[OpenGuides-Dev] crawling openguides for its RDF

hello, as i may have mentioned i've been working on a nodedb map and local information captive portal splash screen server thing for london. earle's been hacking, and the Rev. Rumble helped swat some bugs, thx! it generates node homepages that look like this: (a simplified version goes on the access point splash screen) http://map.wirelesslondon.info/node/44153 with local information crawled from the semantic web, which for practical current purposes just means openguides. (hopefully http://b.evnt.org/ at some point, too.) i had some niggles... - my spider died a few times because it ate toxic XML. i stuck an eval round each page it swallowed, but it'd be nice to not have to. - the RDF links from the HTML category index pages (e.g. http://london.openguides.org/index.cgi?action=index;index_type=category;ind… are broken (they have null values for the GET parameters) - what happened to http://dev.openguides.org (i get host not found?) if there wasn't a machine-interface HOWTO on there, i would like to add one - generally i would like to think about, and work with earle on making patches to, expose more openguides query type functions - e.g. "find things within N metres of X" - in RDF... maybe we can look at the XML character filtering too. -jo