Cool title goes here

You're in Facebook jail!

Thu, 25 Feb 2010 23:34:00 +1100

It’s not the first time that I’ve seen events like this - the Geocities shutdown was just last year, after all. Nobody cares about their data any more. Just stick it up in “the cloud” and forget it. And it’s getting worse, not better - 40% of the traffic goes to the largest 10 internet domains. That means that when you get pissed off with Facebook or MySpace, there’s nowhere else to go, even if you could get your data out in an accessible form.

It’s like the early internet and email all over again. For the most part, AOL users were limited to just emailing each other and accessing AOL approved websites. Facebook is cut from very similar cloth, ditto for Myspace. It’s possible to get your email out of GMail and back it up, but how many people do? Twitter has an API, but how many Facebook users are going to switch if they can’t play Farmville, post their cat pictures or find their friends? There’s some powerful lock-in happening. Facebook and the other incumbents aren’t going to open up their data so that you can take it elsewhere. You’re a modern day sharecropper - while you’re plowing their digital fields, Facebook are raking in the cash. Privacy? What’s that?

So what can we do about it? Movements like The Data Liberation Front and OpenSocial are a step in the right direction, but I don’t think they’re the whole solution. The best step that I can see is to try and break up the Facebook monopoly, the same way that open email won out over AOL’s watered down system. We need an open system that we can bend to our own goals, rather than Facebook’s.

It’s not even that hard - just google for “open source” “social networking” and you’ll find plenty of potential platforms, with one main problem: they’re all geared towards recreating the Facebook-style platform, only on a much smaller scale. Now you have 100 different Facebooks, from your University to the local pigeon-fanciers club.

What’s needed is for all of those small platforms to interact with each other. Why can’t I friend someone from the pigeon-fanciers club and have that link stored on my account on a different server? Oh right - Facebook wants to lock you into their service. GMail manages to send email to other servers, doesn’t it? It’s not much of a stretch from storing the name of people on the same server to people on a different one - it just takes a slightly different mindset.

In fact, I bet I can Knock One Up In An Afternoon(TM)1. And once it’s reasonably feature complete - I’m thinking along the lines of friends, blogging/pages, groups, status updates and cat picture galleries - I’ll bet it’ll have a reasonable chance of becoming popular. Here are some other ideas:

a real security system, so that you can limit any part of your public profile
OAuth and OpenID support
a plugin system so that people can write their own types of updates

Of course there are other hurdles to overcome, both technical and non-technical. But isn’t it better to get started than to cry about network effects? If you’re interested in helping out, let me know - I’m planning on putting up a repository over the next couple of days. I still need to think of a cool name for the overall plan - openfriendly seems a bit lame.

1 - Particularly when I can use Django, django-registration, -profile and Pinax. I’m betting that most of it’s written for me already, and I just have to figure out how to plug it all together.

Update: I’ve put up a repository and a landing page at http://openf.oarsum.com/ to get people interested. Now that I look reasonably respectable I can get on with the actual coding.

The Weasel guide to Settlers of Catan, part three - the second half of the game, and some dirty, dirty tricks

Sat, 23 Jan 2010 23:25:50 +1100

In which I describe the late game of Settlers, some winning tactics and a few nasty tricks.

In the last installment, I covered the set up and starting game; how to get a good start and avoid throwing it away. In this final episode, we’ll look at how to get into a winning position in the middle game, and maneuver for the win in the final game.

The middle game

The critical question for most people is whether it’s safe to trade with a particular person. Obviously, you don’t want to trade with them and have them get a city out of it, while you get a timber to build a road. There are a few tactics you can employ to minimise the chance of this happening.

Trade with as many players as possible. If you do a one-for-one trade with the other three players, you’re ahead by three for their one.
Pay attention to what cards the other players are drawing. If they’ve drawn three ore and a grain in the past few rounds, it’s a pretty good bet what they’re going to need, and what they’ll get out of a trade with you. It’s up to you to try and get as much out of a trade with them as possible. Naturally it won’t always be this straightforward, but you can often tell roughly what someone’s going to do.
Don’t trade cards that you need, or that are valuable. If you’re worried about someone gaining the lead, don’t trade them particular cards. Critical resources typically work better, ie ore and clay. Don’t let on that you have one of these cards either. If it’s that critical, most players will put the robber baron on you and steal the resource anyway. A simple “No, I don’t have any ore.” will leave them guessing as to whether that’s the case. Don’t be tempted to put a “for trade” on the end of it, since it’ll be interpreted as “Yes, I do have ore. Scads of it in fact, but I’m not going to trade it with a filthy weasel like you.” and then hit you with the baron anyway.
Be patient. If you can help it, don’t trade for cards that you may get anyway in a couple of rolls. It hurts you a little, but hurts your opponent a lot more.
Make them pay through the nose. Trade three-for-two, two-for-one or even three-for-one if you can get it. If you’re going to give them a city, they should at least make it worth your while.
Make them burn their cards. Quite a few weasels will be able to build something by trading in three-for-one at a port, but will try and trade anyway, in order to save those few extra cards. Maybe they can build a city and a settlement if you trade with them for what they need. A good giveaway for this is someone who trades right off the bat, but who has a fistful of cards. Make them trade three-for-one, unless they’re going to give you a settlement too.

Ore and grain are good
The key resources that you’ll be looking for from here on in are ore and grain. Building cities is paramount if you want to get to ten points. If you don’t have a city, you’ll need five settlements, the longest road, largest army and a victory point card. That’s difficult, especially with all of the other players trying to stop you. With a couple of cities, you can focus on either building more settlements, the longest road, largest army or those extra couple of victory points.

As with clay, ore is usually the limiting factor, so if it comes down to a choice between ore or grain, choose ore. Grain will normally have a surplus.

Figure out the other players’ strategies
If you want to thwart someone, you first have to figure out what they’re up to. There’s no point in rushing to cut off a player’s access to brick if they’re a dwarf. Well, most dwarves would sell their grandmothers for a brick, but you get the idea. If you know where someone’s going to play, then you can block them, make them pay through the nose for the right resources, hit them with the baron, and so on.

Count other players’ points
Don’t let that dwarf sneak up on you. Pay careful attention to other players’ points. Once you’ve done that, figure out how they’re going to get to ten points and win. Then you can stop them. Often this goes hand-in-hand with the previous point about figuring out their strategy. Also pay close attention to their potential points too. If someone’s on eight, but has a few development cards in hand, they’ve likely got largest army sewn up, and you should go all-out trying to stop them.

The End Game

This is where it all comes down to the wire. If you’re in a close game, there’ll usually be at least one or two other players within reach of the finish, and it’s just a case of who can claw themselves across the line first.

Battening down the hatches
One of the best tools in determining the winner is the robber baron/knight card. It both reduces a player’s hand, and cuts off his supply of critical resource cards. If a couple of players get in on the action, they can switch back and forth, stealing cards from the leader, and dragging him down to their level.

Don’t let this happen to you! Keep back a knight (or even better, two or three), and steal cards back from them. They’ll still have an effect, but you won’t lose as many cards, and if you can deplete their knights before yours, you can often scrape through for the win.

Damn the torpedoes! Full steam ahead!
Before you start building up your knight supply though, assess the opponent’s knight production. Is it higher than yours? You might want to get an extra one, just in case. If it isn’t though, try and gauge how many knights they can produce. Assume that the baron gets put on your key resource, and that they can produce one (or two) knights per turn, as well as rolling a seven once per turn, to be safe. Now, given that, are you still drawing ahead? If so, there’s nothing that they can do to stop you. Instead of your knight cards becoming assets, they’re now hindrances. The resources that you put into three cards could have built a city, and you would still have had a grain and some sheep left over. It’s also safer and easier to build a city than three cards.

Target minimisation
This one’s easy. Don’t make yourself a target if at all possible. Got a stone? And everybody knows you have a stone? Trade it away, or spend it on something like a development card. Just got to eight points? Well, if you’re sensible, you’ll have done it with your last resource card. If you don’t have anythign to steal, some players will look elsewhere, or will put the baron on you, but so that they steal from another player. That increases your chance of starting a feud between them, which means that the baron might not go on you next time.

How can the other players win?
See that obnoxious player on the other side of the board? Yeah, the one with five settlements and longest road? He’s got a development card that he hasn’t used yet, so it’s probably a victory point card. All he needs is a city or two, perhaps the largest army, and he’s won. Want to stop him from winning? Of course you want to stop him from winning - that’s why you need to nobble his ore hex. It’s the only one he’s got, and you’re pretty sure he doesn’t have any stone in his hand. Without any ore, he can’t build cities, and he can’t buy cards to get that damn knight off his one key resource.

Without figuring out what the other players need to do to win, you can’t hold them back long enough to win yourself. Knowing what they need, and what they’re likely to do, is a key element of winning, at all stages of the game.

Weasel tactics

Here are some miscellaneous nasty tricks which you won’t find in the rule book. They won’t necessarily make the difference between winning and losing, but they can make your game a hell of a lot easier if used at the right time.

Build two roads to claim your settlement
Racing for a settlement spot? I know the feeling - you’ve got two clay, two wood and a couple of sheep. All you need is a grain and you’re there, but no bugger will trade with you. Either they don’t have any, or they enjoy watching you grovel or perhaps they’d like that settlement spot instead. Teach them a lesson by building two roads instead, and block them out. If they can’t drive a road up to it, then they can’t build a settlement there. Mwahaha!

Don’t fan out your cards
Nobody wants to look like they’re winning. Nobody who wants to win does, anyway. And nothing says “I’m going to cane the lot of you” like a massive fistful of cards. That’s right, fan them all out, all twelve of them, and wave them around like you just don’t care. Now see who’ll trade with you? That’s right - nobody. If you keep your cards in a nice tight fan, chances are better that they won’t realise just how much stone you have until they trade you a grain, and you cash in your two cities.

Always shuffle your cards
Not that I’d advocate this, since a lot of people consider it tantamount to cheating, but if you watch carefully, you can often spot patterns in the way that your opponents order their cards in their hand. They might put the most recent cards on the left or right, they might put the good cards on the left, and the useless ones on the right. To make life harder for these pilfering simpletons, just shuffle your cards before you allow them to pick one.

If you want to teach them a lesson, make sure that they see where you put your good card (the one that they want) and then swap it for another one (eg. a sheep) when their attention is elsewhere.

The corollary is to never take a card if someone tries to give you one. Often you’ll hear a line like “Oh, don’t worry, they’re all the same” and then give you a card from their hand. This is annoying for two reasons. Firstly, the only way to call them on this is to pretty much imply that they’re a cheat. The second is that it gives away what’s in their hand. Most of the time it’s reasonably obvious, but it’s still annoying. Just take one of their other cards (perhaps the one on the bottom of their pile?) or ask them to shuffle their hand. If you do it all the time as a policy, it’s harder for them to get upset.

Don’t count pretend points
If you’re on five points, your longest road doesn’t count for anything. Sure, it’s an extra two points, but big deal. What else does it get you? Nothing - a big, fat zero. Anyone who’s on a real five points has two extra settlements or a city, and can take it away from you whenever they like. Unless they’re a dwarf, in which case they’re probably more worried about winning the game than your tiny little longest road.

Use chaos factors to your advantage
Most players have a particular style of play. Under most circumstances, you can pick the way they play, and to a certain extent predict their intentions. A true weasel will exploit the weaknesses in their style to his own advantage. Here are some of the more common styles.

The porcupine will build strongly defensive structures, and try to keep to themselves during the game. They have a strong tendency towards dwarfdom, and won’t play knight cards unless the robber is on one of their hexes. If you’re playing with a porcupine, you can generally leave them alone, safe in the knowledge that unless you’re clearly winning and they happen to roll a seven, they won’t be putting the baron on you any time soon.
The grudge player will single out players who have slighted them, and punish them for their impudence. In extreme cases, they’ll even ignore the leader and instead attack whoever has annoyed them. The best way to deal with a grudge player is to try and incite one of the other players to attack them, and then stand well back, or else “assist” the grudge player in smiting the evildoer. If you can manage it, extra entertainment value can be had by nuking them into the stone age so that they can’t get you back, then watching them turn an interesting shade of purple. Be careful in later games though, as the grudge player has a loooong memory.
Number munchkins love having bigger and better numbers. If you have a settlement on a six, and they only have a five, watch out! They’ll be gunning for you, since “you’re winning”. You can often trick them into skewing their distribution of resources by goading them into taking a better number, but of a resource that they don’t need, or already have a lot of. “You can always trade it in at your port…”
Trade munchkins will trade with anyone and everyone, even for stuff that they don’t currently need, but “might do in future”. While some pre-emptive trading can be a good thing, this player takes it to extremes, wheeling and dealing with abandon, and often getting nothing for their troubles. Once you’ve picked someone as a trade munchkin, you’ll want to be trading with them whenever you can. First in, best dressed is usually the rule with trading, so try and predict what they’ll be offering, and if you need it, be ready to jump in as soon as they open their mouth.
Trade fascists are the opposite breed, and will try and use their trading acumen to affect the game, often to the exclusion of all else. They can be a similar breed to the grudge player - a single slight or dodgy trade can put you on their trade embargo blacklist, sometimes for the entire game. The easiest way to deal with them is not necessarily to stay on their good side at all costs, but to try and get everyone else on the blacklist, either by goading them into attacking the trade fascist, by getting them to trade with the person that the trade fascist thinks is winning, or by trading with another player in a manner that they think is “stupid”.

How to get people to trade cards with you when you’re winning
This one is very Machiavellian, and typically only arch-weasels can pull it off. Let’s set the scene. You’re winning by a long stretch, and pretty much everyone is trailing in your wake as you steam for the lead. Pay attention to how the second and third players are treating the person in fourth place. If it looks like they’ve been completely stalled for several turns, offer to trade with them. Most players will automatically refuse to trade with the leading player, but hey, if they’re not going to win anyway, and they’re getting it in the shorts from people who are meant to be friendly, they may be quite happy to join you on the “dark side”, especially when you’re particularly relentless, drawing cards left, right and center. Sometimes they may join you just out of spite. You gain a friend, they get less frustrated, and everyone goes happy. Except for the players in second and third place, but then they’re trying to stop you from winning, aren’t they?

This works particularly well in tournaments where a player’s standing is based on how many points they score. There’s a significant temptation to score an extra point or two, especially when “you’re going to win anyway”.

The tiebreaker - who’s winning, or who’s the better player?
Hmm, decisions, decisions. Player X is on eight points, and player Y is on seven. Who should you hit with the baron? If player Y has won the last two games in a row, stick it on them. Even at a point behind, they’re still more likely to win.

The truth is negotiable
That’s right - you can even argue about how many points you have.

“Sure, I’ve got seven points, but two of those are in longest road, so they don’t really count. I could get that taken off me at any time. I’m not really that powerful. Look at Boris over there. He’s got five points, and he hasn’t played that development card for a while, so it’s probably a victory point card. So he has six points, but I only have five real ones…”

Sometimes this will work, or at least convince them to spread some of the pain Boris’ way. Sometimes though, when you really are winning, there’s no getting out of it, and it’ll only make the other players want to hurt you more. In those cases, it’s best just to take it and smile.

Don’t boast when you’re winning
Notwithstanding the etiquette side of things, boasting when you’re winning is an easy way to get the other players to see just how far ahead you really are, and makes it much more likely that they’ll gang up on you. If your resources get rolled, try not to draw attention to it - just take your mumblemumble stone and hope nobody notices. If it’s a resource that’s in short supply or that you need to win, and you’ve been waving them around for the past couple of turns, don’t be surprised when you get hit with the baron.

Don’t whine when you’re losing
Trade is a two way street. For people to trade with you, they have to expect that you have something that they want. If you’ve been repeating the “Boohoo I don’t have any good cards” mantra, how likely is it that they’re even going to ask? To dig yourself out of the hole you’re in, you’re going to need either phenomenal luck, or a good trading partner. Guess which one is more likely?

The leader is a lightning rod for pain
Normally, the best position to be in is a close second. That way, you have the diversionary tactic of pointing at the leader whenever anyone tries to hurt you. In a four player game where the scores are 5-5-7-8, the player on 7 points should really get a little bit of pain too, if only to bring them back into line, but most people only see the winner. Of course, sometimes you’re winning so heinously that you can go it alone, damn the torpedoes, bring on the baron, etc. etc. But for those times when you can’t, it’s good to have a little help.

Beware the first turn road nazi!
With this trick, place your second settlement on a timber/clay intersection so that you start the game with the resources to build a road. Now you can automatically win a ‘road race’ with someone else that has a road right next to yours. If you can get in before they’ve had a chance to build their road - usually on the very first turn - their road is wasted. Bwah ha ha!

How to make people really hate you
Don’t try this unless you really need a particular resource. You’ll make enemies for life.

Monopoly and knight cards are great, but to use them to their full extent, you need to find out what other people have in their hand. Ask them. Offer to trade whatever you think will get them enthusiastic - a couple of grain for an ore (if you want ore). Once you know that they have ore, hit them with the monopoly card, or put the knight on them and try to draw it.

If that’s not evil enough for you, and you’d instead prefer to annoy people so much that they’ll likely drag you outside afterwards and beat you senseless, try trading a large amount of one resource that everyone wants desperately. Tell them that you’re not in a position to win (perhaps after hemming and hawing for a while to make it convincing) and that you’re spreading the resources around to make it harder for the robber baron - anything that’ll get them to trade. Once you’ve exhausted your supply of ore, or your opponents patience, simply use your trusty monopoly card to get it all back again.

For bonus points, try to continue trading afterwards.

Conclusion
There you have it - how to conclusively win at Settlers, every time, without fail. Well, maybe not quite that good, but at least you won’t get taken for a ride by most players out there, and hopefully you’ll be able to win a few games. Settlers is a fun game, and while there is some element of luck, for the rest of it it’s all skill.

If there are any other nasty tricks that you know of, feel free to let me know in the comments below.

Thanks,

Anthony

The Weasel guide to Settlers of Catan, part two - The First Half

Tue, 22 Dec 2009 23:44:33 +1100

In which I describe the initial game of Settlers, set up and starting strategies.

In the last installment, I gave an overview of the various Settlers strategies around, and how to play them. This time, we’re going to look a bit harder at some of the tactics that you can use to make your opponents’ lives harder. Since there are quite a few tricks, I’ve broken it up into the first half of the game (this post) and the second half (the next one).

Initial Placement

Your initial placement in the game is very important. Get a good head start on the other players and you can streak into the lead, where only a concerted counterattack can hold you back, or even slow you down.

Lies, Damn Lies and Statistics

The critical issue for your initial placement is fairly obvious: How many cards do I get? While it is obvious, there are a few subtleties that you may not notice at first. Try and keep the following in mind:

Individual intersections
First and foremost you should try and maximise the average number of cards that you get from placing on an intersection. There are a couple of ways of doing this, but the chief one is to figure out the chance (out of 36 rolls) that you’ll get a card. Over thirty-six rolls, you should get that many cards. Intersections can vary from 3/36 (an 11 and 12 on a coastal hex) to 12/36 (a six, nine and ten or similar). The intersection totals are distributed roughly the same as a standard 3d6 roll, which means most intersections will be at about a score of 8 to 10.

Overall score
Also pay attention to the overall score, particularly when assessing other player’s positions. Sure, they may have a particularly good eight-five-three number, but if their other placement is shocking, they’re not going to do so well.

Well-roundedness
Fate is fickle, and often likes to play around with the dice. How many times have you gotten that eight grain hex, only to have six after six get rolled instead? To try and avoid this, get access to as many different numbers as possible while maintaining your overall score.

When is a good score not a good score?
This is a bit of a trick question, but try to avoid doing too well out of your initial placement compared to the other players. It’ll single you out as public enemy number one right from the get-go, and you’ll be stalled on the starting line. If your sensibilities rebel at the thought of picking an inferior placement, try and point out better spots to the people who come after you. Remember though, that you still want to come out ahead.

Geometry

Geometry refers here to your overall placement in relation to other settlements. By placing your settlements in certain ways, you hope to limit other people’s access to hexes and resources while keeping some for yourself.

Dense placement
is the optimal solution for packing settlements around a hex. Each settlement is two spaces from another, and there are three settlements around each hex. Generally you’ll use this around your own areas, to maximise the amount of useful land. Notice how you can fit three settlements on the one hex, and still have room for more around the edges.

Sparse placement
is the opposite of dense placement. Each settlement is placed three spaces from the next one, usually starting with two on either side of a valuable resource. The settlements are as far away from each other as they can get without allowing another player to place in between them. This is usually used in other people’s territory, to cut down the amount of available land, although it’s also used by dwarves to limit the expansion of more settlement oriented players. Notice that no settlements can be placed on the central hex in this image, and that there will be many more places where settlements cannot be placed in the hexes outside.

Coastal vs Inland
Coastal hexes are usually the better bet. You’re less likely to get shut in, and more likely to reach a port. Placing directly on the coast is generally a bad idea, unless it’s to annoy someone bt grabbing their port, or there are better numbers or resources on the coast. You get two hexes instead of three, and what’s worse, you might get shut in, since coastal hexes have fewer adjoining intersections than inland hexes. Note the image - red is going to have a hard time getting another settlement from the one displayed.

Other considerations

These are the non-numerical things that you should take into account when placing your initial settlements. They’ll often mean that you take a hit in raw scores, but the positional advantage that you gain may be worth it.

Shutting other players out
involves building around them completely, so that there is physically nowhere for them to build to. As a variation, you can simply build in their way, so that they’re forced to detour around you, spending an extra couple of roads to get where they need to. Players who build on the coast are particularly vulnerable to this trick, and players who are inland are also susceptible towards the middle of the game, when you can deny them a port that they need. See the coastal vs. inland section above for a picture of how this works.

Limiting another player’s expansion
in the early game is important, particularly if they have good clay and/or timber numbers. An early expansion can net them the lion’s share of the resources, which they can then capitalise on at leisure.

Spite!
Often you’ll be better off (and get more satisfaction) by placing on a hex that another player wants. If they have the only clay hex, for example, and you know that they’re going to need grain, obviously you want to place on as many grain numbers as possible, so that you have some leverage when it comes down to getting clay out of them. Often you may be better off denying another player some key resource that they need, than taking an average or below average spot. The same applies to getting in their way if you think they’re going for longest road, or building across their path to a port they might want.

Resource balancing
Don’t get stuck with nothing but sheep and grain. There’s often a lull in trading at the start of the game, so if you have to roll too many numbers for your first expansion you may end up being a turn or two behind the pack. Initially you’ll need two clay, two wood, a sheep and a grain, or two grain and three ore if you’re a dwarf.

Pile in on crowded hexes
If there are two people on a good hex, it can often be worthwhile joining in. It’s far less likely to get the robber put on it, and if it does there’ll be three people trying to get it off, not just one.

Don’t do what other players do
The converse also holds - if there are lots of people trying the same strategy, it often pays to try something else and grab a different part of the map for yourself. They’ll all be struggling with each other while you have plenty of room to expand.

Steer clear of some players
Not that I’d ever play like this, but some players hold grudges, and attack other players purely because they want to get even with some previous slight, or because that player happens to win more often than not. If one of the players is particularly unpopular, try to avoid placing your settlements next to theirs.

The early game

The early game is where it all happens. Empires are won or lost, roads built and settlements founded. It’s also where a lot of games are won or lost. You’ve already (hopefully) made some good choices in your starting positions - here are some tips to help you nail down your win so that other players can’t pry it up and run away with it.

Expand like a loon!
This is one of the critical aspects of settlers. You’ll need to expand dramatically in order to have a chance in the later game. Count your resources, preferably before you even start. To build your third settlement, you’ll typically want two bricks, two wood, a grain and a sheep. What are you missing? If you don’t secure those resources, by whatever means necessary, you’ll falter in the early game and never catch up.

Unless you’re extremely lucky, you’ll likely have at least one gap in the resources that you’ll need. Look around the board and figure out who’s going to have that resource. More importantly, figure out what they’re going to need. Chances are that it’ll be something that you have, or are likely to get. They’ll need to trade at some point, and when they do, you’ll be ready. If they don’t want to trade, you’ll know where to put the robber when you roll your next seven.

Don’t waste your cards
Since cards and resources are critical, it’s important that you don’t waste them, particularly if you only get a resource from a mediocre number. The two biggest areas of waste in the early game are from roads and development cards. It’s often better to sit with more than seven cards rather than possibly waste a road. The only exception is if you are short of timber and clay and fear someone stealing from you. Similarly, the only time that you should buy a card in the early game is if you have the robber on you and can’t get it off, or if you’re a dwarf and you have the cards to spare.

Malthus and exponential growth
Malthus was an economist who proposed in 1798 that populations could be modeled in terms of exponential growth, limited by certain factors in the environment, such as mortality rates and scarce food. The same theory can be applied to help you win at Settlers.

First of all, recognise that resources in Settlers are used to increase your production capacity, which means more resources, which means more production - exponential growth, in other words. The critical issue for each game is to identify those factors which are limiting you, and take steps to alleviate them.

A lack of resources, if your numbers are too poor relative to the other players. This can be relieved by building more settlements, or converting your best settlements into cities.
A lack of the right resources, if you have plenty of sheep or grain, but no clay. This can be helped by trading with other players, building out to a port, buying development cards or building new settlements out to resources that you don’t have.
A lack of space to expand, if someone has built up right next to you. There are only a few ways to counteract this: either try and impose more limitations on your rival, or try and break out before they encircle you.
A lack of pieces, ie. you can’t build more than 15 roads, 5 settlements or four cities.

Note that while you can remove limits from yourself, you can also impose limits upon the other players, by hitting them with the robber and shutting down their good numbers, blocking off their roads or limiting their access to stone later in the game. This is almost always a bad idea unless one player is so far ahead that they are the only threat, and the only way for any of the other players to win is to make a concerted effort to drag them back. To gain the same benefit from removing one of your limits, you need to impose a similar limit on the three other players.

Brick and grain are good
Brick is useful for building roads and settlements, two of the most critical elements in a player’s expansion. It only occurs in three hexes, so it’s often scarce.

Grain is useful for building settlements and cities and for buying development cards. It occurs on four hexes, so it’s often not as scarce as clay.

Timber is less good
Everywhere that timber is used, clay is used too - but timber has four hexes to clay’s three. So clay will almost always be the limiting factor when it comes to building roads and settlements. There is usually a glut of timber, particularly when clay is scarce. But hey, it’s still better than sheep…

Early game hexes are better
The hexes that you secure in the early game are far more valuable than those secured in the middle and end games, so make sure you expand early. There are three key reasons to bear in mind:

Decreasing numbers. As the game goes on, more and more of the good hexes will get taken, leaving the crappy ones for the dwarves. If you get a hex in the end game, chances are it’s a 2 hex bordered by desert.
More time. If you build a settlement straightaway, you’ll get more resources out of it compared to one that you build ten turns later.
Exponential growth. All those extra resources that you’re getting are going into building more settlements. The faster you can build that first settlement, the further ahead you’ll be in another ten turn’s time.

Summary

If you haven’t figured it out by now - grow like your hair’s on fire! You’re competing with the other players in the game, remember? The better your starting position, and the more stuff you can get, the faster you’ll grow, the better the hexes you’ll get, and the more you’ll pull ahead. Cram as many of your resources as you can into developing more resources, and you might just win through.

The Weasel guide to Settlers of Catan part one - The Strategies

Mon, 30 Nov 2009 18:50:00 +1100

In which I describe the various types of Settlers strategies, and their strong and weak points.

This guide was first written in late 2002/early 2003, while I was playing Settlers semi-regularly with friends in Perth. It’s tempting to try and change it, but I haven’t played much Settlers of late, so I’m somewhat rusty and it’s hard to know which bits are right and seem wrong, or wrong but seem right. So I’ll just write it out straight - I lost the original electronic copy, but recently found a long-forgotten hard copy, so I’m retyping it. I also unfortunately lost the original weasel graphic, and have had to improvise using Google images.

Thanks go to Lee Taylor and Jason Funnell, who helped review earlier drafts and provided extra weasel tricks, way back in 2003.

Strategies

There are several basic strategies that can be used to win a game of Settlers. Most of the time you’ll use a couple of them together, but sometimes you can win just using one. Quite often the particular strategy that you play will be dictated by chance. Rather than being able to pick and choose, you might only have one clear strategy that’ll work. A good player will recognise when this is the case, and will be able to play many different strategies as circumstances permit.

City Builder (aka. “The Dwarf”)

This is the normal uber-strategy. For best effect, try and get a balance of stone, wheat and sheep, perhaps with an easy run to a port. Dwarves tend to grow slowly at first, but if they get a few good numbers in the middle game, they can streak for a win with blinding speed. All they need are three cities, largest army and a couple of victory points, and they’re home and hosed.

Sub-type: “Run to the hills!”
Dwarves are really, really good at battering down the hatches, and making trouble for players that try to stop them from winning, since they usually have cards in the right proportions for churning out development cards. Attacking an established dwarf often takes the combined efforts of the rest of the players.

How to play a Dwarf
The main criteria for being a dwarf is stone. Lots of stone. If you can get some grain and sheep too, so much the better, and if you can get to a port of some description that’s pretty much the win. Early on, you’ll be hoarding stone and grain to build up your cities. Once you’ve done that, you need at least one extra settlement, preferably two, followed by victory point cards and the largest army.

An evil trick (if you’re a Nibellung, or dark dwarf) is to buy a development card early on and use the baron to nobble the person who looks like they’re coming second. Naturally you’ll be coming first, and this will allow you to increase your lead before those pesky overlanders grow too powerful.

How to annoy Dwarves
Dwarves need space in the middle game, once they’ve built up their two initial settlements into cities, and start trying for their third city. If you can deny them this, particularly if their third spot is good, they will stagnate. So, build around them in the early-middle game, and they’re as good as toast. Of course, you need to see it coming, but dwarves are usually pretty obvious.

The only other ways to beat dwarves are to take largest army away from them (hard to do, since they usually have the better stone numbers) and/or starve them of cards by hitting them with the robber baron (again, hard to do).

Settlement Builder

This is a more spread out way of playing than the dwarf. It usually involves building settlements as quickly as possible in order to increase your production, often focusing more on quantity than quality. There are a couple of sub-types, the octopus and the road builder.

The octopus
The octopus is similar to the settlement builder, but a bit more choosy over where his settlements are placed. You’ll usually see settlements at the end of long roads, or ‘arms’, where they’ve snuck around to a dwarves’ ore hex.

The road builder
Sometimes one player will have too much wood and clay, and very little grain or sheep. Instead of trying to scrape together enough grain to build a settlement, the road builder often drives over other players can’t get to them. This can be dangerous, since a road doesn’t increase your production capacity, and you can stall if your numbers don’t get rolled for a while.

How to play a Settlement Builder
When placing your initial settlements, try and focus just as much on the balance of resources as how good your numbers are. You should try for a mix of three clay, three timber, a sheep and a grain, so that you don’t have to trade too much to build your settlements. In the early game, you should be able to build out to a couple of ports, to smooth over any wrinkles in the resources that you get. At your third or fourth settlement (depending on how the game’s going) you should start thinking about building your first city, followed by a couple more settlements if you have the room. After this, it’s a case of who gets to the longest road or largest army first.

How to annoy settlement builders
Two words: No Clay. If you can nobble their clay resources early, often the settlement builder doesn’t have the knight card necessary to free themselves. If you can make them stagnate in the early game, there’s a good chance they might no recover in time. If you can’t catch them in the early game, try and stifle their ore or grain supplies so that they can’t build cities. Five settlements plus longest road only gives you seven points, which makes the settlement player unhappy.

Merchant Princes

These players will typically try to get a good supply of a rare and/or valuable resource, usually grain, clay or wood, in the hope that they can trade it for other resources that they need.

The Monopolist
A Monopolist is a Merchant Prince with the only good number for a resource, which forces all of the other players to either trade with the Monopolist, trade for it at a port, or steal it. Whichever way they choose though, they’ll be at a disadvantage.

The Cartel
It’s quite hard for a player to lock down a resource completely by themselves, but it’s possible for two players to place on opposite sides of a hex and lock the resource up between them. This has the advantage that two people are now defending against the robber baron, but the disadvantage is that they are also competing against the other member of the cartel.

The Port Munchkin
The port munchkin is a player who has managed to gain a large amount of a resource, on good numbers, along with that resource’s two for one port. It gives the player a large advantage in terms of gaining rare resources, and is also immune to the aging of resources such as clay. It works particularly well with resources that other players don’t want, such as sheep.

How to play a Merchant Prince
Basically, grab a resource and hawk it for all it’s worth. Hawking it to other players one for one (or one for two if you can) is the best, but you need to be prepared to sell it at a port if necessary. In the early game, clay and grain are usually the most important, but it will vary depending on the specific board that you’re playing on. Always look to expand your coverage to other hexes, since normally you’ll be missing a few in your rush to cover the rare resources.

How to annoy a Merchant Prince
Usually the best method is to fight fire with fire. If you have a resource that they need, then you can swap one for one with them. Typically though, there’s no one strategy that will work exclusively, so be prepared to try a mixture if one strategy fails. If their hex is a clay hex, try a robber baron as early in the game as you can manage. If they don’t have anything to trade, they’ll usually stall. Usually a combination of the robber baron, a trade embargo and building close to them will be enough to finish them off. Watch out for the other players who want the resource so badly that they’re prepared to help out the besieged Merchant - you may find yourself fighting a battle on two fronts.

Machiavelli

Generally this strategy will work in close games where you only need a very slight advantage in order to win. It involves either turning the other players against each other in order to gain advantage, or befriending a player so that they will instead attack other players. Quite often, supporting a player who is competing for longest road is enough to set both them and their competitor at odds, allowing you to concentrate on building cities and consolidating your largest army.

How to play as Machiavelli
Try not to annoy the other players directly. Instead, concentrate on pointing out things that are in your favour, and don’t mention things that are bad for you. Encourage players when they’re doing things that you like, and make frowny faces likes this >B-( when they’re doing something bad. Of course, you’re doing this for their own good - you wouldn’t like to see them being taken advantage of. Take great care to encourage those players who seem particularly malleable. You may sometimes need to put spin on your actions to avoid political fallout.

Of particular note is the devious scheme of ‘propping up’ one of the other players, so that they are dependent on you, while remaining relatively harmless (Letting them have the longest road is good for this). Now you can let them take the lead, cop the flak for you and harass other players while you grow, and then yank the rug out from under them in the end or middle game by withholding key resources.

You might find the other players suspecting you of being a weasel. They’re probably right, but to throw them off the scent you might like to throw out phrases like “enlightened self-interest”.

How to annoy a Machiavellian player
Simply point out to the other players that Mr Machiavelli is a snivelling weasel. Most of his tricks rely on hoodwinking the other players, and playing fast and loose with the truth. Pay particular attention to the facts, including things like how many points the players have, how close each of them are to winning, and resources that the weasel has picked up. I fthis fails, you might try pointing out how reliant the player is on the weasel’s good will - nobody likes being a patsy.

The Generalist

The Generalist will try for a balance of most of the strategies. This is a good way to hedge your bets, particularly in a close game. If two or more of the other players are competing with the same strategy, then the generalist can usually switch to an easier strategy with little trouble.

How to play as a generalist
Put your finger in a lot of pies. You’ll probably want to start with a reasonably balanced distribution of resources, and work out from there: a bit of stone here, some roads there, a couple of cards, maybe a port, all the while maintaining enough diversity to be able to switch from one strategy to another as necessary.

How to annoy a generalist
Most generalists will use a smattering of each strategy. It makes them weaker individually than the more specialised strategies like the dwarf or the merchant. If they look like they’re trying to compete, usually the best defence is to specialise even more, trying to pull ahead on your ‘natural’ talents.

The Producer

The Producer will try and overcome the other players by the sheer weight of cards that they can produce. They’ll typically produce large numbers of cards and trade them in however they can. Their strategy is centred around getting enough cards that they can trade four for one until they can get to a port. Quite often they’ll be general enough that they’ll want this port to be a three for one port.

How to play a producer
Count how likely each resource is to give you cards. Get as many fives, sixes, eights and nines as you can, always concentrating on maximising the number of cards that you’ll get. If you can, place yourself within easy reach of either a three for one port, or a port for your most common resource.

How to annoy a producer
The producer gets lots of cards, but his key weakness is that they’re usually in the wrong combinations to do anything useful. They’ll have to either trade away for what they need, or get to a useful port. If you can block them from getting to a three for one port, they’ll usually wallow for quite a while.

Stay tuned for next week’s exciting episode, where I’ll give you some tips on how to annoy people with your initial placements, and crush people in the early game.

"Cheating" and game design

Tue, 24 Nov 2009 21:50:00 +1100

Like a lot of people on Facebook, I tend to play silly web games. Unlike most people though, I have a very low tolerance for the mind numbing tedium involved, and tend to overanalyse things. Plus, I can program.

I’ll use Farmville as an example, but any other popular competitive web-based game can stand in it’s place.

If you take a look at the Zynga terms of service, you’ll notice the usual restrictions on doing idiot things; spamming and abusing people, sending viruses and generally being an asshat. One term bears closer examination though:

xiv) except as may be the result of standard search engine or Internet browser usage, use or launch, develop or distribute any automated system, including, without limitation, any spider, robot (or “bot”), cheat utility, scraper or offline reader that accesses the Service, or use or launch any unauthorized script or other software; …
xix) engage in cheating or any other activity deemed by Zynga to be in conflict with the spirit or intent of the Service.

In other words, no bots! At first glance rules like this are fair enough. You don’t want people scripting your game, getting lots of goodies, and having an unfair advantage. But once you look a bit closer at the game and the business models involved, things start to look a bit murkier.

Playing Farmville

For those of you who don’t know, Farmville is played by plowing plots of land, planting crops in them and then harvesting them once they’re grown. A typical starter farm in Farmville looks something like the image below. Notice how it’s filled completely with crops - this is the optimal layout for your plots: 12 by 12. If you put trees at the front, then they’ll obscure the fields and you won’t be able to harvest them.

Each time you harvest a field, you get some experience and cash. As time progresses and you level up, you get access to more profitable crops and can make more. Depending on the crop, you’ll get 1 or 2 XP to start, plus 1 for plowing the field. You can find a spreadsheet with all the data that you need here.

For day long crops, we’ll get 3 XP per plot per day, times 144 plots, for 432 XP. Each plot will need three clicks - one to harvest, one to plow and another to replant. Checking that against the list of experience levels will give you an idea of how much grinding you need to do. Your first week might get you to level 10 which is only 1800XP, but to unlock peas is 61000 XP, and 61000 mouse clicks - around 141 days worth of clicking.

As if that’s not enough, Zynga provide tools to automate some of the grinding, with items like Tractors. They can plow four fields at a time but Zynga only provide a limited amount of fuel, not enough to harvest or plant as well. If you want more fuel, you’ll need to pay real money, and keep paying. Hence the desire for scripts, and the terms of service to try and prevent people from scripting - Zynga are protecting their cash cow.

The problem

The deeper problem as I see it, is this:

If I can script your game, your game design is broken.

Seriously. If a script can play a game better than I can, your game is boring and tedious. There’s no fun, except in figuring out the optimal route to riches; the rest is just mere execution. If you don’t want people to “cheat” at your game, make it fun, with no boring bits. Zynga know this, but it’s an easy way to make money, so they’re not going to bother changing their game.

Go read some game design articles, or a book or something if you want the details, but one of the critical parts if you want your game to have lasting appeal is that players need to make real, difficult choices. In the context of Farmville, which crop to plant is not difficult at all - just look it up in the spreadsheet and pick the best one to maximise your income given your level and available free time. Ditto for the placement of your fields, decorations and buildings - it either doesn’t matter, or there is one obvious optimal placement.

Solutions?

So what would I do if I were Zynga? What real choices could you make in the context of a farm game? I think a good first step would be to make the basic game world more complex. More complexity generally means that there are more factors to take into account when making choices; the difficult part is in deciding how much to add.

Weather, for example, could make the choice of which crop to plant a lot more interesting. With no rain forecast for the next few days, perhaps those eggplants can’t grow so well and won’t be quite so profitable. Different plants could have different tolerances for rain, wind, sun and frost, making the choice a bit more interesting than just “What gets me the most money?” The weather report might not be 100% accurate either, but with rain tomorrow followed by good sun, it could be worth the risk of planting a more fragile, but more profitable crop vs. something hardy but cheap.

“What gets me the most money?” could also be open to negotiation. Supply and demand have always been factors for real farmers, so why not introduce it in game? Prices of crops could depend on how many people are growing the same crop. If everyone in your group of friends is planting strawberries, then there’ll be a glut, and the price will go down. Rarer crops, or those which are harder to grow will fetch more of a premium, without having to hard code prices into your game.

Demand from farmers could have a similar effect on the price of seeds, and allowing farmers to collect seeds from their crops, then stockpile and sell them later could add several extra dimensions to the game. Chances for spoilage based on weather and storage conditions would remove surpluses from the game and encourage players to spend in-game money.

No doubt there are other factors which could be brought into play - you just need to read your local agricultural news for inspiration. Labour strikes, local supermarket monopolies and commodity price crashes could all provide excuses for fluctuations in the prices of crops and seeds. More complexity, more choices, more fun.

Coming full circle

In the long term, I see scripting as a good thing, not an evil to be wiped out. You save your precious time upon this earth, plus you avoid the crappy design, even if the developer doesn’t eventually change their game.

If even one of the changes above were made, Farmville would be at least an order of magnitude harder to script, and more fun to boot. The scriptability of a game (short of technical challenges such as getting game state out of Flash) could be seen as a litmus test for it’s quality. If it’s easy to script, it’s a crappy game, but if you need to write an expert system in order to do well, then maybe there’s hope.