Glossary

The words people throw around when they talk about growing on social — explained without the jargon, for founders who’d rather be building.

• Anchor fieldA field that must be present for a document to be what you think it is. The cheapest defence against parsing an error page as data, and why it matters more than any selector.
• API creditA prepaid unit of API usage. Why credits and subscriptions produce very different real costs for the same workload, what makes a credit expire, and the questions to ask before buying a pack.
• Best time to postThe hour and weekday a post goes out, and why it changes reach. What the generic charts leave out, why the answer never transfers between platforms, and the only version of it that holds.
• Build in publicSharing your revenue, decisions and mistakes while you build. What actually makes it work, what makes it backfire, and why it isn't for everyone.
• Data freshnessHow old the data you are reading actually is. Why uptime is the wrong measure for a collector, and the difference between a cached answer and a stale one.
• Decentralized identifier (DID)A stable identifier for an account that does not depend on a handle or a host. Why it matters for tracking anything over time, and the two flavours you will actually meet.
• Engagement rateHow engagement rate is actually calculated, why the same post scores differently on every platform, and why a high rate can still mean nothing.
• FederationA network of independent servers that interoperate. Why a handle without a host identifies nobody, which server is authoritative, and what federation changes about collecting data.
• Follower countThe most requested number in social data, and the easiest to get wrong. The neighbouring fields that look identical, why an anchor is mandatory, and what null means here.
• Headless browserA real browser driven by code with no window. When it is the only thing that works, what it costs against a plain HTTP request, and why it is the last resort rather than the default.
• HookThe first line or first second of a post — the part that decides whether the rest gets read. What makes one work, and why copying viral hooks usually backfires.
• KarmaThe score Reddit and Hacker News attach to an account. It works as a gate rather than a scoreboard, which is why a new founder's first link disappears without anyone disliking it.
• llms.txtA file at the root of a site that tells language models what it contains, in the order that matters. What goes in it, how it differs from a sitemap, and why generating it beats writing it.
• MCP serverA standard way to expose tools and data to AI assistants. What the Model Context Protocol is, what a server actually provides, and why a REST API is not automatically one.
• PaginationFetching a long list in pieces. Offset against cursor, why a silently clamped limit is worse than a rejected one, and how to tell 'no more results' from 'no results'.
• Personal dataInformation relating to an identifiable person — including information that is publicly visible. Why 'it was public' is not a legal defence under GDPR or LGPD, and what changes the moment you store it.
• PollingAsking an API repeatedly whether anything changed. Why it was free under request-based pricing and is a line item under per-resource pricing, and the two things that cut the bill.
• Rate limitA cap on how often you may call an API. Why the number on the pricing page is rarely the one you hit, what a 429 actually means, and the difference between a rate limit and a quota.
• Reach vs impressionsImpressions count times your post was displayed; reach counts people who saw it. Why the gap between them is the number worth watching.
• Residential proxyRouting requests through consumer IP addresses instead of datacenter ones. Why some sources answer one and refuse the other, what it costs, and the intermittency it introduces.
• robots.txtA file at the root of a site telling automated clients what not to fetch. What it is, what it is not, and why it matters even though it is not a law.
• ShadowbanWhat a shadowban actually is, why most suspected shadowbans are something else, and how to tell the difference before you rebuild your whole strategy.
• Show HNThe Hacker News format for something you built that people can try right now. What qualifies, the mistake that gets most launches flagged, and what the traffic is actually worth.
• Silent failureWhen code returns a plausible wrong answer instead of an error. Why it is the most expensive class of bug in data collection, how it happens, and the three patterns that catch it.
• Time seriesThe same measurement repeated on a fixed cadence. Why 'what is trending now' and 'what is rising' are different questions, and why the second one cannot be answered retroactively.
• TLS fingerprintingIdentifying a client by how it negotiates the encrypted connection, before any request is sent. Why a page can return successfully and simply omit its data, and what that costs.
• Trending topicA subject a platform has decided is rising right now. How trending is actually computed, why the label arrives late, and why yesterday's list is unrecoverable.
• Unified schemaOne response shape across many platforms. What a unified schema genuinely solves, the two ways it lies, and why a null in the same field as a number is the honest design.
• Verified badgeA platform's mark on an account — and four different things it can mean. Why one boolean cannot carry them all, and why the honest value is often null.
• Web scrapingReading data from a web page rather than an API. The four ways pages are actually parsed, why the structured-data block beats a regex, and where scraping stops being reliable.
• WebhookThe provider calls you when something happens, instead of you asking. What it saves, the three things it needs from your side, and why most public social data has none.