We strongly recommend that you familiarize yourself with the following sections before installing any networking software:
Remote administration allows you to manage one system at a time, as if you were root on another system. This concept is familiar to most UNIX system administrators through commands such as rlogin, telnet, and rcmd. Your SCO OpenServer Desktop or Enterprise system extends these abilities by allowing you to view and administer remote systems from your management workstation to perform tasks such as:
Distributed administration allows you to manage entities (for example, users and printers) that may be associated with several systems in a network rather than being attached to one system. It is particularly important for enterprise installations where large numbers of systems must be maintained in a consistent stated. Examples of distributed administration include:
Both remote and distributed administration can be done with
SCOadmin managers, which are graphical administration
interfaces supplied with the SCO OpenServer.
For more information on SCOadmin managers, see
``Administering other systems with SCOadmin managers''.
About clients and servers
SCO networking is based on the ``client-server'' model.
In its simplest form, a
``client'' is a program that requests a service and a
``server'' is a program that provides a service. In a
networked environment, client programs frequently issue
RPCs (Remote Procedure Calls) to request that an
operation be performed; a server responds to the RPC
by executing procedures to perform the operation and sending a
response to the client. The terms may also refer to machines;
``server'' may refer to a host whose files or services are made
available through RPCs, and ``client'' to the
requesting host.
The implications of the terms ``client'' and ``server'' may vary in specific networking components. The following table summarizes some of these variations.
Table 1-1 Examples of client-server implementations in SCO networking components
--------------------------------------------------
Component Client Server
--------------------------------------------------
DNS any program that any program that
(Domain queries for host responds to
Name names and queries for host
Service) addresses information
--------------------------------------------------
NFS any host any host that
(Network requesting that ``exports'' file
File a remote file systems for
System) system be remote mounting
mounted locally
--------------------------------------------------
NIS any host that any host that
(Network requests NIS- fulfills requests
Information managed for NIS-managed
Service) information information
across the
network
--------------------------------------------------
SNMP also called also called
(Simple ``management ``agent''; any
Network station''; any program that
Management program that responds to
Protocol) queries for queries from a
system status management
information from station
a remote system
SCO OpenServer systems provide both client and
server administrative functionality; as servers, SCO systems can
easily manage other machines using standard technology, while as
clients, they can be managed remotely using these technologies.

SCO Network Adapter Drivers are available in two forms, both included in your SCO OpenServer system:
Many network adapter vendors provide SCO-compatible network
adapter drivers. Most of these vendors are listed in the
SCO Compatible Hardware Web Pages.
To use an adapter not listed there,
contact the vendor.
Types of networking maintenance tasks
After you decide how your machine fits into the network, you
need to install and configure the appropriate protocol packages
as described in
Chapter 1, ``Configuring network connections'' in Configuring Network Connections.
You also need to update the networking files on other machines so that
they know of the new machine's existence. This configuration ensures,
among other things, that:

The following are not included with SCO OpenServer:
Included in Microsoft LAN Manager for SCO Systems, available separately.
These are available from other vendors.
Networking components
SCO networking functionality is provided in several components
included with the SCO OpenServer distribution. The distribution
products and component packages may be viewed from the Software Manager;
for more information, see
``Examining software packages'' in the SCO OpenServer Handbook.
The following components are included with the SCO OpenServer Connectivity package:
The full LAN Manager server system, including:
The full DCE® (Distributed Computing Environment) runtime system, including:

Protocols are more accurately termed protocol
``stacks'' or protocol ``suites'' because
the communications functions are complex and are usually
divided into independent layers or levels.
A stack is a collection of protocol layers that implement
network communication.
The protocol associated with each layer communicates only with the
layers immediately above and below it, and assumes the support of
underlying layers.
Lower layers are closer to the hardware and
higher layers are closer to the user. The number of layers and
tasks that each layer performs depends on which stack you
are using.
Table 1-2 Supported protocol stacks
-------------------------------------------------------------------
Provided by Provided by other Supported by SCO
Protocol stack SCO System SCO products network drivers
-------------------------------------------------------------------
TCP/IP yes - yes
IPX/SPX yes - yes
NetBEUI - yes yes
SCO OpenServer also provides two operating system packages
that permit UNIX and XENIX® systems
to communicate as part of a remote network:
Although these packages are not protocol stacks, they are discussed here because they provide low-cost alternatives to LANs based on specialized hardware and software.
The relative advantages of the SCO connectivity packages are:
TCP/IP is a set of protocols and programs used to interconnect computer networks and to route traffic among different types of computers. These protocols describe allowable data formats, error handling, message passing, and communication standards. Computer systems that use TCP/IP speak a common language, regardless of hardware or operating system differences.
Many large networks conform to these protocols, including the Internet. Thousands of computers at universities, government agencies, and corporations are connected to a network that follows the TCP/IP protocols. Any machine on the Internet can communicate with any other. Machines on the Internet are referred to as hosts or nodes and are defined by their Internet (or IP) address.
TCP/IP provides:
A datagram consists of header information and a data segment. The header contains information about routing and processing the datagram. Datagrams can be further fragmented into smaller pieces, depending on the physical requirements of the networks they cross. For example, when a gateway sends a datagram to a network that cannot accommodate the datagram as a single packet, the datagram must be split into pieces that are small enough for transmission. The datagram fragment headers contain the information necessary to reassemble the fragments into the complete datagram. Fragments do not necessarily arrive in order; the software module implementing the IP protocol on the destination machine must reassemble the fragments into the original datagram. If any fragments are lost, the entire datagram is discarded.
See also:
The primary purpose of TCP is to avoid the loss, damage, duplication, delay, or misordering of packets that can occur under IP. When IP forwards datagrams, individual datagrams may or may not arrive, and they probably will not arrive in the order in which they were sent. TCP adds the reliability that IP lacks. Also, security provisions such as limiting user access to certain machines can be implemented through TCP.
TCP provides reliability by using checksums (error detection codes) on the data, sequence numbers in the TCP header, positive acknowledgment of data received, and retransmission of unacknowledged data.
See also:
See also:
Translates between DARPA Internet and Ethernet addresses. See arp(ADMN) and arp(ADMP).
Handles error-message and protocol control for TCP/IP. See icmp(ADMP).
Provides both synchronous and asynchronous network connections over a serial line. See ppp(ADMP) and Chapter 11, ``Configuring the Point-to-Point Protocol (PPP)''.
Translates between Ethernet and DARPA Internet addresses.
Enables IP over serial lines. See slip(ADMP) and Chapter 12, ``Configuring the Serial Line Internet Protocol (SLIP)''.
Used by MMDF to send mail via TCP/IP.
Performs distributed network management functions via TCP/IP. See
Chapter 10, ``Configuring the Simple Network Management Protocol (SNMP)''.
Network servers and databases
In the UNIX system, most server programs are started by a
super server, called the ``Internet daemon''. The Internet
daemon, /etc/inetd, acts as a master server for
programs specified in its configuration file, /etc/inetd.conf.
It listens for service requests for these servers, and starts
up the appropriate program whenever a request is received.
The configuration file includes information about:
daytime stream tcp nowait root internalConsult inetd.conf(SFF) for more details on the format of the configuration file and the operation of the Internet daemon. Several data files are used by the network library routines and server programs. Most of these files are host independent and are updated only rarely.
Table 1-3 Network database files
---------------------------------------------------------------------- File Manual reference Use ---------------------------------------------------------------------- /etc/hosts hosts(SFF) host names /etc/networks networks(SFF) network names /etc/services services(SFF) list of known services /etc/protocols protocols(SFF) protocol names /etc/hosts.equiv rshd(ADMN) list of ``trusted'' hosts /etc/ftpusers ftpd(ADMN) list of ``unwelcome'' ftp users /etc/inetd.conf inetd(ADMN) list of servers started by inetdThe files distributed are set up for Internet hosts. Local networks and hosts should be added to the databases to describe the local configuration.
Table 1-4 TCP/IP Commands
--------------------------------------------------------------------
Command Purpose
--------------------------------------------------------------------
ftp file transfer between machines running TCP/IP (these
machines may or may not run the same operating system)
rcmd remote command execution on another UNIX system
rcp file copying between two UNIX systems
rlogin remote login on another UNIX system
ruptime status display of local network machines
rwho display list of users logged on to local network machines.
telnet remote login on a machine running TCP/IP (these
machines may or may not run the same operating system)
IPX is a datagram service protocol which allows individual packets to be sent to and received from user processes. IPX is a connectionless service; it does not support the concept of a connection or reliable delivery. However, guaranteed services like SPX have been built on top of IPX. IPX is used in situations where a guaranteed service is not required or where an occasional lost packet is not critical.
For more information, see ``Internetwork Packet Exchange (IPX)''.
SPX is a connection-based, reliable, sequenced
transport protocol which provides both guaranteed packet
delivery and flow control.
Packets are received in order by the destination endpoint
and the speed at which the sending and receiving processes
manipulate packets is regulated.
SAP (Service Advertising Protocol) is a method by which networks can advertise available network services. SAP allows service nodes (such as file servers, print servers, and application servers) to advertise their services and addresses. The SAP daemon uses the SAP protocol to advertise these services.
For more information, see ``Service Advertising Protocol (SAP)''
The Routing Information Protocol (RIP) facilitates the exchange of routing information on a NetWare network. For more information, see ``Routing Information Protocol (RIP)''.
The Novell Virtual Terminal (NVT) protocol is a guaranteed-delivery, connection-oriented protocol. It is built on top of IPX and uses IPX to send packets and receive positive acknowledgments of packet delivery.
Several IPX/SPX commands, described in detail on their manual pages, provide end users with networking capabilities:
Table 1-5 IPX/SPX Commands
----------------------------------------------------
Command Purpose
----------------------------------------------------
dnvt(PADM) monitor connections to an NVT
server
drouter(PADM) display the contents of the Routing
Information table
getlan(PADM) display information about IPX/SPX
and LLI interfaces
ipx(PADM) start, stop, or restart the IPX/SPX
protocols and the NVT server
nlogin(PADM) enable remote login via IPX and NVT
nping(PADM) test the network connection to a
host
rrouter(PADM) clear and update the Routing
Information table
showsvcs(PADM) monitor and/or validate network
names, addresses, and services
track(PADM) test the operation of the SAP
daemon
See also:
See also:
This is included with Microsoft LAN Manager for SCO Systems.
These are available from various vendors.
See also:
Each protocol supports the following:
Authentication of connection requests with CHAP
(Challenge-Handshake Authentication Protocol) or PAP
(Password Authentication Protocol), which provides additional security.
Multiple levels of debugging data, which appear on the console and in the system logfile.
Error detection through the use of a checksum program.
Connect-time negotiation of IP addresses, authentication methods, compression, and other configurable parameters.
Two-level packet handling, where telnet, rlogin, and ftp packets have a higher priority than all other packets.
Support for multiple protocols. SLIP only supports the IP protocol.
Support for the SCO Global Access browser and other products (such as other World Wide Web browsers) that make use of an httpd server.
Support for SNMP queries for objects in the PPP Link Control Protocol MIB and the PPP Network Control Protocol MIB.
To configure serial lines:
Serial line administration consists of:
UUCP uses a batch method to manage communications traffic, storing (or ``spooling'') requests for later execution when actual contact is made between systems. When UUCP commands are executed, work files and any data files needed are created in /usr/spool/uucp and its subdirectories. The program uucico scans these directories for the instructions contained in any work files and executes them. Although it is possible to execute commands immediately, most systems call other systems according to a daily schedule (usually during the evenings to reduce connection costs).
UUCP and cu configuration consists of:
UUCP and cu administration consists of:
The SCOadmin managers facilitate distributed and remote system administration of SCO OpenServer servers and clients. For information on using these managers, see ``Administering other systems with SCOadmin managers''.
SCO OpenServer supports the following types of servers:
Many applications, such as SCOhelp, SCOcalendar, and many relational database management systems, use a client-server model wherein the data resides on one or more application servers. To access the data, users on client machines run a program on the local CPU, which transfers data over a networking protocol such as TCP/IP. The program may be on a local filesystem or it may be mounted via NFS from a file server. In most cases, the user is unaware that the data does not reside on the local system.
In addition, many multiuser host applications are available for
SCO systems. You can access them remotely via your network
using telnet, rlogin, and other terminal
emulation programs.
File servers,
running programs such as Network File System (NFS),
Microsoft LAN Manager for SCO Systems, and NetWare,
allow users on client machines to transparently access files
from server-exported filesystems on the local host.
Use of file servers enables
you to more effectively balance the disk space load between various
machines on your network and to share files between machines running
different operating systems.
The networked installation capabilities of SCO OpenServer enable you to install the entire operating system or operating system components over the network.
Both mail transfer agents supported by SCO OpenServer, MMDF (the Multichannel Memorandum Distribution Facility) and sendmail, allow for operation over the network and have the ability to interact with other mailers through mail gateways.
Name servers, such as the Domain Name Service (DNS) and the Network Information Service (NIS), maintain repositories of network and host names and addresses, and are queried by programs such as mail and remote login programs operating on client computers. SCO OpenServer supports a variety of local, remote, and distributed name services. Using a name server becomes appropriate as your network grows large or connects to the Internet.
Several programs and protocols provide a framework for managing your users, systems, and network resources.
Most SCOadmin managers allow you to remotely manage system users and resources, such as printers, filesystems, and user accounts.
The Simple Network Management Protocol (SNMP) enables you to remotely manage other network machines and devices running SNMP. For example, you can mark an interface as being down, modify routing information, or generate statistics that pinpoint performance issues.
With the Network Information Service (NIS) or the TCP/IP program
rdist(TC),
you can maintain central repositories of system files, such as
/etc/hosts, /etc/passwd, and /etc/group,
which are either served to or pushed onto other machines as appropriate.
You can print from your local SCO OpenServer system:
Many database programs depend on time synchronization across the network to allow for effective file locking and authentication. The SCO OpenServer Desktop and Enterprise systems include two TCP/IP protocols, the Network Time Protocol (NTP) and the Time Synchronization Protocol (TSP), which allow for time synchronization on Internet-connected and non-Internet-connected networks, respectively.
To effectively manage user accounts across the network:
With remote administration, you can manage one remote system at a time. For example, you can add a printer to another system. To add that printer to additional systems, you need to repeat the procedure, once for each system.
With distributed administration, you can manage multiple systems at one time. For example, you can add a user to multiple systems in one step.
Generally speaking, remote administration entails minimal setup
but does require repetitive steps to accomplish a task (such as
providing a printer definition to multiple systems). Distributed
administration, while requiring additional overhead in configuration,
allows you to perform a task (such as distributing a user to
multiple networked machines) in one step or procedure.
Remote capabilities
The root user can accomplish the following tasks, through
remote administration, by opening another host to manage from
the Host menu of a SCOadmin manager:
Table 1-6 Subsystem authorizations
-----------------------------------------------------------
SCOadmin
Authorization Manager User can:
-----------------------------------------------------------
lp Printer Manager administer printers
backup Backup Manager perform backups
auth Account Manager administer accounts:
adding users, changing
passwords, and so on
passwd Account Manager change user passwords
sysadmin Filesystem Manager mount filesystems
The key enabling technologies for distributed user applications are the supported network protocol stacks and filesharing capabilities such as NFS and SCO Gateway for NetWare filesystems.
Your SCO OpenServer system provides these applications, which can be configured to operate locally or, with the Desktop or Enterprise systems, over the network:
The SCO Network File System (NFS) is a product that enables you to export and mount filesystems across a network. This allows users on a local machine, or client, to access specified files and directories from a remote machine, or server, without the time-consuming process of remote logins or machine-to-machine file copying.
In addition to exporting native UNIX filesystems, SCO NFS can export DOS, LAN Manager, and SCO Gateway for NetWare filesystems.
automount allows remote NFS filesystems to be mounted automatically and transparently. Whenever a user on a client machine running automount invokes a command that accesses a remote file or directory (such as when opening a file with an editor), the hierarchy to which that file or directory belongs is mounted and remains mounted for as long as it is needed. No mounting is done at boot time, reducing significantly the time needed to boot up.
The Network Lock Manager (NLM) network service consists of a loadable device driver and a set of daemons that permit both advisory and mandatory file and record locking on local files. Only advisory file and record locking are supported on remote SCO NFS files. The NLM package includes the SCO NFS Status Monitor, which works in conjunction with the NLM to determine when a remote host has recovered after a crash.
The Remote Execution (REX) service is a remote command service. It allows users to export their user environments to remote servers. These servers execute commands that can access files in the user's current directory and allow the execution of interactive processes such as full-screen editors. REX consists of a set of utilities, commands, and library functions.
File sharing is also provided by these SCO OpenServer interconnectivity components:
Table 1-7 SCO system filesharing interconnectivity
---------------------------------------------------------
Operating system SCO server SCO client
---------------------------------------------------------
DOS LAN Manager* LAN Manager Client
-- SCO Gateway for NetWare
PC-Interface --
Windows LAN Manager* LAN Manager Client
-- SCO Gateway for NetWare
PC-Interface --
OS/2 LAN Manager LAN Manager Client
NetWare -- SCO Gateway for NetWare
To configure file sharing using NFS, you must:
There are five ways to remote mount exported filesystems:
NIS can be used in conjunction with automount to ensure that each NIS system can automatically mount filesystems from the same set of NFS servers. In this case, automount maps are maintained on the NIS master server and distributed as NIS maps to other NIS systems. Local variations can be made to NIS-distributed automount maps. See ``Managing automount with NIS''.
Administering NFS entails the regular execution of common tasks, including:
See also:
You can prepare an installation server, then perform an initial installation or upgrade onto clients with a bootp ROM or a specialized boot floppy.
After initial system load, you can use custom(ADM) to install (or upgrade) software to or from a remote machine. You can also remove, examine, and verify software on remote machines.
When you install a new component via custom, either over the network or from media, the component is placed in a software storage object (SSO) on your hard disk. SSOs allow multiple releases of the same software component to reside on your system simultaneously; users may then (depending on the system's configuration) access the version of their choice.
See also:
Once you initially install your SCO OpenServer system,
no additional tasks are required to perform component installations
and upgrades across the network.
Distributed electronic mail
Electronic mail (e-mail) on the SCO OpenServer system is handled by
two distinct processes:
On a networked system, the MTA on the sending system has the additional task of determining whether the mail should be delivered to a local user or sent to another machine on the network, and, if sent to another machine, what communications channel (such as SMTP, the Simple Mail Transport Protocol) to use. When the MTA on the destination system receives the mail message, that MTA forwards the message to the appropriate user.
SCO OpenServer provides a choice of Mail Transfer Agents:
(Multichannel Memorandum Distribution Facility), the SCO MTA that is included with the SCO UNIX operating system
a commonly used alternative MTA that is also included with the SCO operating system
MMDF offers several substantial benefits over sendmail, including:
Both MTAs can use a variety of hostname-to-Internet address mapping schemes; the most effective when connecting to the Internet is the Domain Name Service (DNS).
To configure electronic mail, you must complete the following procedures:
Administering a mail system includes regularly:
Name services provided by the SCO OpenServer Desktop and Enterprise systems to facilitate this mapping include:
The /etc/hosts file lists pairs of machine names and IP addresses. Update this file on each machine when a network is small, relatively stable (adding or removing machines infrequently), and not connected to the Internet. See the manual page for hosts(SFF) for more information.
You can use the TCP/IP utility, rdist,
to distribute /etc/hosts to a
list of designated machines on your local network if the network
grows to the extent that editing individual files becomes a burden.
See
Chapter 20, ``Distributing files remotely over TCP/IP'',
and the manual page for
rdist(TC)
for more information.
Use DNS to create, maintain, or access a distributed database of names and addresses. These capabilities become critical as your network grows larger or if you connect to the Internet. Many other programs, such as SCO Global Access and various Mail Transfer Agents, either require or make effective use of DNS. See Chapter 6, ``Configuring the Domain Name Service''.
Use NIS to serve /etc/hosts data to diskless or dataless NIS clients, or to serve other NIS clients where restricted access to the actual /etc/hosts file is desirable. See Chapter 9, ``Configuring the Network Information Service (NIS)'', for more information.
Common tasks associated with name services include:
All SCOadmin managers that allow for remote administration issue remote commands (rcmds) over TCP/IP. Each user, including root, who wants to use this capability must have an equivalent account on each system to be managed.
For more information about user equivalence, see ``Establishing user equivalence''. To configure user equivalence, use the User Equivalence Manager.
With NIS, you can:
With SNMP, you can both monitor and control TCP/IP-based networks by querying remote hosts and devices for data (such as the host's status, routing table, and system activity) and by issuing commands to remotely manipulate such data.
SNMP is useful in the following ways:
For more information, see
``How SNMP works''
and
Chapter 10, ``Configuring the Simple Network Management Protocol (SNMP)''.
Distributed printing
Local printing generally refers to sending print jobs via
lp(C)
to a printer attached to your computer via a serial or
parallel port. Distributed printing enables you to use
the network to send print jobs to printers attached to
networked hosts and to send jobs from remote hosts to
a local printer.
To print remotely from your SCO OpenServer Desktop or Enterprise systems, you can:
To configure remote printing, you must complete the following procedures:
Use the Time Synchronization Protocol (TSP) or the Network Time Protocol (NTP), both members of the TCP/IP protocol suite, to synchronize clocks on your network.
Both TSP and NTP are fully supported in SCO OpenServer Release 5. Each has its own advantages, which are fully detailed in ``TSP guidelines'' and ``NTP guidelines''. The primary differences between NTP and TSP are:
NIS designates a single server as ``master'' of files and databases containing system administration information and distributes this information to all systems in an NIS ``domain''. Among the files distributed by default are /etc/passwd and /etc/group; user and group accounts managed by NIS are called distributed accounts. NIS also provides a means to exempt accounts from distributed management, so they are local to a given machine; such accounts are called local accounts. For more information, see ``Administering NIS users and groups''.
Although NIS-distributed accounts allow users to log in to any machine in an NIS domain, they must have a home directory on each of those machines. However, the automount facility can be figured to automatically NFS-mount their home directory wherever they log in. For more information, see ``Distributing home directories''.
The SCOadmin Account Manager provides a convenient, interactive way to perform most administrative tasks on both distributed and local accounts. See ``The Account Manager interface'' in the System Administration Guide.
You can use the useradd(ADM) command to add, modify, and delete remote account information from the command line or in batch jobs.
Common administrative tasks associated with distributed user accounts include:
Connecting to other operating systems
You can serve other operating systems from your SCO OpenServer
server:
Provide print, file, terminal emulation, and other services with PC-Interface. Provide print services with the pcnfsd(NADM) component of NFS.
Provide distributed applications services to NetWare clients with IPX/SPX.
Provide print services with pcnfsd(NADM).
If you install LAN Manager for UNIX Systems, you can also provide print, file, and terminal services to other systems running LAN Manager.
Your SCO OpenServer system can be a client to computers running other operating systems:
Consume print, remote login, and file services offered by LAN Manager servers, by using LAN Manager Client.
Consume print, remote login, and file services offered by NetWare servers, by using SCO Gateway for NetWare.
Consume print, remote login, and file service offered by
LAN Manager servers, by using
LAN Manager Client.
Accessing DOS and OS/2 servers with LAN Manager Client
The LAN Manager Client allows you to connect your SCO OpenServer Desktop
or Enterprise system to resources on these servers:
Users on a LAN Manager Client can:
LAN Manager Client administration consists of:
Users on a SCO Gateway for NetWare system can:
SCO Gateway for NetWare administration, as described in Chapter 3, ``Administering SCO Gateway for NetWare'' in the Guide to Gateways for LAN Servers, consists of:
For more information, see Chapter 1, ``Understanding the PC-Interface server'' in the PC-Interface Guide.
PC-Interface administration consists of:
IPX/SPX administration consists of:
Services that run over the Internet include e-mail, file transfer between systems (FTP), remote use of systems (telnet), and the World Wide Web.
Organizations can use Web technologies and the Internet to make key information available to a worldwide audience of millions of people. It is an effective and inexpensive way to distribute information such as newsletters, financial reports, marketing information, and other important documents. These can be available 24 hours a day, and can easily be kept up-to-date. Because customers or clients download this information directly onto their own computers, printing, mailing, and distribution costs are reduced while service and response time are improved.
The Internet can be used as a tool to sell and market products and services. Organizations can leverage the Internet to provide valuable marketing and sales information to their traditional customers while expanding their reach into new markets. Customers can purchase product 24 hours a day at their own convenience, and the automated tasks of ordering products electronically can streamline the sales process, resulting in a higher level of customer satisfaction.
Providing your organization with access to Internet resources
can enhance productivity. The Internet provides the largest
library of useful information in the world. Employees can use
the Internet to access a wealth of up-to-date information,
including newsletters, reports, and online databases.
For more information on setting up Internet access, see
Chapter 26, ``Configuring Internet services'' in the SCO OpenServer Handbook.
The World Wide Web
The World Wide Web is a global matrix (web) of
interconnected documents on the Internet. Clicking on a
cross-reference in one of these documents displays the
referenced document, even though it might be located on a
computer half-way around the world. This is possible because
all systems connected to the World Wide Web use a common
communication protocol, HTTP (Hypertext
Transfer Protocol), to send and receive documents,
and because all Web documents use the same method,
HTML (Hypertext Markup Language),
to define document formatting. Consult the
Netscape home page at http://www.netscape.com
for information on creating
web pages and HTML documents.
A collection of documents from one organization or individual on the Web is called a ``Web site'' (typically on a single system, although a large site can be supported by multiple systems or several sites could be located on one system). A Web site publishes these documents with software called a ``Web server'', and they are viewed with a ``Web browser''. Documents are typically divided into a number of short Web pages for convenient viewing. The top-level page at a site is referred to as a ``home page''. Web activity is not limited to viewing pages: data can be exchanged by the server and browser and manipulated by either or passed on to other software. This makes it possible to run programs, conduct business, and manage remote computer systems over the Web.
Although the Internet has existed for some time, the introduction
of the World Wide Web caused an explosive growth in its
use. Prior to the Web, navigating the Internet required typing
complicated paths and searching newsgroups for largely
unformatted text. Moving about on the Web is a simple matter
of clicking on interesting references. Searching the World Wide Web
for information is also simple, and the support for graphics, sound,
and motion makes the results of the search more rewarding. Encryption
and other security features have made it possible to buy and sell
services and products online. Increasingly, a presence on the
Web is as much a part of doing business as being connected to
the global telephone or e-mail networks.
Intranets
An ``intranet'' is an internal network that uses World Wide Web
technologies for sharing work and information between departments
and/or remote locations. Intranets can be used to facilitate teamwork
and project collaboration through e-mail and other network-based
applications. Company announcements, health insurance
information, corporate policies, procedure manuals and the like
can be made readily available from a central, easily updated Web site.
Consult the Netscape home page at http://www.netscape.com for more information.