Search Server 2008 Features and Benefits
Search Server 2008 is a standalone search product from Microsoft that is built on the SharePoint platform and extends the Office SharePoint Server 2007 search capabilities. When it is installed, you’ll receive a Shared Services Provider and a Search Center, both of which are fully capable of all of the search functions that you’ve come to expect in SharePoint Server 2007. However, the Search Server 2008 implementation will lack many of the SharePoint Server 2007 farm features because it is a product focused on search and indexing.
The entire reason that a company implements a search engine is because users want to find information via a keyword query. Because we are accustomed to finding information on the Internet using an Internet search engine, it is only natural that we would want to have our own search engine on our internal corporate network to find internally hosted information. Because much of our information is not indexed by an Internet search engine, it stands to reason that having our own indexing engine that indexes our organization’s internal information will help users find the information they need to be successful in their jobs.
Search Server 2008 was built for that purpose and offers the following benefits:
- Search Center, which is a Web site with a query and result set interface that can be exposed to the Internet if needed.
- Improved relevance engine, which shipped with the SharePoint Server 2007 product release and can be fine-tuned within the Search Server 2008 administration interface.
- Ability to target search Web parts with different functionality to SharePoint Server 2007 audiences.
- Ability to build customized search results pages for each scope and audience to produce a highly customized search experience.
- Ability to localize the interface in 25 languages (at the time of this writing).
- Ability to customize the search interfaces using Microsoft Visual Studio and Microsoft Office SharePoint Designer.
- Scalability to crawl any number of documents that you wish. This is important: Unlike other search engine companies, Microsoft does not impose a document limit.
- Federated queries and connectors allow you to execute queries to remote indexes in other data repositories, applications, and services using the Open Search standard.
- Reporting on user query and result activity is built-in with Search Server 2008.
- High availability and load balancing can be used through a combination of front-end and back-end server topologies, which allows you to grow your search infrastructure to meet your needs.
- Secure federated relationships by setting global- or user-level security for your federated search relationships.
- Specify unique crawl inclusion/exclusion behaviors and authentication credentials for specific content sources.
- Remove single items from the index using the single item index removal feature. This action automatically creates a new crawl rule to ensure the single item isn’t crawled again.
Search Server 2008 is built on the OpenSearch 1.1 standard. This standard was developed by the folks at Amazon (https://opensearch.a9.com), and Internet Explorer 7 was built with an OpenSearch client embedded into it. OpenSearch was introduced by A9/Amazon at ETech in March of 2005, and a number of sites adopted the standard during the first six months after its introduction.
More Info There is a growing list of Web sites that are OpenSearch compliant. A list of these sites can be found at https://opensearch.a9.com//opensearch/searches.jsp.
Notes From the Field: Search Providers in Internet Explorer 7 that Use OpenSearch 1.1
Because search is such an important aspect of our users’ daily lives, a browser ought to do something special with search queries. In OpenSearch, the Internet Explorer 7 product team saw the foundation for making this happen.
One of the main goals of the Internet Explorer 7 product team was to make it easy for Internet Explorer 7 users to personalize their lists of search providers. Specifically, the goal was to allow users to target their search queries directly to the site of their choice, such as MSDN, an intranet portal, a SharePoint team site, or some other OpenSearch-compliant search engine. The way a user adds a search provider is similar to adding a favorite.
Site administrators can facilitate this process by providing a link to an OpenSearch definition file for the site in the head tag of the master page by using the following format:
<link title="Company Intranet" type="application/
opensearchdescription+xml" rel="search" href="/Style Library/
XSL Style Sheets/intranet.xml" />
In Internet Explorer, the Search button will respond by turning orange and listing the Intranet Search as a provider. Users may then simply add your search provider to their list of search providers.
The HREF is a link to the OpenSearch 1.1 Description Document for the search query. If the goal is to search the company’s SharePoint intranet, the following XML is used:
<?xml version="1.0" encoding="UTF-8"?>
<OpenSearchDescription xmlns="https://a9.com/-/spec/
opensearch/1.1/">
<ShortName>Company Intranet</ShortName>
<Description>Company Intranet Search</Description>
<Url type="text/html" template="https://intranet.company.com/searchcenter/pages/
results.aspx?k={searchTerms}"/>
<SyndicationRight>open</SyndicationRight>
</OpenSearchDescription>
If the goal is to add a people search, the following would work for SharePoint.
<?xml version="1.0" encoding="UTF-8"?>
<OpenSearchDescription xmlns="https://a9.com/-/spec/
opensearch/1.1/">
<ShortName>Company People</ShortName>
<Description>Company People Search</Description>
<Url type="text/html" template="https://intranet.company.com/searchcenter/
pages/peopleresults.aspx?k={searchTerms}"/>
<SyndicationRight>open</SyndicationRight>
</OpenSearchDescription>
Another alternative is to advertise the availability of search on your site through a link that calls window.external.AddSearchProvider(“URL”), where the URL is a link to the previously mentioned OpenSearch XML file. When the user clicks this link, code in Internet Explorer opens a dialog box asking the user if the search provider should be installed. Note that this process must be initiated by the user. If the client selects to install the search provider, Internet Explorer is configured to allow queries to the remote index using the Description Document for the search provider from the URL.
The OpenSearch options mean that you can also offer search results from other sites. Taken from the Internet Explorer 7 team blog site at MSDN, what follows is an example of a Description Document for searching on MSDN:
<?xml version="1.0" encoding="UTF-8"?>
<OpenSearchDescription xmlns=
"https://a9.com/-/spec/opensearchdescription/1.1/">
<ShortName>MSDN</ShortName>
<Description>MSDN Example Search</Description>
<Tags>MSDN Developer</Tags>
<Contact>admin@msdn.com</Contact>
<Url type="text/html"
template="https://search.microsoft.com/search/
results.aspx?qu={searchTerms}&amp;View={language?}
&amp;p={startPage?}"/>
</OpenSearchDescription>
If a user searches MSDN for “Search Server 2008,” the following URL is passed: https://search.microsoft.com/search/results.aspx?qu=Search%20Server%202008\&View= en-US&p=1.
The OpenSearch standard allows the remote search engine to return results using Really Simple Syndication (RSS), Atom, HTML, or XHTML. The XML for returning the results in both RSS and HTML would look something like this (this XML is taken directly from the Internet Explorer 7 team blog site):
<?xml version="1.0" encoding="UTF-8"?>
<OpenSearchDescription xmlns="https://a9.com/-/spec/opensearchdescription/1.1/”>
<ShortName>MSN</ShortName>
<Description>MSN Example Search</Description>
<Tags>MSN Web</Tags>
<Contact>admin@msn.com</Contact>
<Url type="text/html"
template="https://search.msn.com/results.aspx?
q={searchTerms}"/>
<Url type="application/rss+xml"
template="https://search.msn.com/results.aspx?
q={searchTerms}&amp;format=rss"/>
</OpenSearchDescription>
If the results come back with both HTML and RSS available, then the default setting in Internet Explorer is to choose the first URL specified in the Description Document. When it comes to returning results to Search Server 2008 for federation, be sure that the first URL in the Description Document is formatted to come back with RSS or Atom, not HTML or XHTML, because Search Server 2008 does not support the rendering of HTML or XHTML in the Web parts that display results from remote indexes.
Matthew McDermott, Microsoft MVP, Principal Consultant, Catapult Systems
Now that we’ve learned about some of the main features of Search Server 2008, it’s time to learn about the standard with which Search Server 2008 is compliant. Part of the reason that we’ll spend as much space on this as we do is because a good understanding of the command in the Description Document will help you understand how to build Federated Location Definition (FLD) files from scratch, if needed. While the Search Server 2008 interface will produce much of this XML for you when you build the FLD files from within the Search Server 2008 administration interface, the ability to read and understand the FLD files will be helpful to you when you troubleshoot Search Server 2008.
© Microsoft. All Rights Reserved.