Bottom Line on Electronic Libraries -Reply

Thomas Dowling tdowling at ohiolink.edu
Thu Oct 3 08:10:16 EDT 1996


> From: Elisabeth Roche <ace at Opus1.COM>
> To: Multiple recipients of list <web4lib at library.berkeley.edu>
> Subject: Re: Bottom Line on Electronic Libraries -Reply
> Date: Thursday, October 03, 1996 12:13 AM
> 
> I saw a message today that said Lycos was going to index links directly
to
> images on WWW pages in a new version being developed. The concern being
> expressed on this list was the possible lack of attribution or control
re:
> misuse or copyright of images.
> 
> I wonder if this is true to begin with, and if it is, I imagine
webwriters
> can do something to make sure owner info is included with the image. This
> would mean -better start embedding such information now? What would be
> reasonable?

There are a number of ways to do this.  First and most obvious is to use
some graphics software to include a visible copyright statement on the
image itself.  Since Netscape never signed on to the <FIG> tag in HTML 3.0,
this is also the most straightforward way to include a short caption, etc. 
CNN does this on their home page, for example.  If establishing your
ownership takes precedence over an unobscured image, you can even put your
copyright statement right across the image.

Second, the GIF format includes space for comments.  I believe that PNG
does also (and TIFF?), but that the JPEG file interchange format does not. 
This would seem like a useful place to store a statement of ownership and
other metadata for indexing, but it has been chronically underused and not
a lot of software seems to support viewing, let alone creating and editing,
those comments.

Third, and most secure (if most obscure also) is the black magic of
steganography, which essentially co-opts an imperceptibly small number of
bits in an image and uses them to carry a message.  Since the message can't
even be seen, it can't be cropped out or otherwise edited, and since it's
actually embedded in the image it should survive even if the image is
copied by screen capture (unlike GIF comments).  This makes it a good tool
for determining that an image originated in a specific collection.


Thomas Dowling
tdowling at ohiolink.edu
Ohio Library and Information Network




More information about the Web4lib mailing list