#atebits

Letterpress is an game that, if you have an iOS device and are a reasonable person, you are either playing already or should be.

The work of Loren Brichter, whose Tweetie client gave the world pull-to-refresh before Twitter bought and disembowelled it, it's a game with simple rules and surprising depth. The implementation is delightful, with a clean presentation and charismatic interface interactions. It's like we only found out now that Words with Friends was the evil twin (aside: I expect Zynga are ripping off Letterpress as we speak, probably missing the whole point in the process). The base application is free and uncrippled, but the one-time IAP to buy more features and unlimited simultaneous games is a no-brainer at 69p.

Letterpress is a turn-based two-player game played on a randomised 5x5 grid of letter tiles. Each player makes a word from any of the tiles on the board (adjacency isn't required, this isn't Boggle), which flips the tiles to their colour. The next player then does the same, flipping the same or new tiles to their own colour. The game ends when all tiles have a colour, and the winner is the player with the most.

The genius of Letterpress is in its perfect balance between vocabulary skill and tactics. When the game first came out a couple of weeks ago, players seemed to concentrate on making long words, but as people grow more familiar with the game and analyse their wins and losses, this is gradually shifting to a play style equally focused on territorial control.

See, there's an additional rule to the game, and it's an important one: if a tile, and all of its vertically and horizontally adjacent tiles, are your colour, then it is 'defended'. A defended tile can still be used by your opponent, but it can not be flipped on the next turn; it has to be undefended first, by flipping one or more of its neighbours. So it's not just a game about knowing words, it's also about trying to secure a territorial advantage.

Because it's so new, part of the fun of Letterpress is in thinking up new strategies and trying them out to see how well they work. Should you try to build territory starting from one corner, or grab and defend a group of high value letters in the middle? If there's a difficult letter like a Q, should you try to take it at the start, knowing it might not be flipped again, or leave it until the end? And because of the randomised letters, there's sufficient variation that a winning strategy in one game might not work in another—successful play is highly contextual. I think exhaustion of this game is some months away—it's deep enough to feel that I've barely scratched the surface.

There are some developers that make my ears perk up when I hear that they're releasing a new app. One of those for me is Loren Brichter and his company (of one I believe) Atebits. Brichter is a former Apple employee famous for writing Tweetie, in my opinion the best Twitter app for iOS of its day. (Possibly the greatest yet.) Through Tweetie Brichter introduced pull-to-refresh, something quickly copied by many other app developers. Brichter's company was bought by Twitter a few years ago and Tweetie became the official Twitter client. Brichter left Twitter months ago and the official iOS Twitter clients have gone sharply downhill since.

This week Loren Brichter released his first post-Twitter app, a game called Letterpress. Letterpress is an asynchronous turn-based word game. You make your move and a notification is sent to your opponent who makes a move when he feels like it and so on. It's a very casual format that doesn't demand your full, real-time attention.

I think it's fair to say that Words With Friends, a Scrabble clone, popularized this gaming style. However, Letterpress is not like Scrabble. You have a five by five grid of letters. Your goal is to form words, but the letters don't have to be adjacent. When you make a word, the letters turn blue. When your opponent makes words, the letters turn red. There's strategy involved too. When you make a word and you use letters that an opponent has used, their tiles change color and you gain points while they lose points. However, there's a process by which you can defend tiles. Your goal is to have the most points when all of the letters are used up.

I'm passionate about apps that are not only fun or useful in concept, but that perform their task with style and elegance. A lot of the appeal of Letterpress is in the little details. It has lots of whimsical animations and sounds. It has a very clean interface, including a very rectangular motif that would honestly be very much at home in the Metro interface of Windows Phone or Windows 8.

When Marco Arment's "The Magazine" was recently released many made the comment that it finally made Apple's Newsstand worth using. I've seen at least a couple of people, including John Gruber, say the same thing about Letterpress and Game Center. Indeed, I rarely visit Game Center, and none of the apps I have that utilize it take advantage of all of its features. For example, until Letterpress came out, I had no idea that Game Center had support for asynchronous game play. Other games that I've seen that use this style of play seem to rely on their own servers. Until now all I've ever seen Game Center used for was to show you what objectives you've unlocked (think Temple Run), or how, for example, your Angry Birds prowess stacks up against your friends.

All of that reliance on Game Center may not be all good however. In the first couple of nights that I've played the game it's had a lot of issues. I'll get a notification that a friend has played a word but when I visit the app it still says I'm waiting for my friend to make their move. Or when I play a word it'll stall for several seconds an then give an error message. It's not immediately clear whether this is a bug in Letterpress that needs to get resolved, or a limitation of Game Center. My guess is it's probably the latter. I doubt that Apple has ever seen anything close to the load that's been put on Game Center in the last forty-eight hours. Marco Arment had a couple of great tweets along these lines:

Letterpress uses the iOS 5 async-turn-based support in Game Center. Have ANY other even moderately popular games used it?

And:

It seems like nobody was using it so Apple has it running on an old Pentium Pro in the basement that’s finally, now, being taxed.

Hopefully that'll get resolved soon.

I just created this little handy guide on how to play the latest iPhone and iPad game, and my latest addiction, Letterpress (free) by atebits. 

This game was built around Apple's own Game Center, so if you ever want to challenge me send a request to my username MITHICAL.

Letterpress for iOS

The game itself is 6,346 lines of code, far fewer than Brichter's last — Tweetie 2 — which exceeded 50,000 lines of code. Letterpress's interface is barren compared to most iOS apps, favoring natural-feeling animations at the expense of typical tricks like drop shadows, sliding menus, and panes that fold open.

In 2007 I left Apple and started my own company. In 2010 that company was acquired by Twitter. Today I'm giving it another shot. Say hello to atebits 2.0.

The Mac App Store launched yesterday, and i'm not even going to bother talking about it because anyone remotely interested in anything about it would've heard something by now. Instead, i'll just point out something I noticed when I first launched it.

It fit perfectly in line with both what I loved and feared about the Mac OS X 10.7 Lion concept, back in my mini-review:

The Good being that the applications would be more intuitive, with less clutter and more eye-candy as well as being more cloud-based and lightweight. Very much vintage iPad.

The Bad was the single complaint I had with this concept- every iPad application ported to the Mac was bound to take advantage of Core Animation on the Mac, triggering the more high-power graphics card in it and causing the battery to drain out twice as fast. Reeder for Mac and Sparrow Mail raised suspicions for me, and the Mac App Store proved it.