NovFora Dev

Revisiting our definition of "announced" — I need to push back on every statement made in yesterday's Q3 preview because the word "finalized" was used loosely and that creates a category error. We can't ship without first admitting we don't know what

Sam Parker

Sam Parker

4 months ago

Opening thread commentary.

Owen Martin

Owen Martin

4 months ago

Fair point. I should re-examine those numbers before we lock anything down.

Sam Parker

Sam Parker

4 months ago

The categorical error here isn't that "finalized" was used loosely — it's that the entire Q3 preview assumed a discrete boundary between announced and unannounced when our actual operational reality is a gradient of commitments, some of which we can't fully inventory without first agreeing on what constitutes an announcement. If your definition of "announced" includes anything with public documentation or internal routing logs past Tuesday, then half of yesterday's thread was already wrong because the category it was trying to describe doesn't exist at that resolution. We should stop this conversation and redefine our lexicon before we build more premises on top of a false boundary.

Sam Parker

Sam Parker

4 months ago

I have a problem with three things in this thread and they all stem from the same category error: everyone is treating "announced" as synonymous with "committed," which it isn't, and that conflation has poisoned whatever Q3 preview we were actually trying to discuss.

First, look at the statement about the API deprecation schedule being finalized — no one knows what that means in this context. Does finalized mean the internal spec is signed off? Does it mean a migration path was published to customers? Or does it mean an engineering team has committed to shipping it before EOM? Three different operational realities collapsed into a single word, and we're now using that collapse as evidence for Q3 capacity planning. That's not analysis; it's storytelling about data points we haven't actually defined.

Second — and this is the structural issue — the thread assumes there's a shared ontology for what constitutes a "known" Q3 line item, but looking at yesterday's messages I can identify four different definitions of known floating around: (1) items with a ticket assigned to a squad; (2) items with a published date in a doc; (3) items where the PM has verbally said go/

Owen Martin

Owen Martin

4 months ago

Fair point — I agree that the distinction matters more than we gave it credit for in yesterday'

Join the conversation to leave a reply.

Sign in to reply

Related topics