This is something we would like to do. However, I am planning on using resources from Yahoo Search, and that is a very big machine and it does things in a very particular way.
I would like to someday have per-url metadata (rather than per-bookmark, which is what you have) which would vary from, say, the official title of the book that's represented by an Amazon URL, or maybe the thumbnail of the photograph on Flickr that is bookmarked (we currently do that one via cheating) and so on. Whether the URL has 404'd or been replaced or dissapeared or is now available only in a cache or whatever could be part of this. But this will also take time to develop, and we are hard at work rewriting everything. After we relaunch I think the pace of improvement will pick up a great deal. Additionally, folks writing their own functionality such as this is very valuable. It generates more ideas, it scratches an itch, and so on. By the time we get there we will have seen several tries around what might work and hasn't, and so on. Joshua > > > The best place to put such a link checker are certainly inside > > > del.icio.us itself. > > > > Actually, I don't think so. > > Anyway, the discussion is purely theoretical because we have no idea > on wether Yahoo will consider this. > > That's why I developed disastrous, because we had no commitment of > Yahoo on this service. > > > Reason: There can be URLs, which are not reachable from the outside, > > and thus such a service, if run "from the outside" would wrongly > > label those URLs as broken, while in reality, they are just fine. > > The opposite is also true: Yahoo crawlers probably have better > connections than my LittleProvider.com machine and so will have less > spurious failures. > > My main concern about tools like disastrous is that popular links, > bookmarked by N persons, will be checked N times... > > > > > Yahoo! Groups Links > > > >
