Users of text-based browsersConsider using the old interface at https://old-search.marginalia.nu/, as it uses fewer modern CSS tricks, and should work better than the new UI. It's functionally nearly identical, but just renders it using a different layout.

Marginalia Search

Wait A Moment

The search engine is currently seeing a lot of fairly aggressive bot activity.

Please wait for 3 seconds before proceeding.

Syntax

This is a keyword-based search engine. When entering multiple search terms, the search engine will attempt to match them against documents where the terms occur in close proximity.

Search terms can be excluded with a hyphen.

While the search engine at present does not allow full text search, quotes can be used to specifically search for names or terms in the title. Using quotes will also cause the search engine to be as literal as possible in interpreting the query.

Parentheses can be used to add terms to the query without giving weight to the terms when ranking the search results.

Samples

soup -chicken: Look for keywords that contain soup, but not chicken.
"keyboard": Look for pages containing the exact word keyboard, not keyboards or the like.
"steve mcqueen": Look for pages containing the exact words steve mcqueen in that order, with no words in between.
apology (plato): Look for pages containing apology and plato, but only rank them based on their relevance to apology

Match Indicator

The "coffee stain" indicator next to the search results illustrates the locations in a document the keywords appear. It operates a bit like a clock, starting at the top and working its way around clockwise.

This is by the beginning

This is in the middle

This is toward the end

Language Limitations

The search engine currently does not support any languages other than English.

Support for other languages is planned, but not available right now. Adding support for additional languages and making it work well is somewhat time-consuming, meanwhile having bad support for a language won't make anyone happy.

Webmaster Information

If you wish to add your website to the index, follow the instructions in this git repository, if you do not want to mess with git, you can also email contact@marginalia-search.com with the domain name.

The search engine's crawler uses the user-agent string search.marginalia.nu, and requests come from the IPs indicated in https://search.marginalia.nu/crawler-ips.txt.

If you do not want your website to be crawled, the search engine respects robots.txt. In case of questions, bug reports or concerns, email contact@marginalia-search.com.

Special Keywords

Keyword	Meaning
site:example.com	Display site information about example.com
site:example.com keyword	Search example.com for keyword
site:.example.com keyword*	Search example.com and its subdomains for keyword
browse:example.com	Show similar websites to example.com
ip:127.0.0.1	Search documents hosted at 127.0.0.1
links:example.com	Search documents linking to example.com
tld:edu keyword	Search documents with the top level domain edu.
?tld:edu keyword	Prefer but do not require results with the top level domain edu. This syntax is also possible for links:..., ip:... and site:...
q>5	The amount of javascript and modern features is at least 5 (on a scale 0 to 25)
q<5	The amount of javascript and modern features is at most 5 (on a scale 0 to 25)
year>2005	(beta) The document was ostensibly published in or after 2005
year=2005	(beta) The document was ostensibly published in 2005
year<2005	(beta) The document was ostensibly published in or before 2005
rank>50	The ranking of the website is at least 50 in a span of 1 - 255
rank<50	The ranking of the website is at most 50 in a span of 1 - 255
format:html5	Filter documents using the HTML5 standard. This is typically modern websites.
format:xhtml	Filter documents using the XHTML standard
format:html123	Filter documents using the HTML standards 1, 2, and 3. This is typically very old websites.
generator:wordpress	Filter documents with the specified generator, in this case wordpress
file:zip	Filter documents containing a link to a zip file (most file-endings work)
file:audio	Filter documents containing a link to an audio file
file:video	Filter documents containing a link to a video file
file:archive	Filter documents containing a link to a compressed archive
file:document	Filter documents containing a link to a document
-special:media	Filter out documents with audio or video tags
-special:scripts	Filter out documents with javascript
-special:affiliate	Filter out documents with likely Amazon affiliate links
-special:tracking	Filter out documents with analytics or tracking code
-special:cookies	Filter out documents with cookies