ArchiveGrid Visitor's Guide

Multiple words: When you enter more than one word, for example, yosemite muir roosevelt, descriptions will match if they contain all of your words.

Find a phrase: To find a phrase, enter your terms within double quotes. For example, "harlem renaissance".

Find words near each other: To find matches when your search terms are near one another, put your words in quotes and follow with a tilde "~" and a number indicating how far apart the matching words can be. This technique is known as a proximity search. For example, to find the words "john" and "booth" within a span of four words, "john booth"~4.

Alternate words: To match any of the words you've entered, enter "OR" between words. For example, antietam OR sharpsburg. The symbol "||" can be used in place of the operator "OR".

Exclude words: To eliminate matches on a word, precede the word with NOT. For example, "civil war" NOT spanish. The symbols "-" or "!" can be used in place of the operator "NOT".

Word variations: Wildcard characters can help find different forms of a word. Use "*" to represent any number of characters; for example, archiv* matches archive, archives, archiving, archival, etc. Use "?" to represent one additional character, for example, archive? matches archive and archives.

Increasing the relevance of a search term: If your search includes more than one word, and you'd like matches for one of the words to be given more importance, follow the term with the caret "^" and a number greater than 1. For example, to change the relevance ranking when searching the words lincoln gettysburg to highlight results that emphasize Abraham Lincoln's role, try lincoln^4 gettysburg.

Sophisticated searches: The logic of a search can be controlled with parentheses. For example (antietam OR sharpsburg) AND maryland matches descriptions that include the word "maryland" with either the word "antietam" or "sharpsburg".

When you get your search results, the most relevant matches appear first. Relevance is determined by several factors, including the number of times the search terms appear in the description, the length of the field where matches occur, the number of words in the search, and any emphasis given to one or more of the search terms or one or more of the fields.

Some terms may have different meanings in different contexts: For example, "columbia" may return results that are not as relevant from the searcher's point of view. Adding different words — in this case, "columbia space shuttle" or "columbia university" — can help.

ArchiveGrid is operated by RLG at its headquarters in Mountain View, CA. The service runs on Solaris systems with the Apache Web server and the Apache Tomcat Java services container.

Some of the collection descriptions are harvested by RLG using a proprietary web crawler, customized for assembling EAD-encoded and other forms of archival finding aids for indexing in ArchiveGrid. But most of the collection descriptions are based on catalog records in the RLG Union Catalog. Records describing archival materials are indexed in ArchiveGrid using RLG proprietary applications for extracting, mapping, and indexing of MARC data from the Union Catalog.

The index for searching collection descriptions uses the Lucene text search engine library.

About RLG