The GDS2 multiple choice test
I didn't do so well at the GDS2 multiple-choice test... Although I am not competing, I had looked at the test before, and today I made a (non-serious) attempt to answer the questions. For the record, my score was 30/50. After checking the answers, it turns out that most of the incorrect replies are due to a few major reasons:
1. I am mostly interested in mechanics, color pie, that kind of stuff. I am not interested at all in a card's rarity, or what the development team does and doesn't do at certain rarities. There were several questions asking exactly that, so I would have failed those no matter what.
2. I sometimes didn't understand the question or read it wrong. For example, #30. I mistakenly interpreted that as, designing a set around the Wizard of Oz, rather than creating a deck of existing cards that somehow seem related to it.
3. I didn't do any number crunching. I could have easily done so with magicquery *cough*, but like I said, this was a non-serious attempt. This meant I was sometimes just guessing at how often a mechanism would appear in a color (green has vigilance?). I also didn't feel like combing through the card database in order to find the cycles in #49.
By comparison, the guy who got all 50 questions right, Max McCall, apparently spent 12 hours on this test. 1
Interesting tid-bit: MaRo unintentionally hints at the limitations of Gatherer in one of the explanations (question #35):
This is just a number-crunching question. You had to be careful in your Gatherer search, though, because if you just search for "gain life" in rules text and "sorcery" you'll find a lot of black cards that drain life from creatures and/or players. The majority of actual life-gain spells are instants. Token making, on the other hand, exists mostly on sorceries.
Admittedly, the current version of magicquery cannot answer this any better than Gatherer. I am currently in the process of adding keywords, one set at a time (they're not available for download anywhere yet). Once that is done, I am considering adding "tags" to cards, which can be used to get more information about them. For example (just brainstorming here):
- #painland to indicate lands that do damage to you
- #ping for cards that do 1 damage
- #cantrip for cards that let you draw a card as a side effect
- #lifegain for cards that, well, make you gain life
- #come-into-play for cards with come into play effects (now called "enters the battlefield")
- #power9 as an example of a "group" of cards; cycles could have their own tags, so they are easily found
And so on. Of course, this is a lot of work, and the number of tags that could be used is potentially very large. It would give magicquery a serious edge over other card databases though, and not just for programmers.
One thing magicquery would be capable of right now, is answering questions like #26. It would require writing a small script, but it's perfectly possible to calculate how many enchantments are in each set. (A more interesting problem is, how to make this functionality available without requiring any programming.)
~
1 FYI, his GDS2 card submissions can be found here. I especially like card #8, Omniscience.