Sources and references
The guides on this site describe how HTTP status codes behave and how search engines treat missing pages. Those statements come from the specifications and from the official documentation of the search engines themselves, listed below so you can verify anything you read here.
The HTTP specification
The behaviour of every status code discussed on this site is defined in the HTTP standard. The current documents are RFC 9110, which specifies HTTP semantics including the full status code registry, and RFC 9111, which covers caching. These replaced the older RFC 7230 series. They are the authoritative source for the difference between a 301 and a 302, for the meaning of 410 Gone, and for what a server is expected to do in each case.
Search engine documentation
Statements about how missing pages are treated in search results, about crawl budget, and about soft 404 detection come from the documentation the search engines publish for webmasters.
- Google Search Central documentation
- How HTTP status codes affect Google's crawlers
- Redirects and Google Search
- Google, crawl budget management for large sites
- Search Console, the page indexing report
Crawling and discovery protocols
The rules a crawler is expected to follow, and the formats a site uses to declare what it contains, are public specifications rather than search engine policy.
- RFC 9309, Robots Exclusion Protocol
- The sitemaps.org XML sitemap protocol
- IndexNow protocol documentation
Web standards and accessibility
The recommendations about alternative text on images, and about markup in general, follow the HTML specification and the Web Content Accessibility Guidelines.
- The HTML Living Standard
- WCAG 2.2 quick reference
- W3C Web Accessibility Initiative, the images tutorial
- MDN Web Docs, HTTP response status codes
- MDN Web Docs, mixed content
Server configuration
The redirection examples in our repair guide use standard Apache directives, documented by the Apache Software Foundation.
Graphics used on this site
The site header and logo were produced with cooltext.com.
How the tool itself is built
Bye404 is written in plain PHP and runs on an Apache server. The crawler uses the cURL library for its requests and the DOM extension of PHP to read the HTML of each page. It follows redirections manually so that every hop can be validated, requests only public http and https addresses, and paces itself so that the sites it tests are never flooded.