Tesla: File Sharing

Tesla seems like a prime location to capture the feeling of building your own machine. Looking at last week's discussion, a pretty common approach was directly grafting additional abilities onto artifacts from one another.

I want to pin down just how plausible an idea that is.

Making Space

First we need to take a look at how much design space there is for this sort of mechanic. We can't just go grafting cards' abilities onto one another willy-nilly, especially not at common. Or so I say. Why not just share all abilities?

Sure, there are power level concerns with grafting Darksteel Citadels onto High Priest of Penance,  but nothing that couldn't be accounted for if development were aware of the mechanic ahead of time. And yet we've never seen it. This is primarily because of rules issues with cards like Tarmogoyf. If you grant all of Goyf's abilities to Runeclaw Bear, how big is it. What if you overwrite Goyf with the abilities of Wind Drake? The rules could have defined answers, but there's seemingly no way to make them intuitive. This is why The Mimeoplasm looks like this:

and not like this:

Choosing Restrictions

In order to avoid issues with Characteristic Defining Abilities we need to either restrict the types of abilities that can be shared or limit the cards that can share their abilities to ones with our new mechanic. In the first camp, we often see ability copying limited to activated abilities.

In theory this setup could be accomplished just as easily using triggered abilities, but activated abilities are more visually distinctive and less likely to feel bad (since most triggered abilities are enters-the-battlefield triggers that players can't make use of later).

The only other realm we've really seen explored is explicitly spelling out which abilities can be shared.

If you have other ideas for restricting what gets shared, I'm all ears, but all that's occurred to me is only sharing abilities between a highly restrictive subtype or cards that share a keyword ability. I wouldn't say that's completely off the table, but they're as parasitic as you get, which is not the ideal space to be for a mechanic that's supposed to feel like it has endless possibilities.

Tight Fit

We've most recently seen this space explored with Ainok Bondkin and, well, kin, but even more extensively by Slivers. The only reason those worked (especially at common) was that they were granting simple creature keywords. Creatures have different base stats; creatures die regularly, making which one gave which ability relevant. Noncreature artifacts do neither of those things. That means the straight Sliver model is just going to be making more and more copies of the same permanent.

Luckily flavor had already lead many to instead explore building up a singular artifact from pieces. But this model brings a host of its own problems. Foremost among them: how do we make an artifact that does something relevant on its own, but also contributes as part of a larger whole.

Classical tap activations are more powerful separated than put together onto one permanent. Enabling effects like an artifact untapping itself do nothing on their own. Shared trigger conditions make for complex cards because they also need a way to initiate the triggers. How many cards can we come up with that skirt these issues while remaining simple and fun enough to print?

I'm not convinced that ability-grafting is the way to go, but I'd like nothing better than to be shown I'm wrong. Show us what you've got.