Reading Wikipedia Randomly So You Don't Have To

February 10, 2013 6:41 AM

Random Article. Bunny Ultramod likes hit "random page" on Wikipedia and sometimes he reports what he finds there. [via mefi projects]
posted by The Whelk (20 comments total)

Meh, I already knew about 3 of those things. (The year and a day for murder in England & Wales, the Nepalese royal massacre and the Glasgow kiss).

Wait, I mean that was 30 pages of awesome stuff to read about, plus confirmation that I'm a trivia king.
posted by ambrosen at 7:01 AM on February 10

I will avail myself of this opportunity to say that "Bunny Ultramod" is one of my favorite MeFi handles.
posted by Fists O'Fury at 7:17 AM on February 10 [1 favorite]

The fact that there was a youth movement in Madagascar where everyone dressed like spaghetti western Clint Eastwood fills me with an unexpected joy.
posted by Halloween Jack at 7:31 AM on February 10 [9 favorites]

Way back in the day--I guess 2005 or so?--I was making the point that as useful as wikipedia was, it was still like 20% Star Wars articles. And I said to my classmate, "Watch--one in five hits is a Star Wars link!" So I started clicking the random link button: first click, a town in Germany. Second click, a civic organization in Illinois. Third click, an article about the 1994 World Cup. Fourth click, some former congressman from the 19th century.

I was starting to get worried, and held my breath as I clicked the random article link for the fifth time. And lo and behold! Star Destroyers! And so he accepted my point.

Since then I've never had a Star Wars article pop up when I've clicked on random page (and I've seen edit wars about reducing the number of pages devoted to various sci-fi stuff), but I like to think that, in some secret corner of the internet it's still Star Wars all the way down.
posted by thecaddy at 7:34 AM on February 10

I often while away free afternoons doing this same think on Wikipedia. Thanks!
posted by sundrop at 7:35 AM on February 10

I find that at least 1 in 5 articles I hit is for a European football club.
posted by sundrop at 7:36 AM on February 10 [3 favorites]

thecaddy - in '08 or so a friend and I conducted an informal study "Longer Wiki pages - Secondary Star Wars Character or 19th Century US Presidents" Our hypothesis was supported by the data.
posted by JPD at 7:46 AM on February 10 [5 favorites]

Ha! I've be doing the same thing last couple days as an excuse to find random sketch subjects: just hit "random" until an article comes up with a headshot. My Dr. K Hari Prasad came out better than my Holden Snyder, Soap Opera Character.
posted by cortex at 7:50 AM on February 10 [1 favorite]

Almost 40 percent of the pages I get are just about train stations.
posted by Bunny Ultramod at 8:01 AM on February 10 [3 favorites]

Perhaps if we limited articles to, say 5 kB of size. We could then write a script that would generate every possible article of that size. Then all we would have to do is go through and delete all the ones that were either nonsense or were untrue. I think the advantage of this approach is that you start with the fullest possible wikipedia, and strip it down for quality as time goes on. The disadvantage is that the Random Article feature is going to be nearly useless for a little while.
posted by adoarns at 8:47 AM on February 10 [1 favorite]

"Le nénuphar a ta femme" ... but what does it really mean?
posted by sammyo at 8:47 AM on February 10

Almost 40 percent of the pages I get are just about train stations.

Maybe train stations are considered automatically notable and deletion-proof, whereas other types of things (people, events, products, companies, books, even public schools) are not. You can't really accuse someone of promoting themselves by writing an article on a train station.
posted by goethean at 9:02 AM on February 10

Nice try Thedore Von Penn Station!
posted by The Whelk at 9:06 AM on February 10 [5 favorites]

"Le nénuphar a ta femme" ... but what does it really mean?
Just random crap someone added to the page. It's gone now.
posted by elgilito at 9:16 AM on February 10

It was the subject of a Vanessa Feltz episode on BBC2.
posted by Bunny Ultramod at 9:19 AM on February 10

Perhaps if we limited articles to, say 5 kB of size. We could then write a script that would generate every possible article of that size.

Not possible^*. Even if we limit the articles to the 26 English letters (ignoring case) plus a space, there are around 3.8 x 10⁷³²⁸ such articles (most of which are gibberish, never mind untrue). The number of particles in the known universe is estimated to be around 10⁸².

^* Well, it's possible, but you can't store all the results of your computation, and it's going to take an awfully long time.
posted by axiom at 11:12 AM on February 10

In case anyone is interested in making their web browsing experience a little more ... random, I can help.

I've long used the Wikipedia random article link as my browser home page, ensuring that whenever I open my browser to do anything, I have a chance of learning something new (DISCLAIMER: this may result in you forgetting what it was that you originally opened your browser to do). Eventually this got a little tiresome, so I decided to encapsulate this effect within a custom script that would direct me to one of a list of links with equal probability. For example, this script here will redirect you to either a random Wikipedia article, OR a random TvTropes page:


var random_pages = [
'http://tvtropes.org/pmwiki/randomitem.php?p=',
'http://en.wikipedia.org/wiki/Special:Random'
];
function jump(index) {
window.location.href = random_pages[index];
}
function select(max) {
return (Math.floor(Math.random()*1357) % max);
}
jump(select(random_pages.length));


If you wish to add links to other pages (e.g. other wikis, random webcomic links, or whatever else suits you), merely edit them into the random_pages array.

If you're unfamiliar with the process of setting your browser's home page to a local file, merely do the following:

1. Paste the above code into your text editor (you must make sure that it is surrounded by script tags, like so, sans spaces: < s c r i p t > ...code goes here... < / s c r i p t >)
2. Save the file as something like 'homepage.html'
3. In your browser's Home Page box, enter the path to your file (on a Windows machine, this might look something like this: file:///C:/homepage.html)

It's a little too complex to post in a comment, but I've also written a python script that will automatically generate the above script with a separate link for a random Wikipedia article from a particular *category* (e.g. maybe you have a preference for 'Software Design Patterns' or 'Logical Fallacies' or 'Cities in Kansas' or 'Rivers of Romania' or what have you). MeMail me and I'll send you it too.
posted by Androgenes at 12:37 PM on February 10 [5 favorites]

Maybe train stations are considered automatically notable and deletion-proof

Not really (although it is almost deletion-proof to be sure) -- it's just that there are a LOT of train stations. New York alone has 468 subway stations in active use -- and roughly 200 that are inactive. On the flip side, any given professional sports team is going to have hundreds, if not thousands, of deletion-proof players throughout its history.

to say that "Bunny Ultramod" is one of my favorite MeFi handles.

To underline the awesomeness, it's only slightly made up.
posted by dhartung at 6:24 PM on February 10

Not possible*. Even if we limit the articles to the 26 English letters

Yeah, I know not practically possible. I'm just enamored with the idea of composition as elision, rather like the way the apocryphal Michelangelo said to just chip away the parts of the stone that didn't look like David.

Carry on.
posted by adoarns at 7:38 PM on February 10

To underline the awesomeness, it's only slightly made up.

It is my rock star name. I have sold almost 13 albums!
posted by Bunny Ultramod at 7:25 AM on February 11