What's Phil Mitchell doing behind the partition at the back?

This one from Berlin airport is my favourite… and it doesn't even point to a web URL!

Thanks for the reply. Looks like the opening `{` was missed from the Customer.io unsubscribe link. No need to remove me manually, thanks.

@1030.bsky.social

Our "Urine-free kettle guarantee" sign has people asking a lot of questions already answered by the sign.

Meanwhile, here's the diff-of-the-day for our Amplitude integration…

Anyway… the good news is that we can ignore the docs: the deletion API *does* in fact accept this newly acquired definitely-not-an-integer `id`.

Of course, it returns HTTP 202 rather than the documented HTTP 204 — but it works! 👍

At this point, I'm taking the docs with some big grains of salt.

The deletion API should return HTTP 204 with no response body. So why can I pass a `format` parameter to choose between JSON or CSV? 🤔

(requesting a CSV response will earn you an HTTP 500)

posthog.com/docs/api/per...

Supposedly I can only look up a single user if I know their integer `id` value. But it's possible to search by `distinct_id`, which will return the number I need! 🥳

However, the response has the `uuid` string that the docs mention (but don't explain), along with the EXACT SAME VALUE as the `id`… 😵‍💫

I'm implementing account deletion.

Deleting a PostHog profile means you can't later use the same `distinct_id`. Sounds great — we assign our users a UUID and use that when sending data to PostHog.

So then what's this mysterious integer `id` that the deletion API needs?

posthog.com/docs/api/per...

Oh interesting.. no more info about `previous_attributes` in events 🤔 Plus no more `expand` for API requests!

The excellent Refined GitHub browser extension will at least minimise and dim such PRs by default:
github.com/refined-gith...