I know this is somewhat out of topic. Sorry about that. Technical discussions should go elsewhere.
(11/12/28 6:39), Yao Ziyuan wrote:
If Wikipedia can implement the things I talked about so far (topic-specific chat rooms, forums, topic-specific resource announcement and discovery) right on wikipedia.org, then certainly it's the most ideal solution.
No, it's not. The Web with only one site, wikipedia.org, is not interesting at all.
[snip] then we have to implement all this on a separate website or as a separate program ("Wikipedia Explorer").
s/website/websites/ and that's how the Web behaves already. What might be missing is the direction for readers of an entry to find the "right" place to chat/share. Wikipedia currently implements this as a section, namely, "External Links". Indeed a separate program that aids users to find the right link among all the links in the external links section could be useful.
In any case, asking wikipedia.org to become the only place where all the chatting/sharing happen is just not realistic, and that's not ideal either in my homble opinion.
(11/12/28 6:51), Yao Ziyuan wrote:
[snip] So instead I'm planning a desktop-based browser that simply browses wikipedia.org and provides additional features such as ebook creation, creating a FreeNode chat room for the currently browsed Wikipedia article, creating a virtual forum in a "distributed" manner by storing each user's posts on a Blogger.com blog
I think it is always better to find existing places for chatting/sharing. Your program could maintain a community editable mapping from Wikipedia entries to chat rooms/visual forums.
You mention yet another protocol based on HTTP headers. Believe me, there are already tons of protocols like that. Reusing an existing one might give you a better starting use base. Developing your program as a Flock add-on would have the same effect too. (The ideal situation is indeed if there's Wikipedia mirrors as playgrounds for MediaWiki extension.)
(11/12/28 6:08), Yao Ziyuan wrote:
The idea behind this program idea is that I have long felt Wikipedia is not just an encyclopedia; it's the biggest "ontology" ( http://en.wikipedia.org/wiki/Ontology_(information_science) ) ever made on the Web.
Since you mentioned this, if you haven't heard of Semantic MediaWiki(SMW)[1], an MediaWiki extension, already then you might want to bring your idea to that community too. If I recall correctly, Wikipedia couldn't (or hasn't) adopt SMW for performance reason, but there's a starting project called WikiData[2] from the Germany branch of Wikipedia Foundation.
[1] http://semantic-mediawiki.org/ [2] http://meta.wikimedia.org/wiki/Wikidata
Cheers, Kenny
-- W3C HTML5 Chinese Interest Group: http://lists.w3.org/Archives/Public/public-html-ig-zh/