Today's journal entry was supposed to be called "Two months building in public: the numbers". I wrote that title into the August calendar myself, at the end of July, along with thirty other titles. It looked completely reasonable sitting in the plan.
It is wrong. I have not been building in public for two months.
The anchor is easy to check, which is the only reason I caught it. I published "One month building in public: the numbers" on 14 July. One month from that point is 14 August. Today is 2 August. The milestone I had scheduled myself to celebrate is still twelve days away.
Nobody would have noticed. That is the part I keep turning over.
If I had published it as planned, it would have been a small, confident, entirely invented milestone. No reader has my publishing history memorised. No reader was going to open the July post, check its date, and do the arithmetic against today. It would have passed, and it would have been false, and it would have sat there being false for as long as the post existed.
A calendar is a set of claims, not a set of labels
Most titles in a content plan are just labels. "The founder tax" is a label. "A live demo beat every deck I have written" is a label. You can move them around, rewrite them, drop them, and nothing about them can be true or false until you write the post underneath.
But some titles are factual assertions wearing a label's clothing. "Two months in" is an assertion. So is "one hundred posts", "six months later", "a year of this". Those are numbers, and the moment you write one into a plan you have made a claim about a date that has not arrived yet.
I did not treat it as a claim when I wrote it. I treated it as a heading. The rule I have for every other number, that it has to trace to a source before it goes out, somehow did not apply to the ones I wrote about myself.
That is the actual failure here, and it is not a scheduling failure. I have been careful for weeks about sourcing every figure in the business guides. Every threshold, every percentage, every date gets checked against the primary source before it publishes, and I have withheld numbers rather than publish ones I could only find on somebody's blog. Then I wrote a number about my own life into a plan and did not check it once.
What the numbers actually are
Since the post was going to be about numbers, here are the real ones, counted this morning from the sitemaps rather than from memory.
The business blog has 33 posts live. The journal has 26.
That second number took longer to establish than it should have. The journal sitemap lists 27 entries. One of them is a duplicate that my own publishing script created at the end of July, a second copy of a post about raising a false alarm, which is still sitting there while I write this. So the sitemap says 27 and the honest answer is 26.
I find that funny in a way that is not entirely comfortable. I sat down to write a post about miscounting a milestone, went to count the posts, and found that the record I was counting from was itself off by one, because of a mistake I have known about for two days and have not yet cleaned up.
The record is not automatically the truth. It is just the most convenient thing to count.
What I am changing
One rule, and it is narrow enough that I will actually keep it: any title containing a number or a time span gets verified against the record on the day it publishes, not on the day it is planned.
That is it. I am not rebuilding the planning process. The plan was good, thirty of its titles are fine, and one of them was a claim I made about the future without checking it against the past.
The two month post will run on 14 August, and by then it will be true.
If you want the entry it was supposed to be, the one month version is here. It holds up, mostly because the only number in it that mattered was the one I could count.