Filter categories

[KAREN SCHNEIDER]

I would like to address a technical issue Charles Bearden
raised--though most of my post is not really about Charles at all. 

"Many Internet filters have categories to be filtered that 
can easily be checked and unchecked.  Some even distinguish 
between sex ed, art, and "the naughty stuff," so that 
(depending on the quality of the filter) one can choose not 
to restrict most research."

With all due respect (and I respect Charles quite a bit) I would
point out that:

a) There aren't that many Internet filters (perhaps a couple
dozen, including clients, proxies, and packet filters), and
2) Several do have selectable categories, but
3) Preliminary examination--and I do mean preliminary--indicates
that the categories do not always--in some cases, do not
often--match up with what is being blocked.

This is not an "anti-filter" comment, but a quality-product
comment.  If a filter does not have a category called
"homosexuality," but consistently blocks gay-related resources
regardless of the nature of the content, then the product is not
working as advertised.  If a filter claims to block a type of
resource, but does not, then again, the product is not working as
advertised.  Even if we rule out such egregious tools as
Cybersitter (an extremely primitive tool which on the same
webpage announces that it blocks sites related to homosexuality
AND intolerance... ), we are not getting consistent and
predictable results from these products.  When this problem is
compounded by private site/word lists that cannot be viewed or
edited by the administrator or responsible information specialist
(RIS--a new acronym for our time? ;)  ), then we have a problem. 
We are not putting in our libraries what we intend to offer to the
public, and we cannot diagnose or repair the problem easily or
expeditiously.  Diagnosis, said Andrew Abbott in _The System of
Professions_, is part of the definition of what an expert is;
diagnosis of what happens when information systems go wrong is
part of our professional turf as librarians. 

This isn't just a social justice issue (not that this wouldn't be
enough); it wouldn't be fixed if these software developers
addressed these categories.  It's about how librarians practice
collection development in the new century.  If we are going to
use intermediary tools for selecting electronic resources--and
some of us will; maybe some day ALL of us will, I don't
know--then we must learn to evaluate them, critique them, test
them,  communicate with vendors, etc.  Because I am highly
skeptical of most of the current software tools available, I am
sometimes described as biased.  Well, I don't even pretend to be
objective,  but at least I have talked with vendors, and having run
a small business, I understand (and respect) the need to turn a
profit.  These vendors  can only make money on what people are
willing to buy.  I have a lot more patience with vendors, who are
interpreting user needs and market demands as best they
understand them, than with librarians who uncritically implement
tools that directly  impact core library services.  

As librarians, we  are being paid to think, to judge, and  to make
these decisions, and we are accountable for our actions.  As
librarians, furthermore, we have several core  responsibilities;
one of them is protecting the right to read (also known as
intellectual freedom).  No one else cares about this the way we
do.  This does not translate to "use filters" or "don't use filters" or
"ignore porn" or "worry about porn."  This translates to
"understand the problem, and fix it."  We *know* we can't be
passive about the issue of content on the Internet.  How we
respond--PICS, policies, software, positioned terminals, sign-in
sheets, age restrictions, some or all of the above, etc.--will and
should vary; at bottom, though, we should be agreed that these
are not decisions to be entered into lightly, but carefully and
thoughtfully.  

For those who are interested, The Internet Filter Assessment
Project continues, if not apace, then at least with deliberate
forward motion; our only hitch at this point is that a couple of
vendors have been slow in supplying products.  We're moving
toward the active-testing phase later this week, we hope.  Visit
our project at:

http://www.bluehighways.com/tifap/

Karen G. Schneider/schneider.karen at epamail.epa.gov
Contractor, GCI/Director, US EPA Region 2 Library
http://www.epa.gov/Region2/library/
Watch for: A Practical Guide to Internet Filters (Neal Schuman,
1997)