3 ms·
HN's duplicate article detector is broken. This was previously posted here: https://news.ycombinator.com/item?id=17150550 https://news.ycombinator.com/item?id=1
by craftyguy 8y ago
HN's duplicate article detector is broken. This was previously posted here: https://news.ycombinator.com/item?id=17150550 https://news.ycombinator.com/item?id=17150550
The url is identical. I've noticed this happening quite a bit. Here's another example, in this case an article I posted then someone posting the identical article a few hours later:
https://news.ycombinator.com/item?id=17138682 https://news.ycombinator.com/item?id=17138682
https://news.ycombinator.com/item?id=17143429 https://news.ycombinator.com/item?id=17143429
@dang, what gives?
- andrewla 8y agoMy understanding is that if a post did not get a lot of attention, then the duplicate will be allowed through. > Are reposts ok? > If a story has had significant attention in the last year or so, we kill reposts as duplicates. If not, a small number of reposts is ok. > Please don't delete and repost the same story; accounts that do that eventually lose submission privileges. Deletion is intended for things that shouldn't have been submitted in the first place. [1] https://news.ycombinator.com/newsfaq.html https://news.ycombinator.com/newsfaq.html
- craftyguy 8y agoOk, so how do you explain this one then? https://news.ycombinator.com/item?id=17154175 https://news.ycombinator.com/item?id=17154175 https://news.ycombinator.com/item?id=17157649 https://news.ycombinator.com/item?id=17157649 Both posts are ~6 hours apart, each one has a decent number of points and comments, and each one points to the exact same url.
- qznc 8y agoAfaik HN checks its cache for duplicates. This means if the cache is used up by something else (e.g. a long discussion about GDPR) the old submission gets flushed out quite early and a duplicate submission after only a short time is possible.