[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Bug-gnupedia] Ideas
From: |
Imran Ghory |
Subject: |
[Bug-gnupedia] Ideas |
Date: |
Thu, 18 Jan 2001 23:15:30 -0000 |
Sorry for the disjointed feel to this posting, it's basically a
collection of notes I've made regarding how I would want to see an
"open" encyclopedia.
Keep old version of articles, perhaps by having an "original" version
and a collection of diffs or just by keeping all altered versions
seperate. This is needed so that someone can reference articles
knowing that they won't change in the future.
How to make it searchable, keywords, google-stlye, dmoz-style ?
Directory style struture like dmoz/yahoo, so each article is given a
location which makes it easier to find related articles. Perhaps this
should be centerally controlled to stop a usenet-name-space like
sprawl.
What formats will it support:
Graphics (PNG/jpg)
Text (for sure)
Text styling (italics, bold, emphasis, etc/alternatively content
markup style and leave it up to the client to decide actual formating)
Equations - This could be important for scientific development
Hieroglyphs, etc. - support by plugin system, specialist areas
could have their own plug-ins which allows viewing of non-standard
stuff, also plug-ins which convert these formats into the standard
format used, perhaps as embedded graphics. Perhaps the same
for equations.
A modular data format, so instead of just having an article you
could have a "module" of articles.
Create a format which allows websites to have encyclopedia
sections, so for instances if I had a website regarding prime
numbers I could make some articles which I can group together
into a encyclopedia module which is on my website. (Sort of like
WAP)
A centralised server (which is mirrored of course) could contain a
link to this module along with a list of articles available in that
module.
We could also have a server which can accept articles directly for
those who do not have a website.
Each article should be uniquely identifiable(not by web url), maybe
by MD5 fingerprint. This fingerprint could be used to link to that
article .
A centralized server would basically be a collection of links to
modules. We should also take backups of the modules so we
won't suffer 404 syndrome.
Apart from a centralised server, other servers could be setup which
have links to a differing set of modules. This would allow for
different "distributions" of the encyclopedia.
Multi-platform support, web, linux, macos, windows (but lets leave
windows until we're well on our feet :-)
Are we going to include:
Bibliographys
Stats
Atlas type info
Timelines
Also regararding quality of articles,
1) By allowing for different "distibutions" we can get peer-review
without censorship. The different distros could all have different
levels of filtering allowing people to use a level of filtering whihc they
find most acceptable.
This will also allow the breaking-off of specialist encyclopedias
more easily(as suggested in the announcement).
2) If we get a fair number of encyclopedia modules tied to websites
we likely to get a higher quality as a bad module will reflect badly
on the website.
Imran Ghory
- [Bug-gnupedia] Ideas,
Imran Ghory <=