Comments on Deep Linking Finding

Dear TAG,

I just looked at your Deep Linking finding
(http://www.w3.org/2001/tag/doc/deeplinking.html).

Unfortunately, it's wrong in terms of character encoding (it says it is
UTF-8, but it actually is iso-8859-1). See
http://www.w3.org/2001/tag/doc/deeplinking.html,validate. I hope this
can be fixed soon.

Also, I think it somehow doesn't really address the core of the
deep linking issue, which is that in most situations, trying to
prohibit deep linking (by whatever means, be it policy or http
referer fields of whatever) is a bad idea.

Currently, the finding mostly is concerned with explaining the
difference between access policies and technical enforcement,
and the difference between identifiability (i.e. having an URI/IRI)
and access rights. While these issues are important for Web architecture
overall, they are only accessory for Deep Linking as such.

This is most obvious in section 5, Deep Linking by Analogy.
The analogies there are okay to show the accessory issues, but
don't get to the main point. The first analogy, of a building
with a main front door and some service entrances, is the wrong
analogy for deep linking; the right building analogy is given by
Jakob Nielsen (Deep Linking is Good Linking,
http://www.useit.com/alertbox/20020303.html), who writes:
"A website is like a house with a million entrances: the front
door is simply one among many ways to get in. A good website will
accommodate visitors who choose alternate routes."

Other good analogies would be that forbidding deep linking is
like a book that in the introduction says "The reader is not
allowed to read this book in any other sequence than page-by-page."
or "If you want to cite this book, you can only refer to it as a whole,
but you are not allowed to reference a chapter, section or paragraph."
(and try to e.g. enforce this using copyright law)


Regards,   Martin.

P.S.: On a related note, http://www.w3.org/TR/webarch/#acks has my name
      ASCIIfied while there are several other non-ASCII characters.


#-#-#  Martin J. Du"rst, Assoc. Professor, Aoyama Gakuin University
#-#-#  http://www.sw.it.aoyama.ac.jp      mailto:duerst@it.aoyama.ac.jp

Received on Monday, 24 July 2006 09:25:04 UTC