Gang, I am still fond of the idea to use Google for aggregation of Bundle directories. In my search, I found something called GData;
http://code.google.com/apis/gdata/overview.html At first, it might seem odd, since it revolves around syndication, RSS and the likes, which seems not related. But then is it?? Assume for a second that we use RSS as "Bundle Advertising", we have a simple mechanism for pushing out URLs of available bundles. One could do RSS aggregation and some clients only listen in on that. In my mind it sounds quite neat. I can immediately think of two problems; 1. Number of bundles available will explode beyond what you can practically retrieve in a single RSS "lookup". 2. IIUIC, RSS is very public and you can't have your private, secured feeds. GData from Google could help with both. GData adds a query specification and to me it seems flexible enough. One could throw in the whole Manifest dataset in the snippet as well, and hence make queries for exported packages of certain versions and so on, hence adding to the resolution problem. It also doesn't have, previously discussed, 'delay' of publishing that generic Google searches does, and it allows better control of metadata searched for. WDYT? Cheers Niclas _______________________________________________ general mailing list general@lists.ops4j.org http://lists.ops4j.org/mailman/listinfo/general
