, etc.
Relevant files: 
	/lib/modules/current/kernel/drivers/usb/dc2xx.o
	/etc/modules.conf
	/etc/hotplug/usb.agent

Plug the USB cable into the computer and the camera.  Turn the camera's
mode switch to "connect".  Turn on the camera.  Console should report USB
initialization ending with "USB device 2 (...) not claimed by any active
driver" (possibly followed by "can't load dc2xx....").  As root, give
yourself permission to read the camera:
    chmod 666 /proc/bus/usb/001/002
(or equivalent with chown.)

Make a directory for the images.  Recommended is ~/images/$topic/orig (the
edited images will end up in ~/images/$topic).  cd to that directory, and 
execute gphoto2.  (This is v2.0.)  Command lines:

gphoto2 --camera "Kodak DC280" --summary
gphoto2 --camera "Kodak DC280" -f /DCIM/100DC280 -L		(list images)
gphoto2 --camera "Kodak DC280" -p 5-10		(get images, origin 1)
gphoto2 --camera "Kodak DC280" -P		(get all images, see below)
gphoto2 --camera "Kodak DC280" --shell
    cd or lcd
    ls
    get (must give exact filename, no glob patterns or indices)
    exit (to exit)

When you get all images (-P), it tries to download some administrative files
which is not user readable, and gets an error message.

The images are downloaded under their names as found on the camera, which
are similar to DCP_0685.JPG, with the camera-generated absolute sequence
number.  Full size images vary in size but 120-150 Kb is typical.  

Now make an index of the images.  It will help to use the format:
filename (tab) description.  Use gimp to review the images and fill in
the descriptions, which you will eventually put in the index.html file.
Then edit a copy of the index into a script of "mv" commands to
organize the images semantically.   The convention is to pick
descriptive words for categories, such as wedding, dance, dinner, cake,
to be followed by a number (padded with 0 if more than 9 images in the
category, so you can sort them). Groups of 4 fit best in the standard
index format.  For example:
	mv -i DCP_0685.JPG wedding-1.jpg

Make a directory "orig" and move all the jpeg files into it.  Then:

cd orig
foreach f (*.jpg)
    cp $f ../${f:r}e.jpg
end

(The ending "e" stands for "edited".)  This gives names like wedding-1e.jpg.

Use gimp to edit the images.  You need to shrink the "e" image to 768x500 (to
fit comfortably in a browser on a 800x600 screen), and then shrink it again to
give a thumbnail image of 180x135 (so four will fit across the screen). The
thumbnail is named by changing "e" to "t".  Do any jiggering after shrinking to
768x500, to save computation.  Procedure:

Main - File - Open (or ^O), double click on "e" file.  It opens in a new window.
(Right click) - Tools - Crop - (or hit the knife thingy in the main panel).
    Drag out a rectangle with the left button; adjust by dragging the corner
    boxes.  Hit "Crop" in the dialog box to do it.
(Right click) - Image - Scale - enter width 768 (pixels) for a wide image
    or 500 for a high one.  (An uncropped image scales to 768x509.)
(Right click) - File - Save (or ^S) to save the image.  Take defaults.
Scale it again to thumbnail size.
(Right click) - File - Save as; click on the "e" name but change to "t".
Close the image window.

In the containing directory is a file ../index.pl.  Copy your notes file
again.  Arrange the content in the desired order, then remove comments and 
edit it to say something like this (take the defaults for -t, -c, -d):

../index.pl \
-h '<td valign=top><a href="%f">%t</a><br>' \
wedding-5e.jpg wedding-5t.jpg \
wedding-3e.jpg wedding-3t.jpg \
etc. etc.

Then: sh filename > index.html

Edit the resulting file adding suitable boilerplate around the content.
See any of the index.html files for copy source.  Within the content area,
the above -h is designed for use in 4-across tables, like this:

<h3 align=center>The Wedding</h3>
<table width="100%">
<col width="25%"><col width="25%"><col width="25%"><col width="25%">
<tr>
(four content units)
<tr>
(more content units -- if you have less than four, fill out with empty cells:)
<td>
<td>
</table>


When finished you can run gphoto2 again and delete the images from the camera.
Command line:

gphoto2 --camera "Kodak DC280" -d 1-85

(for example).  But there's a bug, and it will fail to delete some of the 
files.  Do 

gphoto2 --camera "Kodak DC280" -l

to find out what is still there, and keep deleting until you get rid of them
all.