Posted on

Deep Learning Is Not Empowering Us

I see articles that keep pushing how it’s important that every programmer (if not everyone) learns deep learning algorithms and that it will to a democratization of power.

This is stupid.

A deep learning algorithm is worthless without big data. There’s a reason why Google and Facebook spend so much effort giving you free webspace, email, and search engines, and it’s to collect the data. Access to that data is very not-free. Without data of that magnitude all your deep learning algorithm will do is make another ‘hilarious’ list of new vegetables named “plueberry” or w/e.

The reason these articles exist is cause salaries are too high for good NN programmers and the companies need these devs to be as abundant as web devs are now, to keep costs down.

Everyone who says they want to empower you is lying to you. The only thing no one will ever, ever simply give you is power.

Posted on

Google Writing Drone AIs

Re this: https://www.theverge.com/2018/4/4/17199818/google-pentagon-project-maven-pull-out-letter-ceo-sundar-pichai

I’m of two minds on this story. On one hand, I get the that the programmers are unhappy about it, but on the other that algorithm will get written no matter what, so people who are good at implementation may have some sort of obligation to minimize civilian casualties by making sure it does its job well.

That and they’re working for Google. It’s not like they really have some sort of moral high ground just because what they do is starting to feel viscerally wrong as opposed to just wrong in the abstract.

Yeah ok, I’m not really in two minds about it as you might tell, I just said that to try and be less antagonistic about stuff. People just blanche when they can see the results of their actions clearly enough to make it impossible to deny them away. Shut up and write the war machine, Googlers, it’s probably the least damage you’ll do humanity all day.

Posted on

Neural network generated list of neural network generated lists

People have been creating little text neural networks that take things like band names and make lists of more band names that are all a little off. I think this is delightful and not at all getting boring by now so I designed a neural network to design lists of things designed by neural networks.

Here's the best of the neural network designed lists of things we used a neural network to make lists of:

april fool’s bands
death trump children
final catholic saints
mega man lyrics
catholic mistresses
anime super twilight novels
linkin park linkin park songs
start ups villains
death metal board games
lies
neural network designed neural network designed lists
rick and mortiis

Posted on

Algorithmic Failures in Health Care

Verge has a really good story on it.

It's yet another preview of our future. While these are expert systems vs neural networks, but the issues are similar: the algorithm says that you shouldn't get health care and no one can explain why cause the math is a trade secret and no one understands it even when it's revealed. Then, when finally read out, it turns out that it was written by unqualified lowest bidders and contains 900+ errors in implementation.

Two quotes:

Most importantly, when Idaho’s system went haywire, it was impossible for the average person to understand or challenge. A court wrote that “the participants receive no explanation for the denial, have no written standards to refer to for guidance, and often have no family member, guardian, or paid assistance to help them.” The appeals process was difficult to navigate, and Eppink says it was “really meaningless” anyway, as the people who received appeals couldn’t understand the formula, either. They would look at the system and say, “It’s beyond my authority and my expertise to question the quality of this result.”"

There was also no way to effectively challenge the system, as they couldn’t understand what information factored into the changes, De Liban argued. No one seemed able to answer basic questions about the process. “The nurses said, ‘It’s not me; it’s the computer,’” De Liban says.

They always said that learning programming was important, and I suppose they weren't kidding.

Posted on

Space Ninja, review

Slightly obscure review today: Space Ninja from the English-language friendly Japanese company Group SNE.

The box art is way more violent than the game

The game is a worker placement, your workers are agents placed onto various planets in order to influence their individual politics. At scoring time, they earn you points based on who controls planets: player with most agents getting one point for every ‘civilian’ living on the planet, second place getting half that. In case of ties, the earliest agents win.

The heart of the game are the different planets available. 

The game comes with 6 though not all of them are used in each game, and they are very unique. They all follow the same basic format:

  • an area for the population to live on, with a maximum size
  • a row for the player owned ninjas to line up on
  • the effect that happens when you place a ninja
  • and the effect that happens at the end of the row

Most of the planets involve a ‘programming’ mechanic where placing ninjas gives you a say, either by moving populations to voting areas, or paying off officials to pass legislation, or rolling a dice to determine what the fate will be unless someone else shows up and rolls as well. I’m making these sound kinda plot driven, but they’re very euro in practice, heh.

Which leads us to the core mechanic: more population is more points, but only up to the limit. If the population ever crosses over it, the planet ‘explodes’. Half the people die, and more importantly, half the agents die, oldest first. This means that the game is an endless balancing act between the people earning points trying to keep the population high but not too much so, and the players whoa aren’t earning points who are trying to either crash it, or grow it over the limit. And with the complex actions possible on the planets, predicting which is easier is non-trivial.

Programming example: the mandarin planet is set to export a pop counter-clockwise, then shrink their own population while giving everyone a free ‘placeholder’ ninja. Meanwhile, Enlil in the back is at it’s final state where it grows slowly and kills of pops in the neighboring planets.

Btw, did I mention the game is very low luck? Two planets involves a die roll (one to determine the population limit, one to determine how it will behave at turn end), but generally speaking you have a very good idea of what will happen at the end of turn and the only difficulty is predicting the actions of your fellow players. I definitely would not recommend this at 2p since I have a feeling it’d turn into an AP fest.

A very good game, recommended. It takes us about 2 lunches to get through the full 4 rounds, but well worth it if you like the more experimental Japanese euro-game scene.

Posted on

Mystic Vale, short review

We got to play the upcoming Mystic Vale at Kingdom Con in San Diego. A cool entry in the tiny field of ‘card builders’.

Mystic Vale. Tough to avoid glare on this one
Mystic Vale. Tough to avoid glare on this one

The gimmick of the game is the translucent card fragments. You start the game with a hand of cards which will never grow or shrink. Onto each card you then build up more and more effects which all fire whenever the card gets played. Spend those effects to gain more and better effects and acquire points. The one to finish with most points, wins.

So what’s the details? Each card is a little envelope with a single starter ‘background’. Some are blank, some contain an effect. All the starting ones are either a standard “gain 1 gold” or a red icon. The red icon is the other clever bit of the game: you draw cards until 3 of those are visible. You can then press your luck and draw more. No red symbol? Add the effects on the new card to your hand. Red symbol? Bust. Lose this hand, but gain an extra gold for next turn. Adds a nice extra dynamic to the early game especially.

The card ‘stacking’ works really well; it’s visually very much like Gloom. They’re easy to read and clear, though they don’t photo well. One mistake that’s happening in the photo above is I’m overlapping the cards on the right. I learned not to do that cause you can miss some of the trigger effects that don’t have a visual icon on the left.

The feels more like a deck builder then not: you set up combos, you hope they fire in the right order to let you best use the market. The lack of card shedding gives it a different feel obviously, but it’s familiar to make sense quickly.

My only complaint is that the game would probably feel shallow after a few plays, as you explore the range of cards. I fully expect there to be a few expansions that will help with that. Also, I’m now approaching my 200th game of Star Realms with just 1 expansion, so I personally don’t think a tight well balanced deck/card builder has anything wrong with it.

Looking forward to the final release of this. We’ll be picking it up.

Posted on

Kingdom Con Loot

No game purchases at Kingdom Con this year, but we did get a nice bag for carrying stuff, and these mini adventurers for Lords Of Waterdeep

Aren't they adorable?
Aren’t they adorable?

Actually makes it feel like you hire guys instead of shuffling cubes around.

Posted on

BuzzFeed-style board game meme list

Made a stack of buzzfeedy board game memes.

 

When you’re sneaking up on a science victory in 7 Wonders and no one noticed

 

When they try to connect their big meeple to your farm in Carcassonne

 

When you roll 3s even after going to two dice in Machi Koro

 

When every single one of your units is webbed in Neuroshima Hex

 

When in Tragedy Looper you enter last loop and don’t even have a theory

 

When someone publishes a paper in Alchemists and you know it’s wrong

 

When you start 1830 and first action they buy SV

 

When you solve TIME Stories on first playthrough

 

When in Game of Thrones you play Lannisters and the Greyjoys attack both you and the Starks turn 1

 

When he goes for the Scholar’s Mate against you

 

When you get a third Monster Surge in a row in Arkham Horror

 

When you’re trying to explain Mottainai

 

When in Takenoko their quest are the same as yours but worth more points

 

When you’re one point short of winning a seven hour TI3 game and the guy with 3 points suddenly has decision paralysis

 

When you banish someone out of Dead Of Winter and next turn the supply is sabotaged again

 

When literally anything happens in Robinson Crusoe

 

When they first turn gambit into a MegaHauler in Star Realms

 

When no one crashed the market for your goods in Planet Steam

 

When in in an 18XX the bank runs out of 5Ts before you can buy one

 

When you’re given a 0-clue in Codenames for the first time

 

When you’re not the Fake Artist but you don’t know what the object you’re drawing looks like

 

When in Sheriff of Nottingham they’re about to open the saddlebags before you have a chance to try to bribe them

 

When you take down a wyrm with two clones in Nexus Ops

 

When they successfully coup a 4-stability country in Twilight Struggle

 

When someone uses a Guard to guess you have the Princess on first turn in Love Letter

 

When everyone but you forgot to keep a pudding in Sushi Go

 

When someone offers to trade their wool near end of game in Catan

 

When you get at least half of these

Posted on

Euphoria: Build a Better Dystopia review

We picked this guy up from our anniversary purchases. I’m quite fond of it.

Euphoria. Legitimately beautiful board.
Early play in Euphoria. Legitimately beautiful board.

Euphoria is a dice placement eurogame with a great theme and a bunch of interesting twists. You play a middle manager in a world crushing dystopian society and you’re just trying to get your job done. To do so you use your worker dice, and the larger the dice score, the better the dice works in some jobs, but if they get too smart then they will make a break for it and run away from the dystopia. Then you gotta go spend resources to birth new ones from the tanks.

The game also includes 4 factions. All buildings have allegiance to one of them, and every time you use that building it gives that faction a bit more strength. Players have employee cards, active employees give let you use those buildings more efficiently.

Your goal is to put down all your stars onto the board. Doing so requires building markets and trading ancient artifacts (from before the dystopia) in those markets. Getting those requires trading with the Icarites (the sky faction), or digging tunnels to allow some factions to steal from the other factions.

The final tier of dynamics are the aforementioned markets. They’re set up so that not all players can contribute to their building, and once they’re build everyone who didn’t help has a new game rule placed against them until they donate some artifacts.

You should start to get a feel that this game includes lots of things happening at once, and I can vouch that that’s in a good way. First you want to ride the balance of most use you can get out of your workers without losing them cause they got too smart. Second, you want to advance the factions you have employees with so that they get more efficient. Third, you want to make sure to be early to contribute to market building so that you don’t get locked out of game benefits, again for maximum efficiency.

There’s also a nice natural narrative to the game, sort of nice in a euro. The game goes through distinct phases of first general start and dice generation, then a digging/building phase, and then a maximum speed race for artifact generation and placement. This helps prevent the sensation of sameness that occurs with some euro games.

Euphoria, top down
Euphoria, top down

I genuinely like this game. Great theme, great art, great tokens, and most importantly genuinely interesting gameplay. My only quibble is that the box and board feel just a bit light and it makes me a bit worried whenever I move it, but it’s a really really minor concern and in no way makes pulls me away from recommending this game fully.

Posted on

Game of Thrones, the Board Game, review

A classic, for good reason. It’s the area control game that gave backstabbing a good name. Often imitated, never duplicated.

Game of Thrones, I own the blade and the Raven. Good signs.
Game of Thrones, I own the blade, the Raven, and have ships over the east coast. Good signs.

Ameri-gaming done right: start by taking an interesting property, put a slim game on top of it, then make sure that the game ‘feels’ like the property. And that’s the secret of it’s success: GoT:tBG feels right. The backstabbing (sometimes proper invasion, more often in the form of denying support that was promised) is woven integrally into the game. The combat is deterministic which makes the backstabs sting even more as you went through and counted all the math and you have had enough if not for those awful Tyrrels.

The game’s victory condition is holding enough castles. The temporary nature of this means the game is a constant dynamic rotation of who everyone else is ganging up on. The dream is to be 3 castles behind as everyone else fights each other, then to leap across the map on a chain of boats and grabbing 3 in one turn.

Ah, the boats.

Top down GoT
Top down GoT, my boats on right won me the game

The boats allow armies to teleport across the map. They prevent the game from becoming a slow bottleneck of fighting over choke points and allow for grand master moves like distributing 3 armies across 3 different landing zones. Don’t underestimate them or you’ll find yourself with foreigners at your shores sieging your muster spots.

The game’s narrative is driven by power tokens. Instead of using a region militarily, it can be harvested for that abstract currency. That’s then used to help fight against the periodic Wildling attacks and to bid on the 3 tracks which functionally represent turn order, military power, and flexibility with orders respectively. Winning the top tracks gives added benefits: the Throne breaks ties in the other appointments and decides how some event cards play out, the Valarian Steel gives +1 to a combat, and the Raven swaps two orders in response to them displaying, letting the owner respond to someone doing something unexpected.

End of game. Baratheon rode from King's Landing to outside Winterfell
End of game. Baratheon rode from King’s Landing to outside Winterfell

It’s a great game with a caveat: it’s kinda off-balance at 4 and 5 players. Highly recommended at 3 and 6 players though.

Really, my only ‘complaint’ of any sort is that the characters don’t look enough like their TV show versions that I’ve gotten used to. Makes since the game predates it by quite a bit, and to be honest the hand painted art is much nicer than any film stills could be.