Hey! I am currently in my Bachelor Thesis about Serendipity in Information retrieval. I built a prototype which accesses unconventional knowledge from the MediaWiki API. Try it and let me know what you think about it! What would you use it for? Can you think of an application for it? Thanks.
I would certainly use it, probably when I'm creating a new talk/presentation/workshop in order to find connections that I might not think of otherwise.
By its very nature I'm assuming it will be somewhat haphazard and "incomplete", because it's not intended to be a "search engine" - you might want to make that clear in the brief opening blurb, otherwise people will unfairly criticise it for not being wikipedia.
Does it always give the same results, or is there a random component underneath? It seems to have a random component (which is good, for my purposes) - can you tell us a little more about the randomness?
But it's an interesting thing - thank you. For how long will this link remain working?
You are right, I received previous feedback which was biased because of the direct connection to Wikipedia. I'll have to promote it more independent.
Currently, it works with a minimum and maximum parameter.
So if the program finds 500 articles in relation to the search term, it picks a random number between the 100th and the 150th article (whereas the first article would be the most related one). I might add some settings with a slider to customize these parameters, but the current values seemed to work fairly good for now.
The link will be working for another month at least, but I can comment the newer one here, once I improved this first version (also in terms of visual aspects).
There might be some value in having a way that people can leave their email addresses with you, given suitable assurances about what you'll do (or not do!) with them. At the very least you could have contact details in your profile here on HN. If someone really likes your work there's always the (admittedly small) chance you'll get contacted with opportunities.
