|
INTRO |
Spam Indexing |
Mark Up |
MetaTags |
SGML/XML |
Dublin Core |
PICS |
Metametadata |
![]() |
|
|
|
|
![]() |
|
PICS is a system that supports self-rating or self-indexing of Web material. PICS employs a symbol system to classify Web material according to user selected criteria.
The PICS system can be considered a hybrid. It supports both first- and third-party coding. Not only can the document author apply PICS codes, so can third party intermediaries and end users. The system was first conceived as a filtering software, to preclude objectionable material. Because it is designed to support both first- and third party indexing, it can be used to both "filter-out" as well "filter-in" Web documents according to any pre-established criteria set.
Before PICS imbedded in Web documents can be interpreted, PICS codes or ratings must be implemented on specified servers. Those servers may be either client or host based; which to say, they may be interpreted on a local end user server or on the server of a central PICS rating developer. Once interpreted, one of three actions can occur: (1) the code value and if desired the interpretation is shown (This site is rated quality=1, means great), (2) the site is "admitted," or (3) the site is blocked.
Because PICS can be interpreted on a local server, it is possible for individuals or individual institutions like libraries to develop their own PICS classifications, apply those codes to proxy URLs, interpret the codes on their own system when the "original" URL is requested, and implement the desired action. Thus, in principle, a library might encode Web document proxies, and in so doing apply its own quality and authority control to a set of Web documents. Those documents might then be considered as part of the hybrid library collection. Library servers could then either provide access to non-PICS classified documents with some caveat (perhaps: THESE DOCUMENTS NOT EVALUATED BY ABC LIBRARY -- USE AT YOUR OWN RISK) or block non-rated sites for "in-library" computers, but admit everything on "out-library" systems. In this way, PICS can be used to support the selection policy of any given library.
OCLC and its subscriber libraries now catalog Web sites for library catalogs and OPACs. Central clearing institutions, like OCLC, could provide PICS classifications services based on LC or Dewey classification or any number of available thesauri. That third party provider might also maintain the PICS server. Local library systems could then be configured to select or deselect Web material based upon the central classification scheme consistent with their own selection policies.
In addition, end user systems must support PICS interpretation. End users download rating systems to the client hard drive. Browsers must be configured to interpret both imbedded PICS and to act accordingly. Microsoft Internet Explorer 3.0 browsers and higher support the PICS protocols.
PICS can be applied in one of two methods:
<HEAD>-- see http://vancouver-webpages.com/PICS/HOWTO.html
<TITLE>Extra-Fine Good PICS</TITLE>
<META http-equiv="PICS-Label" content='(PICS-1.1
"http://www.goodratings.org/v1.0" indicates the PICS code service provider
labels on "2000.09.05T09:15-0600" client request labels
for "http://www.greateducation.edu/libschool/mypage.html" where PICS are imbeded
ratings (p 10 q 5 r 2) interpret these values
"http://www.retchedratings.org/v3.5" indicates a second PICS code service provider
labels on "2000.09.05T09:15-0600" client request labels
for "http://www.greateducation.edu/libschool/mypage.html" where PICS are imbeded
ratings (xxx 2 badwords 1)interpret these values)'>
</HEAD>
PICS/DSig Standard Library. Available: http://www.w3.org/PICS/refcode/Parser/Overview.html