Group: CiteULike-discussion - Forum Thread

So, here's an odd document: http://www.citeulike.org/article-posts/80546 cited 1094 times, but clearly a whole bunch of different documents, and not just one single document at all.

Here's one version of it: http://www.citeulike.org/user/LaurieEMiller/article/80546 where, rather oddly, it's shown as being authored by 4 CUL users: andresmh, mount_misery, marcus_bm, jfr. But in LaurieEMiller's library, this article has the title: "2008 Long-Term Reliability Assessment, version 1.1", which doesn't apprear in the "my publications" list of those CUL users. For example, for andresmh, that document appears as "Designing a website for creative learning" http://www.citeulike.org/user/andresmh/article/80546

Yours, confused, Andrew

That's the remnant of a old bug where we did very bad validation of the URL. e.g., "http://" - box default - "", and "about:blank" were all allowed.

When an new article is posted, one of the first things we do is check to see if the URL is the same as an existing article and match them if it is. Our ONLY definitions of uniqueness are URLs and URL-derived things (DOIs, PubMedIds, and similar) - titles, etc, don't come into it. (It's actually a bit more complicated than that, but let's skip the details for now.)

The bug was fixed a long time ago, but there is still a lot of crud that needs sorting.