It seems that just about everyone's "surfing the net" these days in one way or another. Just look at any trade press, and you'll see that the number of articles increases on a weekly basis. New magazines, like Web Techniques, are appearing on the market as wellmost of which help you learn new ways to get more out of your surfing hours or make surfing the net easier for others.
Tip: Web Techniques is a magazine devoted to helping you set up and administer your own Internet site. It includes information on finding CGI scripts (or creating your own) as well as Internet-based database management. Some of the technologies include Perl, HTML, Java, and VRML. They'll also tell you about Web security and how to create animated sequences. It's easy to get your own copy of Web Techniques. Just write or fax them:
Web Techniques
P.O. Box 58730
Boulder, CO 80322-8730
Fax: (303) 661-1885
The Internet isn't just a hot topic from the user perspective that we looked at in Chapter 19, "Surfing the Net." The Internet is composed of servers that companies can use for any number of purposes, including information exchange, sales and marketing, and private company networking. So what's this topic doing in a book about Windows NT workstation? After all, most people would say you'd need the server version of Windows NT to set up an Internet serverbut they'd be wrong.
Windows NT 4.0 workstation comes with a special version of Microsoft's Internet Information Server (IIS) software that allows you to create a limited but useful Internet server. In fact, you can use this particular version of IIS in a real-world environment in three ways. The following list tells you about three access methodsthe methods most appropriate to the capabilities of the server software. Remember that the Internet server software supplied with Windows NT workstation is only a peer-to-peer setup and won't handle the heavy-duty requirements of a full-fledged Internet site.
Obviously, I haven't covered every way you can use the Internet server capability provided with Windows NT workstation in this list; it only provides a breakdown of the three techniques for accessing the server. In the past few weeks, I've talked with people who had all kinds of uses for their Internet serversome of which had nothing to do with what most people term business. One person had a small setup that used virtual reality modeling language (VRML) to show off his architectural designs to a small group of customers. I also talked with someone who only needed a way to talk with his close friendshe set up a small, private Internet site to take care of the need.
In this chapter, we're going to look at some of the ins and outs of setting up an Internet server using Windows NT workstation. There isn't any way to cover every aspect of setting up a serversome authors claim that you couldn't even do that with an entire book. I do plan to provide some installation and setup techniques. After you have your server set up, we'll look at some of the things you can do to create a basic Web page or two. We'll also look at some of the exciting new technologies you'll be able to use to enhance that basic Web page.
The first thing we need to do is install the Internet server software. You won't find this item in the Add/Remove Programs dialog box we used to install other Windows NT applications. The Internet server software resides in the INETSRV folder below the main folder for your processor. If you're working with an Intel processor, for example, you'd look in the \I386\INETSRV folder of your installation CD.
Note: For the purposes of this chapter, I'm assuming that you have an Intel processor on your machine. The installation and setup procedures for the Internet server will vary slightly if you use a different processor. Make sure that you review each screen carefully as you follow this procedure when using other processors. Even though the differences are minor, you'll want to take advantage of any features these other processor families offer.
To install the Internet server, follow these steps:
Figure 20.2. The Installations Options dialog box allows you to select which Internet server features you want to use.
Feature | Type of Service |
Internet Service Manager | All |
World Wide Web Service | Small Internet server |
Remote intranet server | |
Gopher Service | Small Internet server |
FTP Service | Small Internet server |
ODBC Drivers & Administration | Remote intranet server |
Local intranet server | |
Help & Sample Files | All |
Figure 20.7. The Microsoft Peer Web Services Setup program displays a success dialog box when you complete the installation process.
Tip: You'll notice an Advanced button in the Install Drivers dialog box. In most cases, Windows NT takes the safe route when it installs drivers on your machine. It checks the version number of any existing driver, and then installs the driver only if it's newer than the existing one. In some cases, you may find that the drivers supplied by another vendor don't work properly with the Internet server. In that case, you can choose to force an installation of the drivers that come with Windows NT. You'll want to do this only when absolutely necessary, though, because overwriting a newer version of a driver file with an old one is never a good idea.
The installation program will automatically create entries on the Start menu for your new server. There are two utilities: one allows you to manage the server, and the other allows you to manage keys. We'll discuss these utilities in the next section. You'll also see an entry for the setup program that we've just talked about. Finally, if you decided to install the documentation, you'll see entries for one or more help files.
Windows NT provides a lot of capabilities as far as TCP/IP goes, and it's really not possible to cover them all in one short section of a chapter. The same flexibility that makes Windows NT so easy to use with TCP/IP, however, makes it nearly impossible for an inexperienced user to configure it. With this in mind, I decided to provide a very fast and easy method for you to configure TCP/IP for a local intranet.
The first thing you'll have to do is install TCP/IP. I show you how to do that in the "Adding New Services or Other Features" section of Chapter 21, "Peer-to-Peer Networking." You'll also find some subsections there that tell you all about some of the utilities provided with TCP/IP. You don't have to understand all of that additional information to use this sectionyour only concern is installing the required TCP/IP support.
After you have TCP/IP installed, you'll want to modify its properties. Just right-click the Network Neighborhood icon, choose Properties from the context menu, and then select the Protocols page of the Network Properties dialog box. You'll see a list of the protocols installed on your machine. Select TCP/IP Protocol and then click the Properties button. Windows NT displays a dialog box similar to the one in Figure 20.8.
Figure 20.8. You use the Microsoft TCP/IP Properties dialog box to set up TCP/IP on your machine for local intranet use.
Look at how I've configured the IP Address page in my dialog box. There are only two fields you need to worry about. The first is the IP Address field; the second is the Subnet Mask field. These two fields work together. An IP address is really composed of two parts. The first part defines your organization, and the second part defines your individual computer.
Notice that the IP address is actually composed of four numbers separated by periods. You can choose up to three of those numbers to represent your organization. The remaining numbers represent your computer. A single entry can contain any value from 0 to 255. All four entries combined make up the IP address. If you were on the Internet, you'd have to apply to InterNIC to get an IP address. On a local intranet, however, you're not really talking to anyone but your own organization. You can use any set of four numbers you want.
I normally select three easy-to-remember numbers for my organization and then use the remaining IP address entry to number the computers on my network sequentially. In this case, I've chosen 200.100.100 to represent my organization and 1 to represent the workstation.
Now comes the question of how the computer knows how to read this address. That's where the subnet mask shown in Figure 20.8 comes in. Like the IP address, it contains four numbers separated by periods. Unlike the IP address, however, you can only choose 0 or 255 for each number. A value of 255 represents an organization entry in the IP address, while a value of 0 represents a node (workstation) in the IP address. As you can see from the figure, I've placed a value of 255 for the first three numbers because I've used those numbers to represent my organization. I placed a 0 in the last position because it represents my workstation.
The final step is getting one computer to talk to another. Every computer on your network must use the same organization number or it won't be able to hear the other computers. In addition, every computer must have a unique node number. Because 1 is already in use, I'll use 2 for the next node. You must configure all the computers on your network before they can talk to the Internet server on your workstation.
Checking these connections is relatively easy as well. Windows NT provides a DOS utility called PING to check your capability to communicate with other TCP/IP workstations. All you need to do is type PING <workstation IP address>. If you want to check the next workstation in the network, for example, you'd type PING 200.100.100.2. PING transmits three packets and listens for a response. If you see three responses from a particular workstation, you'll know that you've configured it properly.
Simply installing the Microsoft Peer Web server software won't get you very far. You need to use two utility programs to configure your new server. The first is Microsoft Internet Service Manager. It allows you to view the current status of your Internet server and configure it as necessary. We'll cover these usage details in the next section. The Key Manager allows you to manage security using a series of keys and certificates. I'll get into what certificate security means in the "Using the Key Manager" section that follows.
You can use the Microsoft Internet Service Manager in a variety of ways. The simplest way is to use it as a simple monitor. It allows you to see the current status of various services offered by your Internet server. Figure 20.9 shows an example of how this utility performs this function.
Figure 20.9. You can monitor the status of various server services by using the Microsoft Internet Service Manager.
Before we go much further, it's important to know how to access the features of this program a bit better. Notice the toolbar at the top of this window. The following paragraphs describe each button:
Connect to a Server: You'll normally see the status of all your local serversat least those you choose to view. This option allows you to connect to another server on your LAN and to view the status of the services it offers as well. This feature allows you to control the services of more than one Internet server from a central location. Figure 20.10 shows the dialog box you'll see after clicking this button. Simply type the name of the server you want to connect to.
Figure 20.10. You use the Connect to Server dialog box to view the status of services offered by other servers.
Find Internet Servers: At times, you may not know which servers are up and running. Clicking this button finds any servers currently running on the LAN. It automatically displays the entire list of servers it finds alongside your local server. Figure 20.11 shows the dialog box you'll see as Microsoft Internet Service Manager searches.
Figure 20.11. Microsoft Internet Service Manager displays the Finding All Servers dialog box as it searches for other servers on the LAN.
Properties: Selecting any service and clicking this button displays a list of properties for that service. We'll talk about the specifics of using this button in a bit. For right now, it's just important to remember that you'll use this button to set the properties for a specific service.
Start Service: This button is enabled only when you select a service that is stopped or paused. It allows you to restart the service.
Stop Service: There aren't going to be many times when you'll need to stop a service, but there are a few. You might want to upload a new root page for your Web site, for example. The important thing to remember is that no one can use a service that is stopped. This means two things. First, you'll always want to make sure that no one is using the service when you stop it. Second, you'll want to make sure that you stop services for an extremely short time if site availability is important to your company.
Pause/Continue Service: You can pause a service to make quick changes. You might need to adjust the contents of a subordinate Web page or add a file to an FTP directory, for example. A pause affects the users of your Web site a lot less than stopping the service does. All the users see is an interval of inactivity; they won't get kicked off the service. Obviously, you'll still want to keep any pauses in service short.
View FTP Servers: Clicking this button displays all the FTP servers loaded on any local or remote Internet servers you're currently attached to.
View Gopher Servers: Clicking this button displays all the Gopher servers loaded on any local or remote Internet servers you're currently attached to.
View WWW Servers: Clicking this button displays all the World Wide Web (WWW) servers loaded on any local or remote Internet servers you're currently attached to.
Now that we've gotten some of the preliminaries out of the way, let's talk about server properties. I'm going to use the FTP Properties dialog box as my focal point for this discussion because it contains all the pages used by the other server types. In other words, every other type of server supported by Microsoft Peer Web Services uses some or all of the pages we'll talk about here. The following sections describe each page in detail. The first thing you'll need to do if you want to follow along is click on an FTP server and then click the Properties button in the Toolbar.
Tip: Most of the changes you make in the Service Properties dialog box won't appear to the user until the next time he logs into your site. In some cases, they won't even appear then unless the user clicks the Refresh (or equivalent) button from his browser. This problem also crops up when you change your Web page or the contents of an FTP site. Because many browsers store (cache) the contents of these pages, you might need to tell users to refresh their browsers in order to see changes.
The Service page is the first one we'll discuss, because it's the first one that appears when you open the Service Properties dialog box. You'll see a dialog box similar to the one shown in Figure 20.12. Notice that there are five pages in this dialog box; most of the other server types only contain four pages.
Figure 20.12. The Service page of the FTP Server Properties dialog box allows you to adjust server-specific properties.
The first two fields in this dialog boxConnection Timeout and Maximum Connectionsallow you to control how connections are made. Connection Timeout defines how long the server waits before disconnecting an inactive connection. The default setting is 600 seconds. The maximum you can specify is 32,767 seconds. In most cases, you'll find that the default setting of this field works just fine. Maximum Connections defines how many people can log into the server at once. The default setting is 1,000which is way too high for the capabilities provided by Microsoft Peer Web Services. You'll want to keep this setting near 10 in most situations. The maximum number of connections is 32,767.
Normally, you'll need to provide some type of method for people to log in. You can approach this in two ways. If you're providing a public service, there isn't any way to know in advance who will log into your server. Microsoft Peer Web Services (and all other Internet servers, for that matter) provide a special login name called Anonymous for these occasions. The middle portion of this page takes care of the anonymous user. The first checkbox, Allow Anonymous Connections, allows an anonymous user to log in. The next two fieldsUsername and Passwordspecify an anonymous user's name and password. You can also choose whether to allow only anonymous users or a combination of both anonymous users and specific users by enabling or disabling the Allow Only Anonymous Connections checkbox.
At the bottom of the Service field, you'll see a Comment field and a Current Sessions button. The Comment field allows you to define a comment that users will see when they log into your system. Clicking the Current Sessions button displays the FTP User Sessions dialog box shown in Figure 20.13.
Figure 20.13. The FTP User Sessions dialog box tells you which users are logged into the selected server.
You'll see three buttons along the right side of the FTP User Sessions dialog box. The Close button obviously closes the dialog box, and the Help button provides access to on-line help. The Refresh button is a handy one to keep in mind. This dialog box doesn't automatically refresh itself; it gives you a snapshot of the current status of the FTP site and that's it. To update your picture, you need to click the Refresh button.
Notice that the Connected Users dialog box provides four kinds of information. First, the icon tells you what kind of user is logged into the FTP site. In this case, I'm showing you an anonymous userthe most common type for an Internet site. Whether you use anonymous users in a intranet setup is up to your network administrator. Obviously, using names and passwords is a bit more secure than nothing at all.
The Messages page of the Service Properties dialog box allows you to personalize your Internet site a bit. Figure 20.14 shows what this page looks like.
Figure 20.14. You can use the Messages page of the Service Properties dialog box to personalize your FTP site.
You can display three messages. The first appears in the Welcome Message field; it welcomes a new user to your site. This is a good place to put any rules you want the user to see immediately. The second message appears in the Exit Message field; it's a lot shorter because all you'll probably say is goodbye. Likewise, the third message in the Maximum Connections Message field tells users that there aren't any more connections available. You might want to place a telephone number in this field so that users who experience trouble logging in will know who to contact for help.
You must specify a secure location on your hard drive for data that you want to share. The Directories page of the Service Properties dialog box shown in Figure 20.15 shows the default setup. Notice that the installation program automatically creates a default directory for you.
Figure 20.15. The Directories page is where you define the location of any data you want to share.
Three buttons appear right below the Directory listbox. The Remove button is pretty obvious. All you need to do is select a single entry in the listbox and then click this button to remove it. You'll want to use this button with carenever delete your root directory or people won't be able to access your site.
Clicking the Add or Edit Properties button displays the dialog box shown in Figure 20.16. The only real difference between the two buttons is that the Add button creates a new entry and the Edit Properties button uses the current entry. The first field contains a physical directory name on your machine. You can use the Browse button to find the directory using a File Open-style dialog box. Another alternative is to provide the universal naming convention (UNC) name for the physical directory you want to add.
Figure 20.16. You can view the properties of an existing directory or add a new one in the Directory Properties dialog box.
Note: The User Name and Password fields in this dialog box only become active if the Directory field contains a UNC path like \\AUX\DRIVE_C\USERS\DEFAULT. AUX, in this case, is the name of a computer on the network, DRIVE_C is the drive letter, and \USERS\DEFAULT is a path on that drive. Because a UNC path could exist anywhere on the network, you need to provide the name and password of the person who has access to that directory.
The next thing you need to decide is what kind of directory entry you'll create: home or virtual. There's only one home directory for any server; it's the first one the user sees when he logs into the site. You can have as many virtual directories as you want; they appear as subdirectories of the home directory to the user but can appear anywhere on your machine. In most cases, you'll want to select a virtual directory to retain the home directory created for you by the installation program.
If you do decide to create a virtual directory, you'll have to type a name for it in the Alias field. This is the directory name the site user will see. The physical directory might be G:\USERS\DEFAULT, for example, but the site user may see \PRIVATE instead.
Finally, you'll need to decide what type of access to give the site user: read or write. Assigning write access doesn't mean that the user automatically gets read access as well. This means you could set up a virtual directory for accepting uploads that no one else could see.
Tip: Even though Windows NT gives you the choice between MS-DOS and UNIX-style directory listings, most browsers will only see the UNIX-style listing. For maximum compatibility, you'll want to select the UNIX setting under Directory Listing Style. Not surprisingly, Internet Explorer falls into the UNIX-style only category. If you intend to use the browser that comes with Windows NT, select UNIX as the directory listing style.
Keeping track of who accesses your Internet server and when is pretty importanteven if you allow anonymous logins. Figure 20.17 shows the Logging page. The Enable Logging checkbox allows you to disable logging if you want to. Notice that there are two techniques for logging access to your Internet server.
Figure 20.17. The Logging page of the Service Properties dialog box allows you to track who has access to your Internet site.
I'll cover the easy way of tracking information first. This is the one the installation program sets up automatically for you. You enable it by selecting the Log to File radio button. All you need to do is look in the \SYSTEM32\LOGFILES folder for any logs. Enabling the Automatically Open New Log checkbox allows you to determine when the service will start a new log. Disabling this checkbox allows you to use a single log file for any length of time. The installation program automatically selects Daily for you. It also generates a log file name automaticallyINyymmdd.LOG, where yy is the two-digit year, mm is the current month, and dd is the current day.
There is another way to log access to your site. The only time you'll select this option is if you work for a large company and have a lot of sites to keep track of. All you need to do is create a data source using the 32-bit ODBC applet in the Control Panel. Because the Internet server only understands SQL databases, you'll need to select a data source that uses an appropriate database engine and data file type, such as Access.
After you create a data source and assign it a data source name (DSN), you can select the Log to SQL/ODBC Database option on the Logging page, as shown in Figure 20.18. I've provided some sample entries for you in this case. The ODBC Data Source Name (DSN) field contains the same DSN you created in the 32-bit ODBC applet. Each SQL database will contain one or more tables. You'll need to provide a specific table name to receive the access data. The last two fields contain the name and password of the person authorized to use the SQL database. I normally use the name of the person responsible for maintaining the Internet server.
Figure 20.18. You have to create a data source in the 32-bit ODBC applet in the Control Panel before you can use the SQL Logging option.
The final page of the Service Properties dialog box is the Advanced page shown in Figure 20.19. Microsoft should have probably called this the Security page, because that's what it helps you maintain. The first two radio buttons allow you to grant or deny access to all computers except those listed in the listbox. This is actually a lot more convenient than it actually looks. You could take a very open approach to site access and only restrict access to troublemakers. On the other hand, you could provide very tight security by denying access to all but a few trusted individuals.
Clicking the Add or Edit button displays the Deny Access On dialog box shown in Figure 20.20. (Microsoft changes the name of this dialog box to Grant Access On if you decide to deny access to all computers but those shown in the listbox.) You can choose whether you want to deny access to a single computer or to an entire group of computers. If you choose a single computer, all you need to do is type its IP address. Denying access to a group of computers requires the addition of a subnet mask entry. The third button associated with the listbox is Remove; you can use it to remove old entries.
Figure 20.19. The Advanced page of the Service Properties dialog box allows you to control access to your site.
Figure 20.20. You use the Deny Access On dialog box to define one or a group of computers that won't be able to access your site.
One other group appears on the Advance page. You might run into a situation where so many people are accessing your company's intranet that you have to limit the amount of network resources the intranet site is using. The computerese for this resource is bandwidth. Think of the cable running between computers on your network as a pipe. The pipe can hold only so much information. If you fill that pipe with intranet data, there won't be any room left for other kinds of data. That's why you would want to limit the amount of resources set aside for the intranet server.
To use this feature, just enable the Limit Network Use by All Internet Services on this Computer checkbox. The Internet server asks whether you're sure you want to take this action; choose Yes. Enabling this checkbox enables the Maximum Network Use field. Use the spin button or type the amount of network resources (in kilobytes per second) that you want to allocate. Because the capacities of networks vary greatly, you'll need to ask your network administrator about the value to enter in this field.
Security is an extremely important issue on the Internet. Read any trade press, and you'll find at least one story on security each week. As companies start making money transactions on the Internet, security becomes an even bigger issue.
One of the ways to improve security on your system is through the use of a key and certificate. The key opens the certificate. Think of a certificate as one of those locked boxes containing your data. You need the key to unlock the box and read the data inside. This is a bit of a simplification, but it does get the idea across.
Let's take a look at the Key Manager utility. When you first open it, you'll see a window like the one shown in Figure 20.21. Notice that there isn't much you can do right now. The number of things you can do will increase as I describe the various features of this program.
You can do a few things right now. The first two buttons allow you to connect and disconnect from other Internet servers. If you click the Connect button, you'll see a dialog box like the one shown in Figure 20.22. All you need to do is provide the UNC name for the server and click OK. The server then is added to the listbox on the left side of the window.
Figure 20.21. The Key Manager utility allows you to create keys to secure your Internet site.
Figure 20.22. The Connect to Server dialog box allows you to monitor key requests for other Internet servers on the network.
The three buttons on the left side of the Toolbar are also active. The first button allows you to send the keys for the current server to the printer. The second button provides general help for the Key Manager utility. The third button provides context-sensitive help. Just click the Help button and then point to whatever you want help with.
Three other buttons are shown on the Toolbar as well. They are enabled only when you select a key. They allow you to use the ClipBoard to cut, copy, and paste keys.
The following sections tell you how to perform the two basic tasks you'll perform in Key Manager: adding and certifying keys. If your company performs a lot of external transactions, you'll want to create keys for them. Your Internet service provider (ISP) will certify the key for you. This process creates a two-way secure link that you can depend on to get your data to you safely.
Looking Ahead: There are a lot of ways to implement security under Windows NT. We'll explore those issues in Chapter 23, "Security Issues."
You can begin adding a key in two ways. You can right-click on a server and choose Create Key Request from the context menu, or you can choose the Key | Create New Key command. Either method displays the Create New Key and Certificate Request dialog box shown in Figure 20.23.
Figure 20.23. Use this dialog box to create or edit a key request.
The first three fields are key specific. The Key Name field allows you to give the key a name that you can use to track it. The Password field is the actual key value; it's the value that will unlock the certificate later. Finally, the Bits field allows you to define the length of your key. A longer key requires more data processing time and increases the size of the data file. What you get in return is a file that is harder for a hacker to decrypt. The Key Manager always sets the longest key size possible as a defaultthe theory being that it's better to err on the side of security rather than data-transfer speed.
The Distinguishing Information group contains all the information that defines your company and your Internet site. You'll begin with all the usual information, including company name and organizational unit (your department, in most cases). The Common Name field contains your Internet address. Don't include the URL you use to access your sitejust the host and domain information. You might provide www.mycompany.com as a Command Name field entry, for example. The Country field contains the two-letter country abbreviation for your country; don't use an extended name here even though the field looks like it will hold it. You can use your state's full name in the State field. Likewise, the Locality field holds the full name for your city.
The final section of this dialog box holds the name of the request file you want to use. You can simply accept the default, or you can use the Browse button to search for an existing file. The Browse button opens a File Open-type dialog box.
After you supply all this information, you can click OK. Key Manager displays the Confirm Password dialog box shown in Figure 20.24. Type the password you selected and then click OK. Two things will happen at this point. First, you'll see a New Key Information dialog box, as shown in Figure 20.25. Second, you'll find a request file added to the directory you specified. Send this file off to your ISP. You'll probably get a certificate back within a few days.
Figure 20.24. Key Manager insists that you confirm your password before saving the key request file.
Figure 20.25. The New Key Information dialog box tells you that the key has been created but not completed.
Click OK to clear the New Key Information dialog box. You'll see a new key added to the server you selected, as shown in Figure 20.26. Notice that you can also see some information now on the right side of the window. The only field you can edit is the Name field. This key isn't complete, so you won't see much status information. You will see the information you provided in the Create New Key and Certificate Request dialog box, however (refer to Figure 20.23).
Figure 20.26. You'll immediately notice a new key added to your server after you complete the key request process.
Note: You should notice something about the new key in Figure 20.26. It has a red line through it. A red line always shows that the key doesn't have a certificate associated with it. We'll go through the certification process in the next section of this chapter.
Key certification is done by VeriSign, a division of Digital. You can contact them through their Web site at
http://www.verisign.com
The process for uploading your key request is pretty straightforwardVeriSign even provides a complete set of instructions for you to follow. You'll get the certificate for your key back in a few days (VeriSign says three to five days in most cases). Make sure that you have your credit card readycertificates aren't free and they do expire after a period of time.
Let's take a look at what you need to do after you get your certificate back. Right-click on the key that you want to certify and choose Install Certificate. As an alternative, you can choose the Key | Install Key Certificate command. You see a File Open dialog box. Use it to find and open your certificate file.
After you open the file, Key Manager asks you to enter your password. This is the same password you used to create the key request, so it's important that you record your key passwords and store them in a secure place, such as a safe.
After you install the certificate, the remainder of the status information for the key (at the right side of the window) is filled in. This includes a start and end date for the certificate. You'll also be able to assign the key to a server connection. Key Manager provides three choices: None, Default, or IP Address. The Default setting automatically assigns the key to the local server. The IP Address setting allows you to select another Internet server on the network.
If you have two workstations available on a network, try setting up the TCP/IP connections required to create an Internet server. Now, use the PING command to test out the connections. You can also use the other utilities I talk about in Chapter 21 to test out the connection.
Set up your own mini-intranet server. Try using various browsers with it to see which ones work best. Do you see any differences in the ways that browsers interact with your Intranet server? Think about how these differences would become even greater if you placed your server on the Internet.
Spend some time surfing the Internet to see how other Webmasters put their sites together. Write down these ideas and use them as a basis for improving your own site. Even if you only run a small company intranet, presentation is a big issue. It affects how often and how well people will use the site you set up.