Ooooh. That would be worth switching over to me by itself, even forgetting all the other really cool stuff you guys are doing.
Beatrice Otter > Oh, God, *everything* involving search and discovery is something we > very much want to improve -- it's impossible to find anything on the > site, pretty much, if you don't already know what you're looking for. > > Site search has always been one of those things that we wanted to > include on LJ, but were stymied due to the complexity of the database > queries involved -- it was very "expensive" in terms of database > resources, because the LJ database was so freakin' huge. (IIRC, it > was no more than a single order of magnitude less than Google's. And > obviously, without all of Google's resources.) > > Ideally, what we want to do is improve tagging so that it's more > flexible and usable, and what I personally want to do down the road > is migrate the "memories" function (which is horribly, horribly > broken and unusable) into something more like del.icio.us-style > bookmarking, so that you could browse each user's bookmarks or browse > all the things bookmarked on the site -- for instance, if you came to > my journal and looked at my bookmark-memories, you'd see that I had a > memory bookmark named "fic", and you could click on that bookmark tag > and see other people's entries that I'd bookmarked and tagged "fic". > This would be separate from my *journal* tags -- journal tags are a > way to organize the contents of my journal, bookmarks would be a way > to organize other people's entries I wanted to find again. > > My vision for this feature does include being able to see, for > instance, "all of my watchlist's bookmarks by tag" (so I could go and > see everything that people on my reading list had bookmarked), and > then another tab for the second-degree (people watched by people I > watched), and another tab for the whole site. I envision happy hours > of noodling around the site and finding things that way ... > > Basically, site discovery (finding new content that you want to read) > is a hard problem to solve, because it's very resource-intensive, but > I very much want to improve it. I don't envision it being something > we'll be able to do by open launch, but it's one of my major > priorities for the third or fourth quarter that we're open. > > --D > > > > -- > Denise Paolucci > [email protected] > Dreamwidth Studios: Open Source, open expression, open operations. > Coming soon! > > _______________________________________________ > dw-discuss mailing list > [email protected] > http://lists.dwscoalition.org/cgi-bin/mailman/listinfo/dw-discuss > _______________________________________________ dw-discuss mailing list [email protected] http://lists.dwscoalition.org/cgi-bin/mailman/listinfo/dw-discuss
