Me: "I'm making serious progress on Badger."
Some other guy: "Excellent. You should probably start thinking about how to integrate it with Echidna."
M: "That won't be hard."
SOG: "You think so?"
M: "Yeah, it's just a bit of glue code, a little hacking, no sweat."
SOG: "I think you're underestimating it. I mean, how are you going to get it in the right format?"
M: "Well, it's just a simple conversion. And the indexes exist."
SOG: "They're not set up to be used in this way."
M: "So we'll set them up so we can use them this way. It's not a problem."
SOG: "But it will take a lot of work."
M: "Well . . . yeah . . . but, I mean, we already know it's theoretically possible, right? I mean, we're just talking a few searches, a few data format changes. Nothing that needs to be invented."
SOG: ". . . And?"
M: ". . . If I can prove it's possible, the hard part is already done. The only question is how much time it'll take."
I mean, seriously. If I know it's possible, I can do it. The only tricky part comes when I don't know it's possible. And in the worst case, all I'd have to do is rewrite our entire database backend. Which would suck, but it won't be necessary, and it would clearly work, and therefore it's possible.
And therefore, while potentially time-consuming, easy.