Making the Intarwebs scientifical

[active_apathy]

This post comes about as a result of a few things, namely:

1. I should update more.
2. Updates have recently been complicated some, by lack of stuff on which to update.
3. Making stuff up can help make an update.
4. Making something int-err-akt-iv can help make made-up stuff fun.

So, what thing could I make up for my readership to assemble and play with?
As a followup to that, what kind of interactive thing could involve the newest members of my readership?
Lastly, what interactive thing could provide material for further posts?

The obvious answer is a poll, but the obvious thing is less interesting than I'd like, and I can't really think of much to poll people on at this very moment. Obviously, this is somewhat problematic.

The solution seems, to me, that it'd be best to open the floor for the communal design of a poll for later posting. The next step, then, is to come up with a topic.

That bit was easy.

[Poll #702203]

For those who haven't had it yet, the EllJay Famous argument is based on a few simple events.

1. Someone suggests I'm EllJay Famous.
2. I disbelieve and therefore disagree. My laundry basket, on the other hand...
3. Argument ensues, typically in a civil manner.

No-one knows how many times this argument's been had, but the outcome is always the same - there isn't one. So, it seems to me that it could be semi-useful to resolve the argument once and for all by:

1. Coming up with a test to determine the EllJay Fame of any given LJer.
2. Finding a way to squish that test into a poll.
3. Posting the poll.
4. Seeing what happens.
5. Optionally, applying the test to determine EllJay Fame in other cases.
   

Just to clarify: by 'test', I mean 'criteria for determining whether a given LJer is EllJay Famous', rather than a clickymemethingy. The test should be fairly simple, and it must be possible to collect the necessary data by way of LJpoll.

Arguably, I could just make up a test all by myself, but this wouldn't really prove anything to other people. I've shown that I like making up wildly unusual and inappropriate explanations of poll statistics, after all. This is where the whole other people bit comes in, so that I have a test which will prove forevermore that I'm right. Unless, of course, it proves something else.

If this is the most ridiculous thing ever on the Intarwebs, do feel free to say so. If not, then everyone's invited for the making-up-a-test bit. And all the other bits too, but they're not happening just yet.

Comments

[aberrantangels]

My personal criteria for EllJay Fame (at least, the most obvious ones) involve being on more than (I'd say) 500 friendslists, being regularly metaquotes'd, and/or having your username recognized even by people who've neither friended you nor been friended by you (I think ginmar and theferrett fit this last point). I'll be able to opine on the relevance of other people's criteria as they post them. And I think it'd be a DSM-IV thing; you don't have to match all the criteria to count as EllJay Famous, just enough've them.

[active-apathy.livejournal.com]

Opining is good, for this is supposed to be collaborative, and collaborative collaboration helps all those attempting to collaborate.

[matgb]

I think "used as an example in the metaquotes userinfo" is as good as it gets.

That and the number of people I click on that have an icon credit pointing this way. Maybe because I tend to click on people from metaquotes but still...

[active-apathy.livejournal.com]

I'm fairly sure that rule's mostly a joke.

[palmer-kun.livejournal.com]

I don't know ginmar but I do know of theferrett.

The fact the latter has made some pretty popular purity tests helps him a lot.

However, active_apathy get's meta'd a LOT more often, and thus is more visible and known in that sphere.

He's more a general celebrity, and she's a niche celebrity.

[eighthcloud.livejournal.com]

... wow.

[active-apathy.livejournal.com]

Wow?

[salzara-tirwen.livejournal.com]

Wow!

[active-apathy.livejournal.com]

$wow

[maggiebloome.livejournal.com]

I figure, you send gophers out with clipboards to the deepest, darkest, remotest icy steppes of livejournal. We skim on our carefully camouflaged hover-cars up to the rawhide tents of various barbarian hordes, being careful not to stampede their horses because no-one's ever friendly when you've just caused them massive amounts of work (I know this from observation, in the wild, of certain people closing other peoples windows when their work is unsaved and nearly finished, grr). Then, utilising our 1337 skillz of Engrish, evocative hand gestures and an ellipse fairy in a cage, we make ourselves understood to the savages. If, upon the mention of the name "active_apathy", they smile and nod and point to their little shrine/postcard/framed metaquotes entry/cave paintings featuring your illustrious presence, then we can confirm that you are, indeed, EllJayFamous.

[jeepcat.livejournal.com]

As polls provide actual numbers on lj (note the numerical results above) they can be used also in finding out if your famous. If people respond in droves to your poll and respond accurately to questions about you and / or your postings, then they know you in some way shape or form that supports the idea of being famous. Follow?
Counters are another good method to an extent. They only let you know how many people have viewed your journal and/or post. Not how many people associate your name with anything on lj.
I do believe one of the first items to be decided upon is to what would be the determining factor of famous. I, personally, want to be the sort of famous where people see my name and associate it, mainly, with something positive. I do not want to be "Paris Hilton famous" where everyone knows your name and that is it. (Sorry but the "female" [for lack of a more public forum word to be used by a lady such as myself] has done nothing but pose nicely for people with cameras and be stupid. To me that is nothing to be famous for - just something to be removed from the genetic pool for!)
Anyhoo... *stepping down from that magic soap box* did I answer the question? What was the question? Who are we talking about here?

[active-apathy.livejournal.com]

People responding in droves to a poll relies on how many know the poll exists. Similarly, cleolinda or iharthdarth are EllJay Famous, but I'd lose at quizzes on m15m or roboclaws.

Counters are a measure of readership, but that one's also flimsy for people who view an entry more than once, or who read them through an RSS feed.

I agree that fame should be positive; Paris, to borrow your example, is 'famous' for a particular kind of slatternly harlotry. That kind of thing qualifies much more as infamy.

You did answer the question, and now I'm kinda poking at your answers so that we can come up with answers that work independently of other factors.

[leila82.livejournal.com]

I saw this linked on theferret's blog, and figured I'd share it with you. There's a site that determines LJ popularity, apparently using Page Rank for the calculations. I'm not really sure how this works, mostly because I entered my own user name and it seemed abnormally high. Also, it's been a long time since I've done any math; I can barely remember what all the symbols mean, at this point. Still, I thought I'd share!

[active-apathy.livejournal.com]

PageRank is an interesting little system, like the trust flow meme that's around the place (or at trustmetrics). I used to know vaguely how it works, but not so much just at the moment; I think a side effect of using it would be that for that meme, if you've been friended by people who are mostly popular, then you will be more popular as a result. If you keep running their thingy over the most popular person on each list, you'll eventually come to someone who's the most popular of their mutual friends.

Oh! Right. That's how it works.

The idea of PageRank is that each and every page on the Intarwebs starts with a certain amount of PageRank, and then shares it out between all the pages it links to. They then repeat the algorithm, plugging in the numbers that just dropped out to see what happens next. Then, they run it again. And again. And again. And so, what they build up is a model of what pages you're 'likely' to get to by just clicking random links.

For this reason, websites optimise their internal links to give important pages a higher PageRank, so that their sites can get more hits from Google.

You can use pretty much the same thing on LJers by treating their friends lists as sets of links - so, each user shares their popularity out between their friends. Rinse, repeat, run algorithm again. Rinse, repeat, run again. Rinse, repeat, run again. Rinse, repeat- I think you see where this is going.

Interestingly, if you're on the flists of a few people of high popularity, on each successive iteration you'll become incredibly more popular. Conversely, if you're on a very long flists owned by unpopular people, then you won't seem very popular at all. Most users will be somewhere between these.

There's various pagerank calculators about on the Intarwebs if you wanted to just play with it for a bit.

Lastly, though, popularity and fame are different things. :)

[leila82.livejournal.com]

Interesting. I was under the assumption that it was basing it more on the number of people who have you friended, than the popularity of each of those people - which actually makes sense. If you're on someone popular's flist, your blog is that much more likely to be encountered by random people, which makes you more popular in turn. Yay understanding!

And you're right about popularity and fame not being the same thing. I also wouldn't say that fame is equivalent to how often your LJ is read. For example, I'd say you're famous on metaquotes, even though not everyone over there has friended your LJ, or reads it. Still, they know of your existence, so you'd have to find some way to include them in your calculations. And then there's the fact that metaquotes has around 7500 users, but LJ has over 9 million journals, so you'd have to decide whether or not being known on metaquotes (or any other given community) is a valid measure of fame. And also what percent would qualify someone as "famous" vs. "not famous" (unless fame is more of a relative thing).

I think I just confused myself.

Update:

[aberrantangels]

warren_ellis has found and taken a test that uses Google PageRank to measure these things scientifically. I leave it to you to discover the results.

[active-apathy.livejournal.com]

See comment above. :)