Friday, March 28, 2008

Once upon a time, I challenged Jen T to a game of Connect 4 over email. After that game, Jen invited me to a website called to play Connect 4 there. Shortly after the game started, Jen disappeared and never finished the game. I figured that she invited me to the site as a way of saying, "Hey, look at all these other people you could be playing against instead of me." Ah well. In any case, it worked, and I have been playing there for a while. I invited my friend Khaldoun there to play "wordgame," a pseudo-Scrabble game, so that we could attempt to relive the glory days of when we lived together and played Scrabble every other night. If only YourTurnMyTurn had the Buffy boardgame...

Essentially, the website is a place you can play some turn-based boardgames for free online. Unlike some other websites, games aren't played in real-time. You have a couple of days to make your move. You can invite other players to casual games, and can set the turn duration to 1, 3, 7, or 14 days. Or, you can enroll in 28-player tournaments. Each day, you get an allotment of 50 moves, with a max of 150 moves. So, you could sort of play in real-time with someone, if you have enough moves. You can also give them some money to upgrade your account and get unlimited moves per day. There are also some limits in the number of games and tournaments you can play in as a normal user that the "VIPs" get extended, but honestly I haven't seen that it would be worth it. You get to play in plenty enough games and tournaments as it is. The unlimited moves thing is the only thing that might be worth it for me.

There are some things I like about the site. 1) It has a pretty good assortment of games, and it has introduced me to some fun games I would otherwise not play. 2) They are constantly adding new games. 3) Playing against Khaldoun lets me chat with him and talk about all sorts of boring mundane things, which is something we otherwise would not get to do since I am notoriously bad at keeping in touch with people unless they are standing in front of me. 4) I like playing in game tournaments.

The site features the following games: backgammon, battleship, castle danger, checkers, chess, connect four, crack the code (a Mastermind clone), dicegame (a Yahtzee clone), dipole, dots and boxes, draughts, go, go-moku, halma (chinese checkers), holomino, kahuna, lost cities, mill, oware, penguin, pente, pick and pack, reversi, rummy (Rummikub clone), streetsoccer, subulata, tablut, tally ho, and wordgame (the Scrabble clone). Of those, the games that I regularly play that I already knew how to play are backgammon and wordgame, basically. However, I'll play some tournaments for dicegame, dots and boxes, and halma. The games I started playing since I joined the site are castle danger (no, it's not caste danger like I mistakenly told Khaldoun, though I bet that would be an awesome game), holomino, lost cities (despite knowing Mat and knowing of the existence of flexgames, I had never actually played this game before), penguin, pente, streetsoccer, and tally ho. I'll give my impressions of how some of these games are and how well they are implemented in the site in separate posts. I'll leave this one for comments on the site itself.

There are also a number of things I do not like about the website. 1) The tournament format is one of the dumbest things I have ever seen. 2) Related to this, the rating system seems kind of screwy. 3) The admins are very resistant to change. 4) The admins seem to prefer trying to dumb down games and giving advantage to poorer players rather than giving games depth and providing more options for all players. 5) Some of the games just don't translate well to the format provided by the website. I'll comment about this last point in posts about the particular games.

First, the tournament system. No, actually, I'll talk about the rating system, since the problems with that add to the problems with the tournaments. I'll say that the rating system isn't terrible, by any means, or at least not really worse than rating systems found elsewhere. I'd say it does a decent job of eventually differentiating player skill level. However, that only happens if 1) you play a number of games, and 2) you play a variety of opponents. For example, Khaldoun and I played each other in wordgame a bunch of times, but our ratings didn't really go anywhere because we play so evenly. Our ratings only went up once I started to play in tournaments (and later when Khaldoun did). Essentially, I would win games against other people, then siphon half the rating points to Khaldoun, so that we would slowly move up in the ratings.

Other than that, there are some illogicalities about the rating system. For example, if you win a game, it is possible to lose rating points if your opponent is much lower in rating than you. I can understand not wanting to reward players for picking out weaker opponents and crushing them to gain rating points. But, losing points when you won seems silly. I think you could easily have a system that scales the reward to the difference in skill level, but still reward players for winning (I know, that's a crazy idea!). Giving low enough rewards for drastic mismatches should effectively discourage those hunting the newbies for rating points.

Also, if two players start out and play two games, splitting wins, they would not end up at the same rating. So, that tells me there is something weird about the math.

Another thing I've noticed is that often when a new game comes out, there will be a tournament started right away for it. So, everyone is at the same skill level. But, inevitably I'll see some guys ahead of me in rating by the end of it even though I have yet to lose a single game and they have lost games. Maybe it would make sense if I sat through and worked through the math. But, it just seems like something is funny. In the end, the ratings aren't that big of a deal, though, especially knowing that the system is wacky. What really bothers me is how it affects tournaments.

The tournaments on start out with 28 players split into 4 groups. Each player within a group plays each other player in the group. The top two from each group then go on to the quarterfinals. Players receive two points for each win, 1 point for a draw, and 0 points for a loss. If there are ties in points by the end of group play, then players are sorted by a calculated variable called Tournament Performance Rating (TPR). And, here is where the problems with the rating system start to taint the tournaments. The TPR is primarily based on player ratings, specifically player ratings at the beginning of the tournament. It also takes into consideration the ratings of all the opponents you faced, but it is primarily based on player ratings. So, if there is a tie in points at the end of the group, someone with the lower rating will have a higher TPR than someone with a higher rating, and will therefore advance. That's fair enough, I suppose, giving a sort of handicap for the players. However, that assumes that the rating system upon which it is based is valid, which it doesn't appear to be.

There are also problems with this approach. 1) Players with provisional ratings, by nature, get sorted incorrectly since they haven't played enough games against a variety of opponents to receive a more accurate rating. 2) Since it only considers ratings at the beginning of tournaments, it cannot handle any changes in skill over time. This wouldn't be a problem if it was a weekend tournament. But, we're talking weeks and months here. I am in three tournaments of tally ho that are predicted to last around a year each (they are the first and last three I will play). Surely player skill will change over the course of a year. Current player rating is taken into consideration, but only as a second tiebreaker after TPR.

As for the particular TPR formula, it seems to be a little screwy as well. I think it too heavily weighs ratings, both one's own and the opponents'. This primarily becomes apparent once you reach the quarterfinal stage. In the quarterfinals, semifinals, and finals, you play 4 games. Not 3, or 5 or some odd number that would make more sense, but 4 games. If there is a tie, then TPR again decides who advances. So, the person with the higher TPR only has to win 2 of the 4 games, whereas the person with the lower TPR has to win 3 out of 4 games, to advance to the next stage. It doesn't matter how close in skill level you and your opponent might be. As I said, I think the TPR formula weighs ratings too heavily. So, you can find yourself in situations where you win all of your group matches, yet still have to win three games in the quarterfinals versus someone who lost games in the group matches, simply because you have a higher rating. I am in a wordgame tournament right now where I am in a quarterfinal matchup with someone with the beginning provisional rating. This person lost 2 of the 6 group games, and I went undefeated. But, I still have the three-win burden because I have a high rating and this guy (who is a good player, but just hasn't been able to be "sorted" yet) only has to win 2.

I think the intention is nice: They wanted to institute a handicap system to give "the little guy" a boost. But, it doesn't really play out that way. Basically, with their system, they are saying, "We don't want the A students to do the best on the test. We want the B students to." They would really like the C or D students to have the best chance, but it's really the B students who do. So, the best player in the tournament might not be the most likely to win it. But, the second-, or fourth-, or whatever-best player does. In skill mismatches, the better player is still going to be able to win 3 out of 4 games. It's in the close matchups where the handicap comes into play. And this situation is more likely to happen the farther you get in the tournament. So, if Khaldoun and I were to enter a tournament, and ended up facing each other in the semis or finals, one of us would have a huge advantage in our chances to advance, even though we are about as closely matched in skill as possible (out of 40 games, I've won 22). If it were a handicap system that somehow scaled to the differences in skill level, like a handicap in a golf tournament, I'd be fine with it. But, you don't have enough games for an all-or-nothing handicap like they have to be appropriate. For example, in golf, the handicap is in number of strokes, and through 18 holes, a handicap of a few strokes would work fine. But, in the tournaments on YourTurnMyTurn, it's just 4 games, and it doesn't matter if you have the equivalent of a 1 stroke handicap or a 100 stroke handicap.

I, and a couple of other players, have tried to get the format changed to an odd-number format. But, the site admins are very resistant to change, and they are also biased towards the weaker players. They have attempted to justify the 4-game system with the fact that players will get an advantage depending on what color they play/whether they go first or not. Therefore, in their argument, you need an even numbered system. However, I think they give way too much emphasis to this point. By the win statistics they keep on their site, there doesn't look like there is a whole lot of advantage for any given color for any given game. Of course, there might be some significant advantage if you actually did some tests. But, it is certainly doesn't look to be true for all, or most, of the games. And, I would argue that the TPR advantage far outweighs any advantage color might bring, particularly the later on in tournaments you go, since presumably the opponents are more closely matched. I think you could just go with a 5 (or 3) game format, and just give the color advantage to the person with the lower TPR, if you want to have a handicap of sorts. But, having an even numbered system takes tiebreaker scenarios out of the hands of the players and puts it into the hands of some shady math. That's what really bugs me about it.

Part of the admins' unwillingness to change is wanting to keep a simple system for all games. If they don't want to do the programming to change things, then there's nothing anybody can really do about it, other than find another site or make your own. It's their site, after all, and they probably won't do anything about it unless a lot of people complain. But, too many either just don't care, or might enjoy the (unfair) benefits the handicap system affords them. And some people think that those of us who are trying to get things changed for the better are just doing it because we are being greedy for wins (since people who actually think about these things and who are being affected the most by the system are those who are generally highly ranked).

The admins' desire to keep the system consistent across games and bias towards lower-ranked players spills into other areas that I think could be improved. I'll comment more on that in posts on the particular games. But, one example is backgammon. In backgammon, normally you play matches for a number of points, and in any given game, you can win a different number of points, depending on how the game plays out. Normally you get 1 point for a win. But, if you manage to get all of your checkers off the board before your opponent has a chance to get any of his off, you win a gammon, which is worth 2 points. If you manage to get all your guys off before your opponent can get any off, and your opponent has some men left in your home board (the last six spaces you are exiting from), then you win a backgammon, which is worth 3 points. Added to this is the doubling cube. At any time in a game, a player can challenge his opponent by doubling the stakes. The person challenged has to decide whether to accept the double, or to forfeit the game at the current stakes. If the person accepts, then they control the cube, and may later offer the same doubling challenge to their opponent, who must then make the same decision: accept the doubled stakes or forfeit at the current stakes. Anyways, all of this adds a ton of strategic depth to the game and is pretty standard on any site that takes backgammon semi-seriously. But, on YourTurnMyTurn, despite backgammon being played three times as much as any other game on the site, every game is a one-point match, and they have no intention of including this feature because it would require them to have a different system for different games. Fine, whatever. But, the game is called backgammon, yet you are unable to actually play for a backgammon. And not having those features really takes more than half the strategy out of the game, and rewards luck a lot more than skill. But, the admins don't really care much for rewarding skilled play, by their own admission, because it gives advanced players more of a benefit. Umm, yes. That's what competitive games are about, right?

Ah well. I'm still playing. If I catch whiff of some other site offering a similar selection of turn-based games that aren't played in real time, and that includes a more rational tournament format, then I'll be there in a heartbeat. In the meantime, I'll be banging my face against the computer whenever I look at my TPR...

Feel free to join up at the site and play me in some games. My handle is monkeyswinkle, and I'll play pretty much anything.

Rating Commercials

I want to be able to rate commercials. I want to be able to say, "Hey, Bud Light, your Dude Madness commercials are great" or "UPS, that long-haired drawing guy has worn out his welcome." I also want to be able to have some influence on how frequently I see these commercials. What I envision is being able to use my remote control to give a rating of some sort, like a 5 star rating scale. Then, somehow the TV station will adjust the frequency of me seeing the different commercials. Even if I give a commercial the lowest rating, it doesn't have to mean that I will never see it, but just infrequently. Also, each time a commercial comes up that I have already rated, it will show me what rating I have for it, and I can choose to keep it there or adjust it. I think it would give some good feedback to the companies to. I imagine they would want to know whether their ad is well liked, or if it is annoying or angering people.

Ok, maybe it might not be feasible yet to do it with TV. But, I think you could certainly do it with online feeds. We're beginning to be able to watch a lot of stuff over the internet. Yet, we still have to deal with commercials. What's worse is that there is a smaller selection of commercials that businesses are putting out on the internet. So, you're stuck watching the same commercial or two every break. For example, I've been watching some NCAA games through CBS sports online. But, honestly, every commercial break has the same two commercials every time. I want to be able to tell them to cut that shit out. I'd rather watch black emptiness than see the same commercials over and over. I know the technology exists, people. Make it happen! And somebody get to work on that TV idea too. Actually, add in the ability to rate everything I see on TV. My wisdom needs to be heard. "Billy Packer, you suck. You get only 1 star from me for your basketball commentating."

Thursday, March 27, 2008

Sightseeing in Midtown

I told a couple people about this. But, I thought it was humorous enough to post here, as well, for those who I might not have told.

A few weeks ago, I drove downtown to my friend Kevin's apartment. He lives on Peachtree between 7th and 8th Streets. Here's a map of the area, so you can follow along. I was driving around looking for parking. I was driving east on 8th Street, coming up to the stop sign at Cypress Drive, when I saw this dude standing on the side of the road in sweatpants, a T-shirt, and a hoodie. Actually, there were a couple of people walking up and down the street, but this guy drew my attention because he reached down to his side, as if he was scratching his hip or tucking his shirt in. But, then, just as I was driving by, he hooked his thumb into his sweatpants and nonchalantly pulled down the front of his pants to show off his goods. I would call it "flashing" except that that might imply some sort of quick, "Hey, look at my weiner!" kind of thing. I guess "exposed" would be a better word, because it was much more slow and deliberate, like "Hey, dude, can I get your opinion on something?". The funny thing is that I kept having to drive around the block to look for a parking spot, and passed him by about two other times. One time, I stopped to let him cross the street in front of me. He waved politely - actually, he flashed a peace-sign to me - in thanks. So, at least he was a nice flasher.

Anyways, I did a little Googling, and supposedly Cypress Street has been somewhat notorious as a place of business for male prostitutes in the past. Supposedly it's been getting better in the last few years, because of police stings and such. They cleaned out some disreputable businesses around there and now there are some nice places like Ecco around there. But, that makes a bit more sense why some random dude would show me his schlong, since it certainly looked like I was cruising. I guess he was just displaying the goods to see if I was interested in buying. It also might explain why it is usually so easy for me to find parking on Cypress and 8th Street.

According to this article, and some other stuff I read, Cypress Street is where the male hustlers do business, Ponce is where the "crack hookers" are (and there's an "open air drug market" around Ponce, Parkway Dr, and Monroe/Boulevard), and the "transvestitutes" do their thing south of 5th Street around Myrtle (a couple of streets east of Peachtree). Also, Atlanta is supposedly one of the prime spots for pedophiles to go to have sex with underage prostitutes. I guess the laws for going overseas for sex with teens and children are much more severe than laws for having sex with American teens and children. Plus, Atlanta as a major airport hub makes it easier for people to fly into town, do their business, and get out of town to go back to the wife and kids. Huh. Who knew?

I also encourage you to take a look at this video in which some guy interviews an Atlanta male prostitute. I dare you to decipher half of what this guy says.

Ahh, I just blogged all over the internet!

I think there's some blog on your shirt now. Sorry about that. Ok, I just think the word blog is funny.

With the encouragement of Mat, I have now created a blog. There will potentially be stuff on it. Let me know what you think of the look of it. Do the colors make you want to vomit?

I'm also not terribly clear on some of the settings. So, if there is something that doesn't work how you think it should work, please let me know, like if you can't post comments or send me hate mail or something.

Review of The Ancestor's Tale (Richard Dawkins)

Here's a little review I wrote for The Ancestor's Tale: A Pilgrimage to the Dawn of Evolution, by Richard Dawkins. Richard Dawkins shares a birthday with one Mathias S. Fleck, which just happened to be yesterday, March 26th. Happy Birthday to both of you!

Apparently, according to the little blue circle on the cover, this book received the 2004 New York Times Book Review of the Year and it is "a notable book" according to the New York Times. I’m not really sure what that means, but I have to say this is a fine book. I think Dawkins is great at sharing the knowledge of his particular scientific expertise in a perfect amount of complexity for a popular audience who is intelligent, but perhaps not as knowledgable in that particular branch of science. This book gives just the amount of depth and breadth that I like to see in a book about science that captures why the subject matter, in this case, natural selection, is incredibly interesting, without going into agonizing detail and making it a dull read. After reading, it leaves you feeling smarter, but still entertained, which is what I hope for from science writing.

In fact, the book is quite a clever and entertaining way to talk about science. Dawkins sets it up as a cousin to Chaucer’s Canterbury Tales, except that in this book, the pilgrimage is back through the history of evolution by natural selection from the human’s point of view. Dawkins makes clear that he could have chosen any of the existing species and told the tale from their point of view. Along the way, the different branches of species join up as pilgrims with the humans on the way back through the course of evolution. Most times more pilgrims join, Dawkins tells a tale or two or three related to one of the species that highlights an important concept related to natural selection. And this is what I really like about the book. As a 600+ page book about science, it can look pretty intimidating. It’s not really a small book, either; it’s 6” X 9”. That’s one reason I had avoided it until now. I knew that it was supposed to be really good, but some of Dawkins’s other books seemed a bit more manageable, such as The Selfish Gene and The Extended Phenotype. But, after reading those two, I wanted to get more of an updated view of natural selection from Dawkins (the other two books were written 25-30 years ago), so I turned back to The Ancestor’s Tale. Despite the thickness of the book, it did not feel in the least like a boring tome, and I think that is in large part to how it was structured, with the overall story peppered with these interesting, manageable tales.

Each tale touches upon some topic related to natural selection. Dawkins talks about everything from methods, such as fossil dating, DNA coding and comparisons, and constructing taxonomies, to social issues like racism. And every tale features interesting and often humorous illustrating examples. For example, in one tale Dawkins gives an explanation of why men are larger than women, and why gorillas have large chests whereas chimpanzees and bonobos have large testicles. Several tales also deal with clearing up myths and misconceptions about evolution and natural selection, from within the scientific community as well as without. So, the book can be approached as a collection of many short, intelligent, entertaining stories, and you could easily read large portions of it while sitting on the toilet without feeling like you won’t be able to follow some long development of a hypothesis or something. As I said, for me as someone with a background in science, but not necessarily evolutionary biology and ecology, the book offered just enough detail to satisfy me. It was a fairly easy read, though I had to go over some sections a few times to really follow the argument. For someone with more of a background in this sort of stuff, I don’t know. I found myself bored with the level of detail of the arguments in The Extended Phenotype, but not The Selfish Gene or The Ancestor’s Tale, if that gives any kind of helpful measure.

Overall, this has become one of my favorite books. It took me a while to read, but I never felt like I wanted to get it over with. I’d recommend it to anyone with an interest in natural selection.