[Bug-gnupedia] A few technical/librarian points

[<address@hidden>]

Hello


1) Full text searching (a la google) is a very useful tool, but it is 
inherently limited. Two documents on identical topics in different languages 
(English and Chinese) will be found by google to be unrelated.  If we use 
classification schemes, keywords and full text searching, then users can use 
their choice of tool, or the tool most appropriate for to answer their 
information need.

2) Links to external sources. References to source material are a key part of 
scholarship, these cannot be left out. For paper sources we need a complete 
normal bibliographic reference, preferably including the ISBN/ISSN, so we can 
create links to the LoC/a bookstore(s). For electronic references we need the 
URL and a date accessed.

3) We probably need a daemon that checks all outgoing links once a month or so 
and notifies the author/maintainers

4) How to stop cut'n'pasting from the web: It should be relatively simple to 
write a script to select phrases from the article and query 
google/alltheweb/whatever for them. Of course this doesn't catch people 
stealing content from sites using robots.txt to keep the search engines away. 
No system is perfect, so we perhaps need a way to report suspected copyright 
infringment.

5) I personally believe we shouldn't have any explicit removal of articles as 
it will cause too much of a backslash from anti-censorship people, but a UI 
that demotes bad/offensive articles off the screens of most readers. Having 
the possibility for articles to get promoted to the `proper' peer-review 
nupedia sounds good as well.

stuart

--    stuart yeates <address@hidden> aka `loam'
"Oh, havoc," cried Pooh, as he let slip the heffalumps of war.
X-no-archive:yes