I really love animals and how many different species there are. When you started talking about Bloomburrow's mechanics not being directly typal, I was hoping that there'd be a lot of one-off animals. But almost all of the cards revealed so far are one of the main 10 animals, with the main exceptions being the calamity beasts.Was there ever a point in development where the focus was more on showcasing a lot of different animal species rather than selecting a smaller number that act as draft archetype groups?

I talked about this in my first Bloomburrow preview column. Here’s a snippet:“Once it was on the schedule, I did a little advance work on the genre to familiarize myself with it. I realized that there were two ways it’s traditionally done.Take #1 – Animals represent groups of people. These people are mice, those people are badgers, and these people are otters. Each animal type has qualities that are consistent among that group, usually things that feel resonant with the real-world animal. In this version, the setting is usually a biome, and all the animals in it are ones who would live in that biome. The animals are roughly proportional to what they would be in the real world.Take #2 – Animals represent individual people. This person’s jumpy, so she’s a frog. That person’s sneaky, so he’s a fox. This other person rushes into things, so they’re a rhino. Each animal is used to represent personality qualities. In this version, the setting is usually something more human in structure, often a city, and the variety of animals is much larger. The animal selection here is not limited by biome, so you can have animals living together that normally would never see each other in the real world. The animals are loosely related in size (a racoon is smaller than an elephant), but the scope of scale is compressed.Take number one is easier for worldbuilding. There are less unique types of animals, and they’re organized by creature type. Because animals are used to express groups of people, they tend to act more similarly to traditional species creature types, like Elves, Goblins, or Merfolk. This pushes us more toward a factioned typal theme.Take number two is easier for design because the designers have access to a lot more animals and can make more individually cool designs. The twelfth Mouse card, for instance, is a lot harder to make different than the first Giraffe. This approach pushes us more toward mechanics that tie into a larger animal theme. It’s more likely we’d create an environment that was about a lot of different animals working together, putting the focus more on individual top-down card design.Aaron was more interested in doing take number one, while I was more interested in doing take number two. So, we did a bunch of market research. It came back exactly even. Half the people we polled preferred take one, and half preferred take two. In a tie, Aaron’s original vision won out, so we did take one. (Also, I believe more people internally wanted to do take one.) I do want to stress that both takes would have allowed us to make a cool set. They just head down different paths and would have ended up in very different places, mechanically and creatively.”

Norin the Wary from Time Spiral and Saltskitter from Future Sight have incredibly similar designs that we haven't seen anywhere else in Magic. Do you know if those designs being in the same block was a coincidence, or is there a story there?Thanks Mark!

I designed both, so there’s a chance they’re subconsciously connected. Norin the Wary was a top-down design to match the character. Saltskitter was just me making something that felt weird.

You said Duskmourn started as 70/80s horror, but also said it's a bottom up set. I thought top down/bottom up was the thing that kicked off the initial idea for a set, and I'd think trying to capture the flavor of 70/80s horror would put Duskmourn in the top down camp. Can you explain the distinction and what makes Duskmourn bottom up?

The difference between top down and bottom up is how is the set structured. I’ll use Ravnica and Innistrad as examples. If you took off all the names, art, creature types (replaced with non flavorable words - “Creature Type A”), and flavor text from both sets and showed them to a player. Ravnica would make sense. It’s core structure is mechanical. You might not get all the flavor of the guilds, but you understand how it’s put together.Innistrad, in contrast, will be harder to understand how the component pieces come together. It will feel more random, because the connective tissue to the structure is the flavor.The idea of top-down and bottom-up was much cleaner back in the day, but we’ve gotten so good combing mechanics and flavor, it’s hard to tell.It’s too early to talk about Duskmourn, so I’ll use Bloomburrow as my example. We started with a flavor, anthropomorphic animals, but quickly decided that having ten two-color pairs, each one a different animal was a great core to the set structure. That structure is mechanical, so technically it’s bottom-up.Let me end by stressing that top-down vs. bottom up in a world where we blend mechanics and flavor so much doesn’t have as much meaning as it once did, except if you’re really in to design.

Got curious and decided to ask, which is more likely: Universes Beyond product for Wizard of Oz or a set featuring top down designs based on Oz?

I don’t understand how those are different.

We already combine mechanics and flavor in every set we do. Yes but you usually take one and use it as a base for the other (hence top down or bottom up) whereas this would mean you take both together as a starting point to create something you would never think of starting from only one of them.

Our design process has us going back and forth, with mechanics influencing flavor decisions and flavor influencing mechanic decisions, so what you’re talking about happens basically every set.Top-down vs. bottom-up is just a technical term talking about how the set is structured. They are two different approaches that work differently from each other. You’re asking for us to make a number that’s even and odd. The good news is if we do our job well, you can’t tell where we started from.

Hi Mark, previously you've said that if R&D does their job well, players shouldn't be able to tell if a set was built from top-down or bottom-up. I think I recall a Reddit discussion where most players said they actually can tell. Wanna do a poll here?

I want to ask a few questions, so I’ll do them on the main site.

"Innistrad is one of my best designs."Yeah, Innistrad was great when it was unique, and gothic horror is an incredibly deep well to draw from. There's tons of source material, and it's easy to make it feel "universal".These days, it feels like every premier set is trying to be Innistrad, unless it's a Storyline Set or a Return Set. The Innistrad-likes all feel like "Magic does [genre]" instead of something actually original, Storyline Sets are practically Return Sets, and the actual Return Sets are half Innistrad-likes anyway.The Modern Horizons sets are the last bastion of bottom-up, Magic doing Magic sets, but they're infrequent, expensive, and overshadowed by how they destroy the Modern metagame every time.I just want a new world that makes me feel like I'm reading the Planeswalker's Guide to Alara again. A world that only makes sense in Magic the Gathering, instead of Magic fitting itself into something that already exists.

I want to begin by stressing something. You’re talking to me, the Head Designer of Magic. I, along with rest of the designers, are in charge of the mechanical execution of design. We create the larger structure of how the set is build, weave in flavorful themes, and design mechanics to play into that structure and theme(s). When I talk about sets I’ve done I’m talking about the part of the process I’m in charge of. With that said, let’s talk about the latest design I led (vision design), Outlaws of Thunder Junction. I hold up its game design to any other set I’ve lead. The mechanics are sharp, useful, and lead towards new play patterns. The limited play and constructed play involving its themes are novel. And while it’s all forward facing, it was designed to be more backwards compatible than previous sets. Plot, in particular, is one of the most exciting new mechanics I’ve designed in a long time. It’s simple in description, but surprisingly deep in strategy. Having the primary focal point of villainy, with a secondary theme of the Western genre, led my team to create something new. Top-down is just an input. Good design can come from any input. Bottom-up design is no better at making good mechanic designs than top-down. I believe current mechanical design is on point, and Outlaws of Thunder Junction, in particular, is my team being on their A game.

Now that we've had a year of sets that explored omen paths as a plot device and an excuse to feature legendary creatures on planes they didn't originate from, how do you feel they affect the set mechanics? And more specifically, do you feel they have a noticeable affect on top-down design? (like for instance, the cowboy-themed set featuring many characters from Magic's past and attempting to stay true to their characters while working in those cowboy-themed set mechanics)

It’s a useful tool to make sets we previously couldn’t.How do you all feel about the impact of the omenpaths?

Hi Mark! What is the origin of phrases "bottom-up" or "top-down" in relation to set design? is it because the first thing you see is flavor (name, art etc) and then you see mechanic?

Actually, I think the terms didn’t start with Magic design.

as someone whoae a big fan of Spaghetti Westerns i wanna say how much i love OTJ so much!!!were theyre any movie references or such you couldnt fir in?.Like the Man with No Name etc

As is usual in a top-down set, not everything fits. I don’t know of any examples off the top of my head though.