Thank you Ben. It is annoying, but I'm not blaming anyone at FF directly. - Derrick

Previous occurrences: http://friendfeed.com/paul..., http://friendfeed.com/bgolub..., http://friendfeed.com/mihai..., http://friendfeed.com/paul..., http://friendfeed.com/paul... - Benjamin Golub

I have always ignored the GUID, choosing instead of figure out rss entry uniqueness based on a combination of title, content, author, date published, and, of course, permalink. I found that too many things change to be able to rely on something like a publisher-provided ID. I'm sure GUID works well 99% of the time, but for some reason, I found it a fun problem to just attack entries by the fact that they remain 99% constant over the course of a day (taking edits and permalink changes into account) to determine uniqueness. - Samuel Clay

ah - thanks ben (solid community alerting award)... - mike "glemak" dunn

Samuel (it is hard writing that name, you will always be Rob in my head): It's a tough call. For private things (aggregators, readers) I think it's ok to have a fancy algorithm that tries to be smart. But for FriendFeed I think false positives would annoy users just as much as duplicates. Can you imagine the frustration if your Tweet never appeared but you *know* it's there? The difference is on FriendFeed the feeds users add are generally filled with stuff they make so they know when something is missing. But in a reader (where the feeds are out of the user's control) if you have a false positive and detect a duplicate that isn't really one the user will probably never know. - Benjamin Golub

Ben: I uploaded some stuff to Flickr on Halloween and it never showed up, so you're right -- it is frustrating! - Gabe