Yes, we are sloppy

We created this site in just a few days of prompting. Really we were bitter because we had spent years working on APIblaze.com and it got completely lost in the slop, even worse, because we posted, we began receiving TONs of slop from bots offering their suckage to use on our suckage. Hilarious. So we thought, wouldn't it be nice if we could distinguish what is pure slop and what isn't. Of course, we'll get how to grade wrong. What makes something sloppy or its opposite (crafty) is up for debate and improvement. For now, at least, this site grades itself by its own rules. Which means we are openly slop too.

What the number says

Two things are measured, each out of 100, and the rules are public and generated from the code.

Quality asks whether this is real, maintained work by more than one person over time. It reads the github repository for signals you cannot buy: how many of the last 52 weeks had a commit, scaled by how old the repo is; how many other humans have put in two or more commits; whether anyone has ever filed an issue. Stars, forks, badges and READMEs are not used, because they are the first thing a sloppy project fakes. But admittedly, someone could submit the wrong repo for a website, at least they have to own it and it'd be hard to fake years of improvements and collaboration, but not impossible.

Popularity asks whether the people who find it actually use it. It reads the owner's own Google Analytics: active users in the last 30 days, and how many came back. Under a hundred users is zero reach, however loyal those users are, because from the outside "nobody there" and "nobody there yet" look the same.

Average the two, subtract from 100, and you have how sloppy it is.

How to argue with it

If you think a rule is wrong, the rule is a pull request away. Every grade can be appealed in public, and an appeal that points at a rule misfiring gets the rule fixed for everyone, never a private exception. That is the only kind of fairness we know how to make.

Watch the number.