cni-modernization: Re: Pointer Control
Re: Pointer Control
David Walker (DHWalker@uci.edu)
Thu, 30 Jun 1994 10:50:51 -0700
Message-Id: <199406301750.AA21829@mothra.nts.uci.edu>
Date: Thu, 30 Jun 1994 10:50:51 -0700
To: cni-modernization@cni.org
From: David Walker <DHWalker@uci.edu>
Subject: Re: Pointer Control
Dan,
It seems to me that the issue is who should decide if information
owners (authors, publishers, advertisers, etc.) care about who's
referencing their materials. My suggestion is simply to make the
information available to them. Some may care (may even get brownie
points), and many will not. Those who do not care can ignore the
information.
It occurs to me, though, that I'm assuming that the author/owner
is the person who gets the statistics. That, of course, may not be
true. When it's not, I could imagine a number of inappropriate uses
of such information. Perhaps we need to develop a culture in which
usage data belongs to the owners of the information being used?
David Walker
<dhwalker@uci.edu>
At 12:50 PM 6/27/94 -0400, Dan Lester wrote:
>On Sun, 26 Jun 1994 13:42:40 -0400 David Walker said:
>>The issue of knowing what pointers exist to a particular document has
>>always concerned me, if for no other reason than it allows an information
>>provider know who to tell if they reorganize their information (and change
>>pointers). It also strikes me that having *control* over who links to your
>
> The reorganization angle seems to be being taken care of by public
>announcements and the various worms, etc, that search out the info.
>This area can only get better.
>
>>information is not really the issue in most cases; it's usually just a
>>matter of being able to find out who has linked to your information, after
>>the fact.
>
> But why do I care? Do I get some sort of brownie points if I was
>connected to by Harvard, or The University of Tasmania?
>
> This is exactly the problem with traditional library statistics,
>where we've attached great value to the NUMBER of books checked out,
>the NUMBER of reference questions, etc. Yes, these are useful for
>workload measurement and staffing planning. However, checking out
>more books isn't especially better. I'm sure we've ALL checked out
>books, or bought books, that sat unread until they were due, or
>overdue, or forever. Just because someone connected to my server
>doesn't mean that it was useful to them, educational for direct or
>cross reference uses, or anything else. Once again, measuring
>packet count is workload....but means nothing otherwise.
>
>>Protocol designers, how about adding information to request packets
>>indicating the URL of the document that contained the pointer for the
>>current request? That would allow authors to learn who has cited their
>>work.
>
> As a firm believer in not having secrets, I'm in some ways the wrong
>person to be advocating privacy issues, but they may be relevant here.
>Also, it is just that much more data to compute, transmit, tally,
>and produce in some report format. Most of us keep lots of statistics
>that take time and computing, but are totally useless. Why more?
>
>cyclops
>
> Dan Lester Internet: alileste@idbsu.idbsu.edu
> Network Information Coordinator WWW: http://cyclops.idbsu.edu
> Boise State University Library
> Boise, Idaho 83725 In the kingdom of the blind, the
> 208-385-1235 one-eyed man is king. Erasmus. 1523
![[CNI Home Page]](/Images/home.gif)
|
