bug-gne
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Bug-gnupedia]Location of data is also an issue


From: Gary Benson
Subject: [Bug-gnupedia]Location of data is also an issue
Date: Wed, 17 Jan 2001 13:24:34 +0000

Hi,

There seems to be no common concensus upon where the data will be
stored. One of the problems with the web is its essentially transient
nature; if the GNUpedia is essentially an indexing system for entries on
external webservers then it will start containing broken links. It could
also be `spammed' in the sense that someone creates an entry which is
approved, accepted and indexed, and then replaces the page with a
redirect to a porn site.

My preference would be for the data to be stored on a central server and
mirrored in a similar manner to Freshmeat - a central (perhaps
firewalled) server which generates dynamic content and mirror sites
which download pages from the central server and serve them statically.

In view of the data format, it would probably be nice to have a
standardised internal format and allow people to submit entries in the
format of their choice: HTML, plain text, some kind of wiki-like markup
language, even dirty old MS word perhaps. Since the software running the
site will (presumably!) be GPLd, it would be a simple matter for people
to submit convertors that run either on the server itself (my
preference) or on the submittor's local machine. In that way, the format
of the data on the server becomes at least partially irrelevant from the
user's point of view.

For these purposes, user-submitted HTML would (IMO) be a particularly
bad choice, considering how fundamentally bad HTML is as a markup
language and how bad most HTML generators are; you want the data on the
server to be compact (so no extraneous <font> tags, no Javascript for
$DEITY's sake) and heirachical (<H1>'s instead of <FONT size=16>'s). A
lack of heirachy would seriously hinder the implementation of searches
and cross-indexing schemes (Aardvark - see also mammals,anteaters...).

Currently the web sucks, and despite the best will in the world this
lowers the overall quality of services provided by things like the Open
Directory. Lets not repeat their mistakes...

Gary
[Reverse me <address@hidden>]



reply via email to

[Prev in Thread] Current Thread [Next in Thread]