The server was down from 10:32am - 11:15am ET (server time) today for a grand total of 43 minutes. The cause was the same as the 2+ hour outage last September. This embarassingly means I indeed never got around to applying those long-overdue updates! I'll try to get to them soon to avoid this in the future. Sorry for the trouble, all.
Congrats to Myriad  for leading <Escapism> to victory in the January 2014 clan competition and securing a George Washington platinum trophy! Monthly energy has been reset and a new round has begun.
|5||<Peace>||ACE 10 ||
|7||<Reservoir Dogs>||Mr Green ||1,255,476||(x1.81)|
|9||<Berzerker Barbarians>||mrvain ||
|10||<Green Day>||GD02 ||
Starting tonight with the end of January, you'll notice two changes in the usual end-of-month happenings:
- All end-of-month trophies will be awarded automatically
- Clan and bot historical results will be instantly available
In the past, players have had to wait for me to do some manual work before they received certain trophies like the ones based on monthly energy. Clan and bot results also weren't available until this happened. Automating this process eliminates a tedious monthly maintenance task of mine and frees up more of my time for building interesting things like tournaments. Expect to see a few more of these maintenance-killing updates in the coming months (hint: self-serve perm star moves, self-serve logo uploads, and my end-of-month announcement post).
As with all changes, but especially a big one like this, be on the lookout for anything unusual. I've gone to great pains to ensure the end-of-month script will produce the correct output by running it many, many times and diffing it against the expected output, but things can and do go wrong. That being said, I have high hopes that nothing bad will happen. :)
Deleted clans are now styled differently than active clans:
One neat thing about this is that it provides a new perspective when looking at historical results (currently the only place deleted clan names can appear). For example, did you know that of the 84 clans that finished the game's first official month (July 2011), a mere 6 still exist?
As an extra bonus, deleted clans also now have simple profile pages listing their historical results. If you don't remember, until now, navigating to a deleted clan resulted in an error.
Are you wondering why you can't find your old clan? Data was backfilled with historical clan results. This means that if your clan was deleted the same month it was created, it was not backfilled. Going forward, all clan names will be archived.
And finally, why make this change? With tournaments coming soon, there will be an additional place that deleted clan names can appear: historical tournament results. The primary motivation for this change was to clean up how deleted clan names are stored in the database, specifically so that handling them in historical tournament results would be cleaner. The player-visible changes I pointed out are just some now-low-hanging fruit that were made accessible by said changes.
I read about the Internet's 1% rule the other day and found it interesting:
In Internet culture, the 1% rule is a rule of thumb pertaining to participation in an internet community, stating that only 1% of the users of a website actively create new content, while the other 99% of the participants only lurk.
I've heard similar things about sites like Reddit where the vast majority of people just lurk and this got me to thinking: how does this apply to bots4? I did some simple command-line hacking on the server logs to get a rough idea of how many unique people visit the site and how many unique people are active on the game forums.
This command shows how many unique IP addresses visited the site in December 2013:
$ cut -d ' ' -f 1 201312*.log | sort -u | wc -l 6635
And this command shows how many unique IP addresses made POST requests on pages containing the substring "forum" (which is an approximation of how many unique IPs wrote on the forums):
$ grep POST 201312*.log | grep forum | cut -d ' ' -f 1 | sort -u | wc -l 546
This isn't a perfect measurement for a few reasons (for instance, IP addresses don't correspond exactly one-to-one with unique people, but it's a good approximation), but 546 / 6,635 = 8.2%. This is higher than the direct quote above, but if you read the Wikipedia article, it talks about content creators and content contributors, which tend to correspond to 1% and 9%, respectively.
So what's the takeaway? As an active forum member myself, I think it's easy to forget that there's a silent majority that plays this game and doesn't really communicate too broadly. There are a ton of familiar faces, but there's an order of magnitude more of unfamiliar faces. It would probably be beneficial to everyone for me to find ways to more easily solicit opinions from this silent majority, either directly (quick polls) or indirectly (usage metrics).