No trackers, no pixels: open our page source and count
Privacy policies describe intentions. Page source describes behavior. This post takes the second route: load any Peopoli page, open devtools, and count the third-party requests. The honest answer is: you won't need to count.
What we don't load
No Google Analytics. No Meta pixel. No Sentry. No third-party CDN, no font CDN, no cookie-consent theater because there are no third-party cookies to consent to. The only external service in the stack is the payment processor — Stripe — because card processing requires exactly that.
What we run instead
The stack is self-hosted, on hardware we own: mail (JMAP via our own James deployment), object storage (S3-compatible, our cluster), search, translation, content classification, and the AI models themselves. When Chart drafts a diagram or the translation pipeline renders your posts in 55 languages, that inference happens on our GPUs — your content is not being fed to a third-party model provider's training pipeline. That last sentence is increasingly the sharpest one we can write.
Speed is a privacy side-effect
The marketing site you're reading is static HTML and one small stylesheet — no framework runtime, no third-party JS, system fonts, one self-hosted image where it matters. That's why it renders before you finish the sentence. Core Web Vitals aren't a growth hack here; they're what's left when you refuse to load someone else's script. Fast is what an honest page looks like.
The receipts rule
Every privacy sentence we publish has to survive one question: could an engineer at a competitor publicly refute it in a paragraph? “No trackers” survives it because it's checkable in view-source, today, by anyone. That's the standard we hold ourselves to — and the one we invite you to hold us to.
Claim your @username — free.
Self-service registration. No waitlist, no credit card, no trial clock.
Your public profile lives at me.peopoli.com/@yourname