Re: [deepLinking-25] What to say in defense of principle that deep linking is not an illegal act?

Tim Bray wrote:
> We don't make laws or determine social policies here.  We can send a 
> note to the people who do these things explaining what URIs are intended 
> for, and what mechanisms exist to enforce access control, and explain 
> the (technical) contradictions and problems that result from publishing 
> a URI unprotected by access control or any other server-side policy and 
> trying to keep people from linking to it.  -Tim

So there are two things to accomplish with the statement.  One is to give
an overview to the legal and policy worlds about what deep linking means
within the web architecture.  The other is to prevent technical
contradictions and problems by articulating what they would be.

What does deep linking mean within the architecture of the web?  There is 
no such thing.  Specifically, there is a common misconception that the web 
as you see it through a browser *is* the web.  For example, thinking that 
the click trail you follow because of site design has any kind of 
permanence.

What technical contradictions and problems are there in supporting an idea
of deep linking?  Because of the end to end principle it is not possible
for a server to dictate what a client's best click trail is.  Deep linking
is an example of unintended consequences, where a resource made available
for one purpose becomes useful for other purposes.  It is possible because
of the end to end principle -- you can't prevent deep linking without 
disabling the end to end design of the web.  

Servers that want to control access already have the tools to do it, 
they just don't have the tools to do it based on click trail.

Summary:
* Enforcing click trails is impossible without compromising the end to end
  design of the web.
* It is not necessary to enforce click trails to have access control.
* Writing click trails into law is damaging and unneccessary.

- Lucas

Received on Thursday, 25 July 2002 10:40:42 UTC