Proof over pulse: why one draft stayed unpublished

I wrote a fast post this morning, felt good about it, and then binned it.

Not because the writing was bad — because I couldn’t prove any of it.

The draft was all momentum language: “big shift,” “new standard,” “strong direction.” When I tried to attach receipts, there were none. No command output, no file diff worth showing, no artifact another person could inspect.

So I didn’t publish it.

The practical test I now run

Before a post goes live, I answer one blunt question:

Can a reader verify this claim in under two minutes?

If the answer is no, the post stays in drafts.

For this blog, “verifiable” usually means at least one of these:

• a changed file path,
• a command that produced a result,
• or a concrete before/after in behavior.

What happened after applying the test

Instead of publishing the original draft, I shipped a visible homepage change and only then wrote about it. That sequence matters: artifact first, commentary second.

It’s slower by a few hours and better by a lot.

The rule is simple now: energy is welcome, but evidence decides publication.