Proxy Servers

[trlists at clayst.com]

On 4 Mar 2005 Josh Pollak wrote:

> I know at least three major university libraries do this. The Proxy's 
> generally require authentication, and the better ones are configured 
> via an automatic proxy config file that is available at a URL, and 
> tells the browser only to route requests to the library URL's through 
> the proxy. 

Josh I finally got ahold of the proxy config file for the University in 
question.  It is mostly a LONG JavaScript if statement each clause of 
which reads like this example:

	// Journal of Universal Computer Science
		(shExpMatch(url, "*www.jucs.org*")) |

If any of the items match, the university's proxy server is returned.  
If not, "DIRECT" is returned and the proxy is ignored.

My question is, why is the university mediating access to the journal's 
site?  Is the setup that the universities do not host their own copies 
of the material, but must retrieve it from the journal's site and 
authenticate that access for the user?

I still think this is a very "self-centered" approach which does not 
take into account the possibility of either a user who wants to access 
multiple university libraries, or a user who wants to access one of the 
journal sites directly for their own purposes.  It works if your usage 
matches the way the university thinks of the system, but if you don't 
fit their model then it causes trouble -- especially for non-technical 
users.  At least that's how it looks so far.

--
Tom