Game Notation

I thought of something interesting recently. It's pretty clear by now that just about everyone likes fenrislorsrai's ruleset, and the ruleset is a pretty nice system. The way that game setups grow is intriguing to me. And so that led to this question -- can anyone devise a notation that we could use for transcribing actual played games into text, like chess notation does for chess? I've got some ideas, but none that could be easily taught to 3rd graders. If I work out something that I like especially much, I'll post it.

The main reason that I'm bringing this up is that not only would it be useful, but it also seems like an absolute necessity when discussing game balance -- communicating example games, for instance, or discussing potential "overpowered" setups or card juxtapositions.

This seems like a delightfully tricky problem. Anyone have any clever ideas?

Hm, tricky. Notation works readily for chess because there's a limited amount of pieces: with Phylo, you'd have to avoid abbreviation overlap with other cards.

You'd have to indicate the positions of cards relative to others for notation to work properly. So maybe something like: [Card Type][Card ID in letters or numbers][Position played relative to the following card][Card(s) this card was played next to]. This might look like: "s-bea-ne-h-cit", or "species: beaver played northeast of habitat: city.

Of course, you have to decide if your cardinal directions are relative to the player who goes first, or the player who's turn it happens to be. It's more complex than chess, but it can work.

I was originally thinking about describing the game in terms of a decentralized network where each card has a certain number of "slots" that can be filled by a new play, and then working out the notation as a series of operations on the initial state, again reminiscent of chess notation, which simply mark what card is being played and what it's connected to. The issue I'm running into is finding a way to specify the individual card(s) that new plays connect to.

One thought for identification is to use a Cartesian coordinate system with the central starting card as 0,0 -- but orientation is tricky.

Anyone else have thoughts? Any other ways to do things?

Wootfish wrote:
One thought for identification is to use a Cartesian coordinate system with the central starting card as 0,0 -- but orientation is tricky.

This would be my idea, too.
The starting player is player 1, his side is the negative vertical axis, everything on the left site of the starting card, from his POV, is negative. Every card has exact coordinates, they are (n/2 | m/2) with n and m element of Z. (the /2 because of only half touching cards).

With the prevalance of cameras, a few snapshots of the position are probably going to be the best method of reporting until we get a standard position method.

Playtest reposts are REALLY important and I hope to have some after tomorrow's game meet.

On slight side jaunt, Phylo now has a entry on BoardGameGeek and I'll probably type up the playtest reports for the Sessions section of the entry and include pictures of the positions.

Phylo's entry:
http://www.boardgamegeek.com/boardgame/71594/phylo

Yeah, thre's nothing there right now. It JUST got added to the database this week.

I'm inclined to post them over there and link here because the reports automatically hit BGG's front page and subdomain forums so promote the game. (it also generates more Momentum on Google since BGG has a WAY higher page rank than this forum)

To see what it does look like with actual content, here's the sessions section for a DIFFERENT game I'm currently playtesting:
http://boardgamegeek.com/forum/8007/rage/sessions

The top three entries are all playtesting the same card set, but in 3 different deck designs. If you read them in order, you can see I refer to earlier reports on some cards.

That gives people a vague idea of what sort of thing we're looking for in a playtest report.


And I cannot stress this enough:
The most valuable playtest report is that you used a card and it was fun and you had no questions about how it worked!

This lets people know make more like this. If there's no reports that are good, it looks like the cards aren't fun and aren't being used.

Snapshots of positions will do, but to actually annotate how a game develops, you'd need a snapshot of each move -- and that could easily take quite a few pictures, making it rather inefficient.

As an initial suggestion for notation, how about just listing the two half-card points on the cartesian grid that each card covers? The order they are listed in could signify the orientation of the card -- put the cell that the top of the card covers first, say. To identify what card is being played, we could just name it. We'd have to use scientific names, not common names, as sometimes multiple species have the same common name.

As an added bonus, using the Latin names encourages memorization of them, without forcing it -- people can still Google the names if they're unsure what's what. Additionally, adding card look-up by common or latin name to the site might be a good idea.

I agree Cartesian coordinates would be the most reasonable way to note location. For the cards, we just need to find the easiest and most efficient way of sorting them. It wouldn't be ideal, but we could easily attach a number to each unique card, basically a bar code, for identification.

ID numbers wouldn't work. If they are actually unique to individual cards, they would have no significance to someone who's trying to play through the game with their own cards. Simply giving the scientific name or (in the case of habitat/event cards) title of the card being played and the location where it's being played, if applicable, ought to be enough, I think.

If we have different expansion sets each one could have a three letters code.
Every card could then be identified with this three letters and three numbers.

But wouldn't that be even more confusing than just using scientific names?