4 ms·
A workaround approach - disregarding any existing standards - would be to simply give the user a random, unique string to search for and then try to detect that
by Nextgrid 4y ago
A workaround approach - disregarding any existing standards - would be to simply give the user a random, unique string to search for and then try to detect that string in the URL (typically as a query parameter) - then the browser just has to replace that string with the user's actual search query in subsequent searches.
- deafpolygon 4y agoThis is probably how Firefox works, because I get suggested a new search engine every time I search on a new site.
- lstamour 4y agoFirefox uses OpenSearch. https://support.mozilla.org/en-US/kb/add-or-remove-search-engine-firefox https://support.mozilla.org/en-US/kb/add-or-remove-search-en... Chrome uses HTML5 search inputs: https://martin-thoma.com/search-engine-autodiscovery/ https://martin-thoma.com/search-engine-autodiscovery/