Site Server - Site Server 3.0 Technical White Paper

Article
08/28/2007

Workers face an "information glut" and may not be able to find or share the information they need. By deploying Microsoft® Site Server 3.0, businesses can use the intranet to efficiently gather the collective expertise of the organization from wherever it resides.

1. Abstract

White Paper

Businesses are increasingly using intranets as a way for employees to easily publish, search, and deliver the information they need to make decisions and work more productively. Employees, however, face an "information glut" and may not be able find or share the information they need. By deploying Microsoft® Site Server 3.0, businesses can use the intranet to efficiently gather the collective expertise of the organization from wherever it resides—in Web sites, databases, file servers, e-mail—and deliver it to facilitate the sharing of knowledge and to improve business productivity.

Introduction

In order to realize the benefits of an intranet, businesses need a solution that allows them to efficiently manage the information flow across their business. This relates to information about both the content on the site as well as the site users. It includes:

Structured ways to get all types of content into the site.
Facilities to organize and manage information about the content and users.
Efficient delivery mechanisms to provide targeted information to users.
Mechanisms to measure and optimize site structure and site usage.

Traditionally, businesses have tried to build such capabilities into an intranet either by acquiring disparate, nonintegrated server applications or by building their own server applications from the ground up. But both methods are costly and inefficient: the first, because it requires that administrators and developers wrestle with independently developed applications and systems, and the second, because it can result in incompatible applications and unpredictable maintenance costs.

2. Introducing Microsoft Site Server 3.0

Microsoft® Site Server 3.0, a member of the Microsoft BackOffice® server family, provides a powerful alternative to the traditional methods of intranet development and maintenance. Microsoft Site Server 3.0 is the powerful intranet server, optimized for Microsoft Windows NT® Server operating system with Internet Information Server, for publishing and finding information easier and faster. By deploying Site Server 3.0, businesses can use the intranet to efficiently gather the collective expertise of the organization from wherever it resides—in Web sites, databases, file servers, e-mail—and deliver it to facilitate the sharing of knowledge and to improve business productivity.

Site Server 3.0 includes a unique set of features that are designed to work together to optimize information sharing across the organization:

Content Management provides a structured publishing process for multiple content authors to submit, tag, and edit content through a drag-and-drop Web interface. Site editors can then approve, edit, and enforce uniform guidelines for content.
Content Deployment enables administrators to deploy content securely and robustly across multiple distributed servers.
Search enables users to perform full-text and property searches across various stores and formats, including HTTP, file systems, Exchange files, and databases.
Personalization & Membership provide easy authoring and targeting of personalized information based on user profiles and behaviors and the ability to present search results in highly customizable user view and personalized Web pages.
Push enables businesses to create delivery channels for Microsoft Internet Explorer 4.0. Channel agents, based on Microsoft Active Channel™ Server, enable intranet developers and administrators to create channels from databases and file systems as well as Search and Index Server. The Active Channel Multicaster saves valuable network bandwidth by using multicast technology to deliver channels.
Knowledge Manager is a centralized Web-based application that integrates the Site Server knowledge management features to enable users to easily browse, search, share, and subscribe to relevant information.
Analysis transforms raw hits recorded in server log files into valuable information about the requests, visits, and users that interact with an intranet. This allows businesses to measure the effectiveness of an intranet. Analysis also captures content and site structure to identify issues, such as pages with long load times and out-of-date content.

3. The Power of Metadata

The ability of Site Server to provide businesses end-to-end information management for their intranet depends not only on a powerful set of features and full integration with Microsoft platform technologies, but also on the ability to catalog and share "metadata"—information about information. Metadata is created and updated by the various constituents who interact with an intranet site. Site Server provides the capabilities to capture this metadata and make use of it to provide a more powerful and complete business solution on the intranet:

authors "tag" content to identify its nature and define when, where, and how it is to be deployed and delivered to users,
administrators can set up systems to gather and store user profiles and site-usage information, and
users can define what content they want delivered proactively to them.

Using Site Server, content authors, developers, administrators, and users can share and make use of metadata. As such, it provides a consistent approach to the gathering, analysis, and integration of information. It also provides a way to coordinate all content and usage data for the best match between content and user and the best approach to making content easy for users to access.

Figure 1: Metadata input and delivery with Site Server 3.0.

Site Server enables the integrated sharing of data, and each of the Site Server features ties into this metadata model. Content attributes are stored in database catalogs, while user attributes are stored in a directory. The remainder of this paper details the primary features of Site Server and how metadata sharing is supported.

Establishing Basic Publishing Principles

To ensure a consistent and manageable way of creating content and getting it to an intranet, Site Server provides two primary publishing capabilities: Content Management and Content Deployment.

Content Management

The content to be placed on an intranet site is published from a highly distributed environment, in potentially dozens of source formats, and from hundreds of content authors. In such an environment, intranet site managers need a method of presenting content in a uniform way and maintaining professional publishing standards. They also need to prepare content so it is easily accessible by users.

Site Server addresses the need for presentation and publication standards by helping managers establish fundamental principles of structured publishing:

separation of content from presentation.
a formal editorial-approval process, including the use of submission templates.
support for distributed and remote submission, editing, and deletion.
support for specialization of contributors, by enabling them to create content using their choice of word processor, spreadsheet, or other desktop applications.

Site Server addresses the need to make content easily accessible by users through content types and content tags. Content types define the various kinds and formats of documents that are destined for deployment on an intranet, such as sales proposals, customer-satisfaction surveys, legal pleadings, and product demonstrations. Site editors can define content type based on subject matter, purpose, format, audience, and so on, and any combination of these criteria. Site Server supports an unlimited number of content types and stores all content of each type in its own directory.

Figure 2: Content Management process with Site Server 3.0.

Content tags identify aspects of the content type, such as author, title, and date of submission. For example, an author posting a document to the intranet is prompted for tags within his or her browser. Following the prompts, the author can tag the job description so that users can access it later by searching on criteria such as author, title, subject, job title, and pay scale. The author does not need to know any HTML to ensure that the tags are inserted correctly, because Site Server does this automatically. Subsequently, the site manager can tag the job description according to content properties, approval settings, expiration dates, and access codes.

Content tags are then used by Microsoft Index Server, a component of Windows NT Server, for indexing content and for displaying content in view pages. View pages are HTML pages with scripted rules that read content types and tags to determine what is displayed.

Content Deployment

Through Content Deployment, Site Server enables developers and administrators to stage and deploy content reliably, securely, and efficiently across an intranet. Content deployment simplifies and automates the distribution of files, server applications—including Microsoft ActiveX® Server Components and Java applets, directories, access-control lists, and other data from one intranet server to one or more servers.

Deployments are managed through projects and can be rolled back to previous versions as required. An administrator can run HTML-based management reports to track deployments. Such reports can simplify the rebuilding of a site and help builders and administrators avoid performing a full redeployment or rollback when it may not be necessary.

Site Server uses the TCP/IP networking and Windows NT authentication mechanisms to create secure server connections and various methods of data distribution, including data-validation and restart capabilities for reliable deployment across all types of networks. For additional security, Site Server supports data transmission using the MD5 hash digital signature.

Content Deployment servers deploy and retrieve content as instructed in their project definitions and routing tables. Although you can use a single server to deploy content between directories, most Content Deployment installations are more complex where multiple servers are involved and each one performs a different function.

Regardless of the configuration, Content Deployment servers can fill two roles in the staging and deployment process: staging and deploying content (staging server) and receiving content (end point server). All staging servers operate in the Windows NT environment and should have a network or Internet connection to other Content Deployment servers. End points can be Windows NT- or UNIX-based servers.

Finding Information Across the Intranet

On a sophisticated intranet site, only a portion of the content that is valuable to users resides on the intranet Web server(s). The bulk of content accessed by users resides elsewhere in corporate files, databases, and Exchange servers and on newswires and the Internet. For users to have the best, most complete information available, all this content must be as accessible to users as the content that resides on the intranet server.

Site Server provides a robust, secure, and configurable way to give users access to content from a vast assortment of sources. It provides industry-standard search algorithms supporting full-text, property, and single-query searches across multiple stores and formats including HTTP, NTFS, Exchange public folders, and Microsoft SQL Server™ and other ODBC databases.

The searches implemented by these algorithms encompass four operations: gathering documents, extracting information from them, indexing the information, and propagating the index to help users find the content they need.

Gathering content

The Search server in Site Server gathers content with the help of three kinds of crawls: Web crawl, file crawl, and Microsoft Exchange crawl.

In a crawl of Web documents, Search uses HTTP to collect a designated start page, taking links found on that page and following them to other pages to collect documents. Search can crawl documents on any kind of Web server.
In a crawl of a file system, such as in shared folders in Windows NT Server or the Windows® 95 operating system, Search starts the crawl from a designated file directory. Using the file protocol, Search collects documents in that directory and in its subdirectories, while preserving the documents' security Access Control Lists.
In a crawl of a Microsoft Exchange server, Search starts from a public Microsoft Exchange folder. Using the Exch protocol, Search gathers all messages and their attachments in the folder.

Using existing or created links, Search tracks the documents it collects in a transaction log. In a Web crawl, Search uses the links it finds in each document; in a file crawl, it creates links for all the documents in the starting directory and its subdirectories; and in a Microsoft Exchange crawl, it creates links for all the messages in the public folder. In each kind of crawl, Search marks the links as pending in the transaction log. After Search collects a document, it marks that document's link as done and follows the next pending link in the transaction log to another document, collects it, and so on.

Search also maintains a crawl history, which is a record of all links that have been crawled. Before adding a new link to the transaction log, Search checks the crawl history to ensure that links are not crawled twice. In case of a power failure or other disruption, Search uses the transaction log and crawl history to resume the crawl from where it left off.

The first time Search performs a crawl, it executes a full crawl: starting with an empty catalog, following every link in the transaction log, and collecting every document. To save time and bandwidth on subsequent crawls, Search performs incremental crawls, starting with the previous catalog and gathering only those documents that have changed since the last crawl.

Extracting Information

When Site Server Search gathers a document, it opens it and extracts information that it will use to create an index and a catalog. This information includes the full-text contents, links, and document-specific properties such as the author's name or file size. To accommodate the diversity of document formats, Search uses filters, which are plug-in modules available for each file format that conforms to a Microsoft standard IFilter interface. Search includes filters for common formats such as HTML, Microsoft Office documents, and plain text files. Filters for PDF and other file formats are available from third parties.

Because documents may contain a variety of natural languages, Search automatically detects the language of each document it encounters. Like other search technology, Search reads the HTML language tag. However, customer research has shown that some administrators do not fill in this HTML tag. Therefore, Search uses patented language-detection code to determine the language of particular information even if the HTML tag is not completed. Search then stores the language as a property that can be searched. This is important for organizations, such as multinational corporations, that have intranet content in several languages. The administrator can simply set up Site Server Search to crawl the entire intranet, and end users can search for content only in the language in which they are interested.

Creating an Index

After extracting the desired information from a document, Site Server Search indexes its words and their locations. For example, the index might indicate that the word "offer" is word No. 423 in Default.htm and word No. 638 and No. 982 in Products.doc. This index enables a quick and efficient search for documents containing certain words, for example, advanced search operations such as phrase searches (searching for "white elephant" means looking for documents with "white" followed by "elephant") and proximity searches (searching for all documents in which "big" occurs within 50 words of "house").

In order to minimize network traffic, Search supports distributed indexing, which places indexers where the data resides. In addition, to prevent the index from becoming bloated with words that do not help site users find documents, Search ignores "noise words" such as "a," "and," and "the." Search provides customizable noise-word lists for nine languages.

Compiling and Propagating

Once Site Server Search has finished indexing each document, it compiles the information into a catalog. Then it propagates the catalogs, that is, copying them to hosts that can be searched by site users. Such searches can be limited in keeping with security requirements by use of a "mixed" catalog. This way, site administrators can include both public and secure documents in a single catalog, while at the same time providing broad access to the former and limited access to the latter.

Two types of authentication supported by Search make mixed catalogs possible: NTLM (for documents on a Windows NT file system and Microsoft Exchange messages) and WWW-Basic (used by all HTTP servers). For example, when crawling for documents on a Windows NT file system, Search includes Windows NT Access Control Lists in document catalogs by using an account with administrator privileges. When site users search a catalog, they are limited to those entries for documents to which they have access. Any documents that Site Server gathers from HTTP and FAT volumes will be available to all site users.

4. Personalized Delivery

The Personalization & Membership (P&M) capabilities of Site Server increase the efficiency of intranet-site processes by making content more relevant to users. The personalization capabilities identify the content that is to be matched with specific users and the membership capabilities help direct the content to those users.

To enable the personalization of intranet pages, P&M provides the Rule Builder tool and a set of design-time controls for rapid development with Microsoft FrontPage® 98 Web site creation tool or Microsoft Visual InterDev™ Web development system. The Rule Builder facilitates the creation of rules based on user properties. Rules can be defined within sections of an intranet page or e-mail message that define what content to display, or other action to take, and under what conditions. Rules can be built to personalize content using simple name tags or to specify sophisticated sets of condition tests, content-retrieval parameters, and output formatting based on user information and/or system variables such as date or time. Rules can be organized into rule sets and organized by priority.

For security, P&M works with Windows NT Server to provide user authentication and access control to protect both content and the Membership Directory. Credentials can be stored either in the Windows NT Server directory database (using Windows NT Authentication mode) or in the Membership Directory (using Membership Authentication mode).

Enlisting Push for Proactive Delivery

With the Push model, Site Server uses three different methods to provide content to users: direct mail, channels, and personalized Web pages. Direct mail is delivered to the user's mail client, while channels and personalized Web pages are delivered to a user's Web browser.

Channels

Channels provide an organized and structured way to deliver a Web page, any type of document or file, and applications. Channels provide immediate desktop visibility, access to a wide audience, and a way to structure and organize information for users. To help users access the information they want, a channel can be focused on a particular theme or subject. For example, a channel that delivers departmental updates might contain minutes from the department's last meeting, a recent press release, a spreadsheet of the organization's latest product offerings, and a Web page with URLs to relevant sites.

Push relies on two technologies when using channels to deliver personalized content to users: the Active Channel Server and the Active Channel Multicaster.

Active Channel Server

The Active Channel Server provides extensive channel-management and channel-delivery options and can help manage network bandwidth by sending a single copy of data when delivering information to multiple users. The Active Channel Server uses Channel Definition Format (CDF) technology to automate information delivery to users' desktops. All information about a channel and its items is stored in a CDF file, which the Active Channel Server presents to users as a subscription link on a Web page. Once a user subscribes to a channel, the user's browser retrieves the CDF file, reads its contents, and places links to the channel and its content items on the user's computer. If the user's browser precaches such items, the user can view them offline. Otherwise, the user must be online to access a channel's items. Users are notified automatically of any changes or updates to the channel.

CDF files use HTML syntax and contain tags that store information about channels and their content items, including URLs and scheduling information. Ultimately, CDF files specify which channels users see, how content items are displayed, and when channels are delivered to users.

For most delivery methods, CDF files are the key to delivering channels to users. Client browsers receive CDF files, interpret the file tags, and display links to channels and their content items. Users need not go to the intranet in order to retrieve a content item. They simply click the link they want, and their browser retrieves the content item.

Note that users must have a browser that supports CDF technology before they can subscribe to and receive channels. Because Active Channel Server places all URLs for channels and content items within Anchor tags, any HTML-enabled client browser can read the CDF files generated by Active Channel Server. Microsoft Internet Explorer, version 4.0 and later, provides full support for CDF technology.

Site Server Push offers the following methods for delivering channels and their content items to users: Managed Push, Personalized Push, Multicast Delivery, and Mail Delivery. With Managed Push and Personalized Push, the user's browser periodically requests updates from a Web server, pulling the updated information to the user's computer. The information is delivered to the user's desktop without any action by the user.

With Multicast Delivery, the Active Channel Server first generates an updated CDF file, and then Active Channel Multicaster crawls the CDF file and delivers the data contained in the file to client computers. To receive multicast files, client computers must have the Internet Explorer Multicast Delivery Agent installed. (For more on the Active Channel Multicaster and multicast delivery, see the following section).

With Mail Delivery, users receive an e-mail message that contains links to channels and content items. To access the channels, users simply click the links.

Active Channel Multicaster

By enabling multicast delivery, Active Channel Multicaster saves network bandwidth in two ways. First, it distributes channels and content items to users who request them rather than to all users on the network. Second, it delivers the content items themselves, rather than a CDF file pointing to content items on a Web server. Client browsers can cache the content items for offline browsing, thereby reducing network traffic.

Active Channel Multicaster can distribute any Web content, including CDF files. When distributing CDF files, it gathers the content items referenced in the file and multicasts them to self-selected users. These content items may originate from a CDF-based channel or a Web site.

Once a site administrator has determined which channels to multicast, the administrator assigns the channels to a multicast project. If the content is not defined by a CDF, the intranet site where it is found can be assigned to a multicast project. Once administrators have configured their multicast projects on their Active Channel Multicaster host, it begins multicast announcements. In these announcements, the host directs the client computer where to receive the multicast and identifies the content source.

Using Knowledge Manager

Knowledge Manager is a Web-based sample application that demonstrates how the Site Server Search, P&M, and Push features can be combined to address the problems of managing information delivery on an intranet. Using Knowledge Manager, users can have a single place to go and search, browse, and share information within their organization.

Knowledge Manager is comprised of five areas:

The Search Center: Users can either search for information from previously built search catalogs or browse through content in a structure that an administrator defines through the Site Vocabulary.
My Brief: Users can create their own collections of knowledge, called knowledge briefs. By default, all users have one personal knowledge brief called My Brief.
Shared Briefs: Users can create or access public briefs to share knowledge.
Brief Delivery: Users can specify whether they want to receive briefs through e-mail or channels.
Channel Area: Users can receive additional information through channels. The Channels area is the one place where users can find available channels.

Knowledge Manger includes a sample site vocabulary that can easily be extended and has established security guidelines:

Users can see only their own private briefs.
Search results presented to users correspond to user permissions.
Windows NT access control lists are used to manage, set up, and administer Knowledge Manager.

Knowledge Manager maintains three separate areas of security: for site users, for information, and for administration. To maintain site-user security, Knowledge Manager uses Windows NT Challenge/Response authentication for site users who are logged on and Allow Anonymous for site users who are not logged on. This means that all site users have immediate access to Knowledge Manager pages regardless of whether they are logged on. To keep briefs secure, Knowledge Manager stores briefs and sections of briefs based on the site user's globally unique identifier.

To maintain information security, Knowledge Manager enables administrators to set up Search catalogs so that each site user has access only to information he or she has permission to view. For example, employee data may be available to the Human Resources department, but not to all employees.

To maintain administration security, Knowledge Manager defines a Site Server Knowledge Administrators group and a Delivery Administrators group. Individuals in the Site Server Knowledge Administrators group can modify any aspect of the Search, Push, and P&M configuration, including creating Search catalogs and modifying the Site Vocabulary. Individuals in the Delivery group can perform all the functions of the Knowledge Administrators group in addition to other administration functions.

Conducting Comprehensive Web Site Analysis

Business want ways to measure success of an intranet based on how effective a site is at attracting employees and their continued reuse of a site. Another key element to maximizing intranet investment is to ensure a site's consistency, quality, and conformance to policy.

To address these needs, the Site Server analysis features provide three types of Web analysis—usage, content, and user.

Usage: Analysis transforms raw hits recorded in server log files into valuable information about the requests, visits, and users that interact with a Web site in order to maximize return on Web investment.
Content: Analysis can crawl an intranet capturing content, content attributes, and site structure. This allows you to identify problems, such as duplicates and orphans, unavailable resources, and content with excessive load sizes.
User: Analysis integrates with other Site Server features and BackOffice servers, including Personalization, Search, and Proxy Server, to build greater data richness into your usage and content analysis and to expand the breadth of analysis opportunities.

Usage Analysis: Maximizing ROI

Each time a visitor interacts with an intranet site, the server's log file records a "hit." This information details information, such as where intranet users came from, what they saw, how much time they spent visiting certain areas, and what they saw last. Site Server Analysis transforms the large amount of raw log-file data into useful reports, such as on the number of repeat visitors, that detail the success of a business's intranet site.

Importing Server Logs and Other Data Source Files

Analysis provides two tools for importing usage data: Usage Import and Custom Import.

Usage Import imports log files from diverse intranet servers and applies inference algorithms to reconstruct the actual requests, visits, and users interacting with the intranet before storing the information in a database from which reports can be produced.
Custom Import augments server log file data by importing user attributes (self-describing attributes recorded in Personalization), content attributes (such as topic, author, and type for each entity), and custom data files (such as department employee or job category files). Site Server is the first product to support the integration of usage data from a variety of sources.

Analysis extends beyond mere hit analysis and applies sophisticated inference algorithms to reconstruct more meaningful measures of site activity such as requests, visits, and users. For example, when a user requests a Web page with three graphics, the log file records it as four hits (or more, counting overhead), but Analysis interprets the activity (accurately) as a single request. Analysis defines a visit as a series of requests by a user and determines that a visit is over when a user does not request any more content for a specified period of time (default timeout period is 30 minutes). To identify a user, Analysis looks first for cookies, second for registered usernames, and third for host names.

Usage Import automatically detects more than 25 log file formats including those from Microsoft, Netscape, Lotus Notes (including separate referrer and user-agent logs), and Apache. Additionally, for Internet Information Server 4.0 users, Analysis automatically installs a user-identification filter to generate unique cookies for each user and configures Internet Information Server to extend the standard logging (W3C extended format) to include referrer, user agent, cookie data, and query strings.

Usage Import also includes support for analysis of Search query strings and proxy server log files from Microsoft and Netscape. During the import process, Analysis automatically detects and stores information from query strings generated from Search. This information includes the actual query, catalog name, and number of items returned. You can use this information to improve employee productivity with your search service by identifying what employees are looking for and the success of their queries. Identifying what Internet resources employees use most is made easier with support for proxy server logs. IT professionals can evaluate if the organization is productively using the Internet as well as measure the bandwidth savings from caching popular Web sites.

Usage Import provides enterprise-level server management, aggregation of data from multiple servers and sites, and maintenance of log file data inventory. Usage Import can also analyze sites based on multiple Web servers. For large sites distributed or replicated across multiple servers, analyzing user activity one log at a time often results in: (1) overstating the number of visits, and (2) incomplete visit profiles. During import, multiple log files can be merged together to unify visits that span log files. The import process can also process out-of-sequence log files eliminating the need for any preprocessing.

The Analysis database enables the cross-site sharing of user and content attributes. The database, stored in either Microsoft Access or Microsoft SQL Server, is organized into four categories: organizations, users, visits, and requests. These categories are related because users belong to organizations, users generate visits, and visits generate requests.

Reporting Usage Data

The Report Writer queries the Analysis database and generates reports about usage and site activity. With Report Writer, administrators, content developers, and managers can create their own reports or choose from among 46 predefined reports, including reports on intranet usage by department and content topic. They can base their reports on data from a wide variety of sources, including other Site Server components and external databases, and produce them in HTML, Microsoft Word, or Microsoft Excel formats. These reports can include summary information (or provide detail on specific users) on request and visit trends, navigational patterns, and server load patterns, among other areas of interest.

Administrators can create their own reports by modifying existing predefined reports or building new ones from scratch. For power users, more than 200 report-building dimensions provide the flexibility to configure data to answer the most pressing questions about a site. To begin the creation of a new report, a calculation inventory offers more than 60 predefined calculations that can be dragged and dropped into a new report definition. (A report definition consists of one or more calculations).

Users can also execute and view the results of any predefined report (including their own creations) from a Web browser.

Analysis also enables intranet administrators and others to perform deeper usage analysis with innovative usage data visualizations. With the Hyperbolic View found in Content Analyzer (see details later in this section), administrators can animate their site structure with usage data. Through an intuitive interface, administrators can use Content Analyzer to identify how traffic moves through their site, targeting high- and low-traffic zones as well as the most-traveled path through a site.

User Analysis: Targeting a Message

A key promise of an intranet is its ability to make employees more productive and to reduce costs. Delivering the right content at the right time can help make the intranet more effective for employees, thereby maximizing a business's intranet investment. To this end, Analysis works in conjunction with Personalization to enable administrators to deliver targeted messages over personalized pages, e-mail, or channels. Analysis details the effectiveness of targeting efforts and creates targeted user lists for email campaigns based on users' online behavior. Personalization can also update users' profiles based on these user lists.

You can use targeted e-mail campaigns to deliver specific messages to particular segments of your employees. Campaigns can be generated to inform employees when new products are released, benefit policies change, and other company news and information become available that is relevant to a particular segment of your employees. Using business rules designed in the Direct Mailer, Analysis creates a user list based on users' online behavior. For example, the Direct Mailer could request that Analysis provide a list of all sales people that have visited a particular product's section of an intranet at least three times in the prior month. An e-mail could then be sent to this group asking for feedback on the effectiveness of the section, for instance.

Content Analysis: Ensuring Quality and Consistency

The Analysis tool known as Content Analyzer maps intranet sites to extract vital content-attribute information and create site maps from the results. Content Analyzer uses a "spider" to record the links on a page and then follow those links to locate new pages and new links, storing all the information in a site map. Each map provides a catalog of the Web objects in a site (such as HTML pages, Java and PDF files, graphics, links) and properties associated with those objects. Once created, a site map provides a visual representation of a site that can be used extensively for navigation, site analysis, and site maintenance.

Searching Content

Powerful search capabilities in Analysis enable intranet administrators to locate any information in their site by searching for resources such as text strings, property (for example, size or URL), object type, author, expiration date, or other user-defined parameters. They can search the entire map or only what's currently displayed. Through Content Analyzer, they also can access a set of common searches, such as pages with load size greater than 32K, top entry pages, and pages with external referrers.

To simplify the process whereby content authors insert their name, last edit date, and other content attributes directly into their page, Site Server provides the Tag Tool, a feature of Site Server Publishing. With this tool, authors can incorporate META tags in their pages. They can then search and sort on the META tag data and identify, for instance, their own pages and when the page was last changed.

5. Using the Microsoft Platforms Technology

Site Server is a member of the BackOffice family and designed to run on Windows NT Server and Microsoft Internet Information Server. As such, it is optimized for those products and can help reduce the complexity, increase the performance, and lower the cost of intranet ownership.

Integration with Windows NT Server

The integration of Site Server with Windows NT Server provides businesses many advantages, particularly in terms of administration and security. With Site Server, businesses have a choice of administration options:

Microsoft Management Console, a Windows-based interface, allows administrators to manage all Site Server capabilities from a single server to multiple remote servers.
WebAdmin, a browser-based application, provides administrators a way to manage a server or set of servers. WebAdmin is especially useful for remote administration of servers not running Windows NT Server.
A command-prompt interface enables administrators to manage Site Server capabilities by entering text-based commands. The command-prompt interface enables administration using Windows Script Host.

For security, Site Server supports Windows NT Server logon, the standard challenge/response security mechanism provided with Windows NT Server for creating user permissions and access rights to personalized pages. In addition, a business with existing Windows NT Server installations can enjoy a single, common security mechanism for both its corporate network and its file and intranet servers.

Integration with Internet Information Server

The integration of Site Server with Internet Information Server provides additional benefits to intranet developers, administrators, and users. Site Server can be used to build intranet sites that are viewable by any Web browser. Through an Internet Information Server data-security protocol layered between HTTP and TCP/IP, Site Server supports data encryption, server authentication, and message integrity for a TCP/IP connection. Through support for both HTTP and FTP, Site Server can pull content for distribution to multiple intranet servers.

With Site Server and Internet Information Server 4.0, developers can use standard Internet protocols to create sophisticated client/server applications, thereby reducing development and management costs. Support for industry-standard protocols also means that business managers and site developers can easily integrate Site Server into an existing information-technology infrastructure.

Integration with Other Microsoft Technologies

Site Server also takes advantage of many other technologies available within the Microsoft BackOffice family. Dynamic content can be created and delivered through the Active Server Pages technology and accessed through APIs such as the Active Data Object. Active User Object and ADSI simplify the access of user information, and Index Server supports full-text indexing. Site Server also builds on other Microsoft products, such as FrontPage and Visual InterDev, to provide a comprehensive intranet development and authoring solution. Such a fully integrated solution enables developers to take advantage of existing knowledge and expertise in the creation and deployment of intranet content.

As part of its integrated design, Site Server also is extensible through documented APIs and code samples included in the software developers kit (SDK). For even greater flexibility, Site Server is compatible with UNIX server environments. This means that a site developer can analyze data from UNIX-based intranet servers and deploy content to UNIX-based production servers so as to integrate Site Server in virtually any intranet site environment.

6. The Microsoft Site Server 3.0 Solution

Businesses are increasingly using intranets as a way for employees to easily publish, search, and deliver the information they need to make decisions and work more productively. Employees, however, face an "information glut" and may not be able find or share the information they need. By deploying Site Server 3.0, businesses can use the intranet to efficiently gather the collective expertise of the organization from wherever it resides—in Web sites, databases, file servers, e-mail—and deliver it to facilitate the sharing of knowledge and to improve business productivity.

Whether you are creating a new site or enhancing an existing one, Microsoft Site Server 3.0 enables you to:

Publish information easily. Control how information is published to the Web by providing authors an easy-to-use, structured content submission, posting, approval, and deployment process.
Search for information throughout your organization. With the rapid growth of intranets, it is increasingly difficult for users to find necessary business information. Site Server enables enterprise-wide searches across Web sites, network file servers, ODBC databases, and Microsoft Exchange Server folders.
Deliver relevant information to users and analyze site usage. The enormous amount of information available on an intranet can be overwhelming. Deliver the information that users need by using personalized Web pages, channels, and e-mail. By measuring the effectiveness of your intranet, you can ensure that it is adding to the productivity of your business.

For More Information

For the latest information on Site Server, check out our World Wide Web site at https://www.microsoft.com/siteserver/default.htm.