Friday, June 28, 2013

Most Guilds Are Small

This finding was a little odd -- most guilds in two games I looked at were small, very small. Many were either as small as possible or smaller than the smallest group size that the code allows. Specifically, the two games were both Sony games, Planetside 2 and EverQuest 2. Sony provides a census for both games, so you can scrape all the guilds for EQ2 but the API for PS2 is different so you can't for outfits (guilds) there, but there was a secondary website which has an unknown but large sample of outfits in PS2.

This is somewhat odd because we generally consider the point of a guild, at least from the point of the developers, is to allow and foster long-term group cohesion among many players so those players can go and do things that you can't do in solo play or small group play. However, a lot of players apparently don't actually use guilds in this way at all. They are using them, I think (at least in EQ2 which I play), for personal satisfaction in a variety of ways. (Personal satisfaction of leveling, design their own guild hall, "I have a guild", a "home" for the player's quiver of alts if not in a larger guild, a guild for their family members...)

Google doesn't have great table support, but I'll put some data here anyways.

For PS2, having an outfit of two players... well I'm not really sure what that does (in terms of the benefit).

PS2 (Capped at 21 here in this table but there are many larger outfits.)

Outfit SizeHow Many
222,315
312,030
47,052
54,416
62,901
72,010
81,398
91,035
10757
11591
12477
13390
14310
15300
16221
17174
18141
19149
20127
21107
There were some much larger outfits, but you see the pattern (remember, this is an unknown and non-random sample).
Here's a graph, capped at 50 on the X-axis.


There were similar results for EQ2, here shown by number of accounts. (PS2 and EQ2 do characters and accounts a little differently.) This, a census not a sample, is the subset of guilds which I determined were "active" during a 4-week period a few months ago when I scraped the census. There are far more guilds registered in EQ2 but most aren't active. Most guilds are small, but most accounts are in a large guild. A "group", as defined in code, well a full group, is 6 people, so guilds with 1-5 people can't fill a full group on their own (although with the recent additions of mercenaries they can, sort of).

Note that the "Accounts" column is non-linear. This is for the following reason: Full group = 6, x2 raid = 12, full raid (x4) = 24. Once it hits 24 I scale it.

AccountsNo. of GuildsTotal Accts.
1-51,3424,205
6-111,1819,140
12-174346,101
18-232354,791
24-4954518,425
50-7425815,629
75-9917114,883
100-14915318,471
150-1998113,794
200-249388,354
250+6939,730

And, a visual image of all that, I think each bar width is 5 characters. (Oops initially I put in the wrong chart, it had characters, not accounts...) Ok I am not sure why there is a discrepancy between the table and the chart with regard to the first two categories... that is.... slightly disturbing, but overall the point still holds.



Both are long tail. But that people are making really small guilds, ones that are unusably small in terms of many types of gameplay, is decidedly of note.

Thursday, June 27, 2013

Academic Journals and Double Billing

So, we academics write articles and send them to journals. Other academics who are on the editorial boards may review them for review, or actually review them. Other academics also review the papers.

On the back end, academic libraries complain about the high prices that publishers charge academic libraries for those same journals.

Let's review who pays for the work of those academics in the first paragraph.

  • Time spent on research for an article, paid for by: University
  • Time spent writing an article, paid for by: University
  • Time spent doing whatever the ed board does, paid for by: University
  • Time spent reviewing articles, paid for by: University
You will note that at no point in this process does a typical journal publisher actually pay for any of the time spent working, it is all paid for by the university where the academic resides (or, perhaps a grant, which is still not the publisher).

I am probably missing a few nuances but, usually this kind of behavior is called double billing, and it is frowned upon.

Wednesday, June 26, 2013

Big Data Divide

About a month ago I was at CeDEM, the Conference for E-Democracy and Open Government, in Krems, Austria. Several projects were focused on communication between local municipalities and the citizenry, with a focus on communication and knowledge sharing that could go either way (such as, a municipality learning what local citizens don't know, and getting that information to them). So a lot of it concerned data collection and analysis.

That should sound familiar, given recent events. I pointed out how there were two sectors that already had the three needed factors:

  1. Data collection capability,
  2. Data (have been collecting),
  3. Data analysis capability.
These two sectors are, of course, intelligence agencies for at least the US and perhaps Britain (and other governments on a smaller scale, most likely), as well as the information companies themselves (in this case I mean information conduit providers, like Google, AT&T, and Verizon, to name a few).

Sure we could pass laws forcing these two sectors to help out local governments provide better services (one eternal promise of not just the internet, but of the information future -- it's always a day away), but I doubt that will happen. That's rather sad, and falls short of the "of the people, by the people, for the people" ideal from US President Lincoln.

Tuesday, June 25, 2013

The NSA and "Unreasonable"

The 4th Amendment to the US Constitution, part of the Bill of Rights (the first ten amendments, as a package, minus two), ratified in 1791.

Perhaps the, or a, pivotal section relies upon the word "unreasonable":

The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated...
Is it unreasonable to have the government have access to data that the giant info-companies already have access to and have in fact generated and searched through themselves? It certainly seems unparalleled, and I don't for the life of me see how it is actually legal, but given that other organizations have been doing this for years for their own good I have a bit of a hard time seeing how it is unreasonable to have some other people do the exactly the same thing as others while looking for terrorists instead of for their own capital gains.

People, especially at Wired Magazine, used to love to name-drop Bentham's panopticon back in the 1990s, but that idea hasn't been used much if at all in the recent dust up. Perhaps it seemed cool until it happened.

Edit: Last week's New Yorker lead article does mention Bentham, but since I was off at ICA in London I only saw it today. The article also astutely points out how we knew the NSA was doing this for years, and knew this years ago.