Feature Set	Description
Full-text Search Engine Indexes the web	spider and index up to 10 million pages per seat federated search lets you search multiple seats to create multi-billion document indexes supports html, pdf, and text document types supports other document types using external conversion script supports https and secure sites supports image search: gif, jpg, png, bmp, tga, pcx supports cookies, robot compliant specify crawl instructions for directory categories crawl websites and directory categories crawl user submitted sites and links create metadata rules or assign metadata to directory listings and categories export new links from document data to a CSV directory import file
User Accounts Submit url listings	submit url listings specify url categories, premium keywords, title, description free or paid accounts toolbar with popup blocker page and search alerts bookmarking tool with search and tag clouds custom spidering rules for url listings spider and index urls immediately, by schedule, or after review rich backend user administration
Directory Engine	imports over 500,000 categories and 5 million pages from ODP add any number of metadata fields to categories and listings complete search and browse functionality users can purchase listings and keywords customize content and presentation use your own directory using ODBC import, file import, or RDF import export directory to ODBC, or file
Metasearch Engine Searches web and intranet search engines seamlessly	crawls the web to discover over millions search forms aggregates specialty web and intranet database searches supports metadata extraction let users create their own metasearch sets paid search result feed aggregation
Wiki Engine Import from wikimedia or start your own	supports millions of pages indexes pages after editing for immediate searchability support templates for complex rendering control page editing access incorporate Wikipedia extracts onto search result pages
Tag Engine Bookmarks and tags	import user bookmarks use NRS toolbar for bookmarking button share or keep private tag clouds and tag lists
Intelligent Agent Monitors searches and pages	notifies you when new search results appear in saved searches deep query functionality lets you specify complex expressions create reports that aggregate alerts. make the reports public so other users also be notified.
Mail Engine Organizes and shares your results	save search results organize alerts in folders e-mail content to your buddy lists
Plugin Engine Use plugins for extra functionality	develop your own plugin tie in to users and crawling full-text search
HTML Administration Gives complete customization and application building	purchase as you go more documents, users, and support build complete research applications administer users, templates, server settings view web logs XSL editor manage server seats: crawling/indexing/databases
Web Services & Database SDK Provides API functionality to develop apps	.NET, Java, C++, Perl database interfaces integrate search and tracking into your app using XML services

NRS is further composed of the following architectural components:

Base Services	Description
WEBSERVER	By incorporating a webserver, NRS does not need to rely on external webservers, simplifying installation and integration. NRS serves pages as XML or HTML, so if you desire to use your webserver, NRS can act as your content provider in XML format. For robust security, NRS supports HTTPS mode and self-issued SSL certificates.
XSLT ENGINE	All pages are generated using an XSLT engine that transforms XML into HTML using an XSL stylesheet. NRS lets you build your own XSL templates for total control over presentation. Through an administration interface, content can be aggregated as one XML stream. NRS uses the Open Source Sablotron XSLT engine, or, if detected, can also use Microsoft's MSXML engine on the Windows platform.
CRAWLER	An industry conforming crawler and caching engine provides excellent performance for metasearches, content aggregation, and background crawling to refresh search indexes. Supports http,https, and cookies.
INDEXER	A scalable full-text indexer can index 3 million pages in 24 hours. NRS builds indexes that feature for every word occurance a popularity ranking score and category information.
WIKI	A wiki rendering engine allows for the development of complex Wiki templates and full-text search.
MAIL	NRS features SMTP, POP3, mail send, and HTML client modules for a comprehensive mail solution. Mailboxes let you organize agent notification e-mails into folders, and receive trade newsletters.
DATABASE	Object-relational database technology provides a performance edge by optimizing all metadata relationships between objects. The application database engine is powered by Gigabase, a database that scales to terabytes.

Feature		Description
DIRECTORY
	- Schedule	Schedule directory updates at specific times
	- Import	Import from ODP, ODBC, or from file
	- Export	Export to ODBC or file
	- Fields	Add extra fields to listings
	- Shadow Updates	Import new directory in background and swap out with old one
CRAWLING
	- Schedule	Schedule crawler to run at specific times
	- Protocols	Supports http, https, and cookies
	- Documents	Supports html,text,pdf, images, and other filetypes through an external converter script
	- Languages	Supports all languages
	- Stay on site	Keep crawler on site (includes subdomains)
	- Stay on path	Keep crawler on same URL path
	- Follow cgi	Control access to dynamic URLs
	- Index forms	Index search engine forms found on pages
	- Index links	Allow "link:url" query which returns urls pointing to page
	- Robot check	Optionally check robot.txt file
	- Depth	Maximum directory depth to crawl
	- Distance	Maximum hops crawler can make from start URL
	- Max pages	Maximum pages to crawl for given site
	- Politeness	How long to wait between page retrievals to same domain
	- History	Number of historical versions to keep of a page
	- Excludes	Prevent crawler access to directories, pages, or domains
	- Includes	Allow crawler access to a directory, page, or domain
	- Refresh rate	How many days between recrawls of a document
INDEXING
	- Popularity	Activate link analysis to determine URL popularity ranking
	- Incremental	Incrementally index newly spidered pages
	- Metadata	Create rules to extract metadata from full-text document, directory fields, metatags, and other sources. Use regexp or templatized expressions.
	- Rank boost	Boost rank of a given collection
SEARCHING
	- Operators	AND, OR, ANDNOT, -, +, ""
	- Duplicates	Filter out duplicates
	- Collections	Search across one or more collection
	- Metadata	Filter search results by selecting metadata values, and sort results by metadata value
	- Remote	Search a remote index on another server
	- Highlighting	Customize highlighting
	- Cached	View cached page with highlighting
	- Historical	View differences between historical versions of the document
METASEARCH
	-Discovery	Enter url of search page, NRS will retrieve page and import search box definition
	-Fields	Provide a name and description for each search
	-Streaming	Control streaming to browser. Send search results to user browser as they come in or once they are all in.
	-Robot	Follow robots.txt before using search resource
	-Autoparse	Extract search results automatically without a parse rule
	-Parse rule	Provide a parse rule to extract specific metadata fields in the search results
	-Parameter override	Provide new parameters to the search engine.
	-Input fiddling	Modify query before retrieving search results
	-Cache	Control length of time to cache search results from a source
NEWS
	-Parse rule	Provide parse rule to extract news titles, descriptions, and other fields.
	-Display options	Specify when to render news: home page, query page,..
LINKS
	-Related links	Provide a list of related links to display on search/news page
XML
	-Aggregation	Aggregate XML streams and transform each to standardize output
ODBC
	-SQL Query	Incorporate SQL query results as XML into the template
INDEX
	-Search results	Incorporate other search result xml streams from main index
SEARCHING
	-Highlighting	Customize highlighting
HOSTING
	-UI access	Manage collections and templates from HTML UI
	-Templates	Create templates that provide functionality and HTML customization
	-Applications	Organize templates into grouped categories and subcategories in a tabbed HTML interface
	-Host	Run server under a hostname
	-Virtual host	Provide subdomain redirection for multiple solution hosting off one server
SDK
	- .NET	Easy to use .NET client to database
	- Data access	Access database using Perl/C++/C/Java or command-line tools
	- XML	Get XML feeds.

See Also:

Search cluster architecture
Application architecture