Conversation

A parallel to “perfect is the enemy of good”: “exists is the enemy of good”.

https://seancoates.com/blogs/exists-is-the-enemy-of-good

2
4
5

My latest hobbyhorse of this is markdown. It’s not bad!!! But how can you justify something better given that it’s incompatible?

RT @alpha A parallel to “perfect is the enemy of good”: “exists is the enemy of good”.

https://seancoates.com/blogs/exists-is-the-enemy-of-good

1
1
1

@simrob I’m having this exact argument about https://githubnext.com/projects/rtgh (which uses a rich text format); people are very attached to Markdown, and to the way things already are

1
0
0

@jakedonham if you want to talk about my experience designing (w/ Chris) and deploying Camperdown at Brilliant I’d be game - Camperdown is nominally (legally but not usefully) open sourced, it’s not perfect but I think it pulls off “better than markdown” - I’m having to restrain myself from using it on the next gen of Twelf Wiki because damnit I have to use markdown there to keep shit simple and forward maintainable

2
0
0

@alpha this is amazing. Thank you.

0
0
1

@simrob @jakedonham I think what's missing in "exists is the enemy of the good" is what "good" actually means; I reckon that in the mind of the person saying the phrase, they do have specific attributes of the solution that they care about. Like for markdown, idk, markdown is "not good" in that it's got weird corner cases or whatever. A "cleaner"-designed markup language might lack such corner cases.

1
0
0

@simrob @jakedonham but "exists" is a kind of "good", also, I'd argue. "simple and forward maintainable" is a kind of "good". "there exists a community of people already familiar with it" is a kind of "good".

So "exists is the enemy of the good" is observing that "good₁ is in tension with good₂", which is *absolutely* true, but I don't think it's anything to wring our hands over.

1
0
0

@simrob @jakedonham I agree that it is possible to be in local-not-global-optima states where you're reluctant to tear down the bridge to build a better one, but the negative utils brought about by *not having a bridge* while the new one is being constructed are *real*.

1
0
0

@simrob @jakedonham tho maybe I've been traumatized too many times by people claiming to build better bridges and running way over schedule and delivering bridges that actually weren't obviously better...

2
0
0

@simrob @jakedonham (at the end of the day I doubt I actually disagree much with OOP. He's pointing out one extreme that should be avoided, and I'm merely contemplating the opposite extreme, and reckoning it should be avoided)

0
0
0

@jcreed @jakedonham It's a really good point. I don't *know* if JS would have been better if Eich had had another week, for example.

The purest example of this in my mind is the cognitive overhead of the way that makefiles depend on tabs in an essential way. that could have been avoided with an investment of retraining and effort many decades ago - we could have done better and they **knew that at the time** and now lots is worse for lots of people forever

0
0
0

@simrob this looks cool, is it what drives the scripts on brilliant?

1
0
0

@jakedonham yeah, the interactive bits are Elm mostly but all the text is Camperdown as well as like multiple choice question config

0
0
0