Finite crawl is not a promise of coverage

How Oernoe Search works describes crawling as a real machine with limits: crawlers visit pages they already know about, follow public links, come back when a page changes, respect robots.txt, and honour crawl rate limits. The Search homepage says the quiet half of the same fact: results come from websites and knowledge entries in the Oernoe index, and not every page on the web is indexed. This essay is Search limits writing about crawl honesty. It is not a remake of the finite-index essay that taught missing as ordinary, and it is not a remake of the guide-versus-results-page essay that separated documentation on www from retrieval on search.oernoe.com. The hinge here is narrower: describing how crawl works is not a promise that your URL will be covered.

## What the guide actually says about crawl

How Search works keeps an index of public web pages Search has fetched. That sentence already refuses the fantasy of a live mirror of the open web. The index exists so Search can return results without asking every site on the web in real time. Continuous crawling is named: crawlers work around the clock to visit pages and stay current with new and updated content. Respecting robots is named: Search respects robots.txt files and crawl rate limits set by website owners. Privacy-first crawling is named: the guide says the crawler does not store IP addresses, device information, or identifying data during crawling, and does not harvest personal data, extract email addresses for advertising lists, track user identities, or store behavioral signals. Content analysis is named: publicly visible page content, structure, and links.

Those bullets are mechanism. Mechanism has consequences. A page behind a robots disallow is not “down in Oernoe Search”; it is unavailable to the index by the publisher’s own rule. A crawl rate limit that slows discovery is not an outage; it is a polite refusal to hammer a host. A page that has never been linked from anything the crawler already knows is not guaranteed to appear tomorrow because someone wishes the index were larger. Continuous does not mean complete. Respect does not mean override. Analysis does not mean omniscience.

The guide’s comparison FAQ refuses a second fantasy: size-match marketing. Oernoe Search is a working search engine at search.oernoe.com. It is not a claim that the index matches every other index in size. The difference the guide will stand behind is narrower: Search does not use personal advertising profiles to rank results, and the publisher site discloses Google ads when they appear on selected pages. Crawl honesty and size humility are the same family. Both refuse to let privacy language or product pride invent coverage the machine does not have.

## What the Search homepage prints

A guest open of search.oernoe.com on 8 September 2026 shows product language, not publisher language. The title is Oernoe Search. Tips teach site filters, quoted phrases, switching between Websites and Knowledge, and a slash shortcut back to search. The tip footer states the limit that matters for this essay: results come from websites and knowledge entries in the Oernoe index, and not every page on the web is indexed.

That sentence is crawl honesty translated into results-page English. It does not explain robots.txt. It does not list ranking factors. It does not need to. It tells the person staring at an empty or thin hit list that absence can be true without Status turning red. Earlier journal writing already argued that missing is ordinary, and that tips on the Search homepage are limits printed in public rather than magic operators. This page does not re-argue those acceptance postures. It asks a different question: when the company describes crawl in a finished guide, what promise did it actually make?

The answer is process, not inventory. The guide promises that crawl has rules. The homepage promises that the index is a subset. Neither document promises that a particular public URL is inside that subset, that a page published an hour ago has already been fetched, or that robots-respecting refusal will be overridden because the site owner prefers visibility.

## Crawl honesty versus coverage promises

Coverage promises sound generous and fail under inspection. “We crawl the web” can be read as “we have your page.” “Continuous crawling” can be read as “your update is already live in results.” “We follow links” can be read as “any URL you paste into support will appear.” None of those readings survive the guide’s own machinery. Fetch is required. Discovery is required. Robots permission is required. Rate limits apply. The index is what was fetched and kept, not what someone believes ought to have been fetched.

Honesty about crawl is useful precisely because it refuses those readings in advance. It tells a site owner why a Disallow line is doing what it was written to do. It tells a reader why a brand-new post can be absent without implying censorship. It tells a reviewer why privacy-first crawling is not a slogan that enlarges the index. Corrections already had to unwind a different overclaim when older copy talked as if the whole network had zero tracking while the publisher applied for AdSense. Crawl coverage is the same class of risk: a soft sentence that outruns the system.

About’s product honesty belongs in the same family. Searching as if an unlisted product must appear is asking the index to invent a directory entry. Homepage directory truth stays on www. Crawl truth stays in How Search works and on the Search tip footer. Confusing those truths produces bad mail: people ask support@ why Search does not show a page the crawler was never allowed to fetch, or why a noindexed promotional leftover must rank as if Search were obligated to resurrect it.

## Distinct from finite-index and guide-versus-results essays

The finite-index essay taught an emotional and procedural baseline: start from ordinary missing, then decide whether the gap needs a sharper query, another engine, a publisher correction, or a careful support message. That essay’s hinge was acceptance of absence. This essay’s hinge is the crawl description itself. Finite index is the stored set. Finite crawl is the process that feeds the set. You can accept missing and still misread crawl prose as a coverage SLA. This page exists to block that misread.

The guide-versus-results essay taught category: documentation lives on www; retrieval lives on search.oernoe.com; ads may appear on the finished guide and do not appear inside Search. That essay’s hinge was hostname and job. This essay assumes that split and stays inside Search limits: even when you are reading the right document about crawl, the sentences are still not a promise that every public page will be covered. Mechanism text is still mechanism text.

Ranking prose in How Search works makes the same cut from another angle. Content relevance, authority, link analysis, freshness, aggregated satisfaction signals, mobile-friendliness, and page speed can order documents the index already holds. They cannot rank a page the crawler never fetched. Privacy-preserving ranking is a claim about how stored documents are ordered and about what personal advertising files are not built. It is not a claim that the stored set is complete. Readers who hear private search and translate it into must have my page are asking privacy language to do crawl work again.

## Robots, rate limits, and intentional refusal

Robots.txt respect is the cleanest example of crawl honesty. A publisher that blocks crawlers is exercising a public rule. Search that honours the rule is behaving as documented. The absence that follows is not a secret demotion and not a Status incident. It is the system working. Crawl rate limits are softer but belong to the same class. They trade speed for host health. A guide that celebrated privacy-first crawling while ignoring robots would be theatre. How Search works refuses that theatre. The results page inherits the consequence.

Corrections’ robots story on www is related but not identical. The publisher once shipped a robots.txt that missed live application routes because trailing-slash Disallow lines did not match bare paths. Empty shells became fetchable. The fix used noindex headers and a narrower Disallow set so crawlers could read the header. That episode is about the publisher’s own routes. Search’s robots respect is about other people’s routes. Both stories teach the same discipline: write the rule you intend, then live with what the rule excludes. Do not paper over exclusion with a coverage slogan.

## Funding does not buy infinite crawl

How Search works answers how the company makes money if Search does not profile users: optional premium features and contextual Google ads on selected publisher pages such as that guide. Search queries are not sold or used to build advertising profiles. Ads on the guide are disclosed in Privacy and Cookies. Cookie Policy adds that Search queries typed at search.oernoe.com are not an input to those units. Privacy section 3 repeats that Oernoe does not use search history to create advertising profiles and does not sell account or search data.

That funding model lives on www. It does not pay for infinite crawl. Expecting monetization of selected publisher essays to force universal indexing is a category error. Expecting ads.txt verification or a google-adsense-account meta tag to enlarge the Search index is the same error with different paperwork. Publisher inventory and Search coverage are different machines on different hosts. Crawl honesty keeps them from being mashed into one marketing sentence.

## What this essay refuses

It will not invent coverage percentages, crawl counts, or index sizes. It will not claim every public company has a Knowledge panel. It will not claim a page published minutes ago must already be present. It will not claim Status green enlarges crawl coverage. It will not walk Search chrome control by control. It will not treat the guide as the results page. It will not claim that respecting robots is optional when a site owner prefers to rank.

It will also not claim that every absence is good. Some gaps are still worth filling later. Continuous crawling exists because the company wants the index to improve. Improvement is not the same as a promise that a named URL is covered today. The useful correction is linguistic: when you read crawl prose, hear process. When you need coverage this index may not have, use another engine like an adult using finite tools. When you file support@oernoe.com about Search emptiness, bring the query, the hostname, the time, and what a known-good query returned, instead of a theory that the crawl section guaranteed your page.

## How a reader should use crawl honesty

Read How Search works for the mechanism: known pages, public links, revisits, robots, rate limits, public content only. Read the Search homepage tip footer for the plain limit: not every page on the web is indexed. Read About when you need the live services directory, not when you need the crawler’s todo list. Read Corrections when a publisher document overclaimed; do not send a third-party robots block to hello@ as if it were a typo on www.

If you run a site and want it considered, make it fetchable, linkable, and allowed. If you blocked the crawler, believe your own robots file. If you are a reader staring at a thin list, rewrite the query, try the other tab, try a known public string, and only then treat emptiness as a possible bug. Finite crawl is how Search stays honest about being a real engine with a real subset. It is not a soft promise that coverage will catch every URL you care about. Process first. Coverage only where the process actually reached.