AARNet:, ActiveX, ATM, bandwidth, bots, browser, cable Internet, cache:, CDF, CGI, Channel Definition Format, client:, cookies, DHCP, DHTML, digital signature, DNS, domain, Domain Name Server, download, DSL, Dynamic Host Configuration Protocol, dynamic Hypertext Markup Language, ecommerce, email:, encryption, extranet, File Transfer Protocol, finger, firewall, FTP, GIF, gopher, HTML, HTTP, Hypertext Markup Language, Hypertext Transfer Protocol, IETF, IMAP, Internet Engineering Task Force, Internet Relay Chat, intranet, IP address, IRC, ISDN, ISP, Java, JavaScript, JPEG, MUD, multicast, NNTP, packet, peer-to-peer, ping, plug-ins, point of presence, Point to Point Protocol, POP, port, PPP, protocol, proxy server, Push, Secure Sockets Layer, Serial Line Internet Protocol, server, SLIP, SMTP, spam, SSL, TCP/IP, Telnet, Telstra Internet, Transmission Control Protocol/Internet Protocol, upload, URL, Usenet, VBScript, W3C, Web, Web phones, Winsock, World Wide Web, XML

ActiveX: The original meaning of the term ActiveX has been largely lost in the shifting sands of Microsoft's marketing. It originally referred to a method of embedding "programs within programs" within Microsoft operating systems. If you had both Microsoft Word and Internet Explorer on your system, for , and you double-clicked on a Word document from within Explorer, then Word would actually open up within the Internet Explorer window, and the document could be accessed through it. So what's this got to do with the Internet? The Internet connection with ActiveX is a product of Microsoft marketing, which originally touted ActiveX as an alternative to Java: you could download small programs (called ActiveX controls) from the Web and run them within the context of your browser. These controls were expected to be used to provide the same features Java Applets provide on Web pages: pretty animations, unique menu systems and so on.

Today, ActiveX refers only to those controls, and not to the document embedding (the term COM+ is now used to refer to the document embedding features of Windows). ActiveX controls are not widely used today because of security issues (an ActiveX control has free reign over your system) and because they are not cross platform - although there are now implementations of ActiveX for several other operating systems, ActiveX controls need to be written for specific platforms.[Top]

ATM: asynchronous transfer mode. A monolithic networking protocol that provides numerous advantages over traditional networking protocols, such as TCP/IP and Ethernet. The most important things the average Web surfer needs to know about ATM are that is considered the wave of the future for networks (including the Internet) by many, and that its unique methods of transferring data make it exceptionally good at handling multimedia applications, such as real-time voice chats, videoconferencing and Internet conferencing. It can emulate other network protocols, including TCP/IP, to provide backward compatibility. It is not likely to appear on desktops for quite a few years yet, however.[Top]

bandwidth: The most talked about issue on the Net, bandwidth simply refers to the amount of data that can be transferred within a given period of time. It is usually referred to as a data rate (such as 33.6 kilobits per second), though it might also be referred to by frequency (measured in hertz, which means cycles per second; for , 100MHz refers to 100 million cycles per second.) Bandwidth is a key issue for the Internet because it is a measure of how responsive Web pages and other Internet services are -- the more of it the better. It's because of a lack of available bandwidth at the present time that the World Wide Web is so often called the World Wide Wait.[Top]

bots: Also known as agents, bots are computer programs that run automatically, are sent to do a task in lieu of a 'real person', and are capable of reacting autonomously to events. The name derives (as you've probably guessed) from robots, and they can take many forms on the Internet. Among other things, bots appear on IRC as fake personalities which give automated responses and comments, in MUDs and MOOs as characters who walk around and perform actions based on their programmed artificial intelligence, and in online games such as Quake as extra players. They are usually designed to appear as real people to the software and the world (to fool unsuspecting Net-goers and servers). [Top]

browser: The main tool for surfing the Web, the browser is both an HTML file viewer and HTTP client application. It is used to read Web pages and navigate hypertext links. Recent developments have resulted in moves towards monolithic applications which incorporate not only Web browsing but FTP transfers, Usenet reading, email reading and sending, and even HTML editing. The two most popular browsers, Netscape Navigator and Internet Explorer, are also extensible via third-party plug-ins, Java and ActiveX controls, giving them the capacity to be almost universal client applications. [Top]

cable Internet: The potential of the pay television cable networks to carry data has long been recognised, and last year Telstra launched Australia's first cable Internet service. The cable networks offer massive bandwidth (often measured in megabits per second), solving many of the problems of the World Wide Wait. High costs for the service and limited access options have so far stunted its growth somewhat, although Telstra and Optus still have hope for the service. Accessing the Internet via the cable networks requires a device called (funnily enough) a cable modem, and transmission can be asymmetric, meaning that downloads are faster than uploads (this is the case on Telstra's network). Cable Internet was originally devised by US cable companies that wanted a piece of the lucrative telecommunications market. Of course, in Australia, the cable companies and telecommunications carriers are one and the same.[Top]

cache: In general terms, a cache is a place where you store data to provide fast access later. When it comes to the Internet, there are several things to which it can refer. One is the cache on your personal computer -- with most browsers, when you access a Web site, most of the data from that site will be stored on your hard disk or in memory. If you go back to that site later on, it will load the site from the hard disk instead of the Web, which is, of course, much faster. ISPs can also cache Web sites and data on their servers to speed up download time and reduce traffic on the Internet. Downloading a Web site which has been cached on your ISP's server will be a lot faster because you don't have to draw the information across a heavily congested Internet.[Top]

CDF: Channel Definition Format. A specification from Microsoft which is intended to provide a standard way for push vendors to deliver push content to users on the Internet, as opposed to the proprietary applications now in use. CDF is incorporated into Microsoft Internet Explorer 4, but Netscape Communications has refused to include a receiver for CDF in its browser.[Top]

CGI: common gateway interface. A CGI program offers a standards-based way of receiving input from an end user and transferring it to a server. Many Web pages have forms embedded in them which use CGI to receive information from the user and transfer it to the server, which then processes that information.[Top]

Channel Definition Format: See CDF.[Top]

client: A computer that accesses a server is considered a client of that server. Client/server computing (as it is known) is the predominant network model of the Internet, although peer-to-peer networking also operates in some situations.[Top]

cookies: Derived from the Unix term 'magic cookies', cookies are small messages sent from a Web server to the computer, not to help the user, but for identifying the end computer and providing some configuration information to the server for future site accesses -- kind of like digital dog tags. Cookies are stored as harmless text files in a directory on the client's computer, and are automatically accessed the next time a user visits the site at which they received the cookie. Cookies can store a variety of information about users, but they were originally intended to enable customised views of Web sites which persist for future accesses to the same site. If you filled out a form with your name on it the first time you accessed a site, for , the second time you access that site, you might see a welcome sign with your name on it. When they were originally implemented in browsers, there were grave privacy concerns regarding the use of cookies, because they could theoretically be used to track an individual's site accesses and collect data on a user that could be used for nefarious commercial purposes. These concerns prompted browser programmers to give the option to reject cookies received from Web sites, although most browsers have the 'accept cookie' feature on by default, and users might have to go hunting through some 'advanced' configuration settings to turn it off.[Top]

DHCP: Dynamic Host Configuration Protocol. A set of standards for automatically assigning IP addresses to devices connected to a TCP/IP network. Essentially, instead of saying 'this is my IP address' (called static addressing), the device connects to a DHCP server, which assigns the device an IP address from a pool of available addresses. For as long as that session lasts, the device will use that IP address as its Internet address. When the session ends, such as when the user turns of their computer, the IP address will be returned to the available pool. DHCP is handy because end users and network administrators do not have to keep track of addresses; instead, software can take care of the automatic assignation and management of addresses.[Top]

DHTML: See HTML.[Top]

digital signature: A key authentication technique, used to prove that the sender of a particular message is exactly who they say they are. Email addresses are easily forged, for , but if you receive an email message which has been digitally signed it means that the originator is not a fake. (It does not mean that there is some kind of graphic at the bottom of the screen with the person's signature on it -- digital signatures are invisible to the end user.) To prevent digital signatures from getting into the wrong hands, encryption techniques are usually used to transfer messages with digital signatures.[Top]

DNS: domain name service. While TCP/IP understands only numeric addresses (see IP address), people don't work as well with numbers as with names. Because of this, the inventors of the Web devised the domain name system (see domain), which essentially has the job of linking names, such as www.apcmag.com, with IP addresses, such as 203.18.241.20 (try typing the IP address straight into your browser -- it will work exactly the same as typing the Web address does). Numerous servers around the world, called DNS servers, have the job of keeping this directory of domain names -- the phone books of the Web, if you like. When a user requests a Web site by name (instead of number), the computer will make a call to the DNS server, find out the IP address of the destination site, and then request the contents of the site by IP address. This is why you have to list your DNS servers when setting up TCP/IP on your PC, so your computer knows where to find the machine to translate the Web addresses into something the network the Internet is based on can understand. Without it, we'd have to remember an awful lot of numeric addresses.[Top]

domain: To 'own' a particular Web or email address (such as http://apcmag.com or joe@apcmag.com), companies and individuals have to register what is called a domain. Registered domains are then listed on DNS servers around the world, enabling people to access a particular site or address using the domain name rather than the IP address of the destination. Domain names are usually divided into two or three segments, separated by periods. The first of the three is the unique identifier for the company, group or individual, such as apcmag, microsoft or hotmail. The second term refers to the kind of organisation, most commonly com (for company), org (for organisation), net (for network), edu (for educational institution) or gov (for government). A revamp of this system is currently in the works, but for the moment these are the primary domain types. The third part of the domain name, if present, signifies the country the domain belongs to. Australia, for , has a '.au' extension. If there is no country given, then the domain is registered as a US domain, which does not necessarily mean that the server is in the US, but that the domain was registered in the US. Using this information, we can look at a domain name like ozemail.com.au, and see what the main organisation is (OzEmail), what type of organisation it is (a company) and the what the location is (Australia).[Top]

With Web and other forms of Internet servers, the owner of a domain is entitled to put any number of words before the domain name in a URL. OzEmail, for , has the right to have Internet sites such as help.ozemail.com.au, downloads.ozemail.com.au and jan.was.here.ozemail.com.au, if desired. The two most commonly used prefixes are www. (as in www.ozemail.com.au), a convention signifying that the site is a World Wide Web site, and ftp. (as in ftp.ozemail.com.au), indicating an FTP site. Email addresses work similarly, and any name can be used before the @ symbol. To use the OzEmail example again, OzEmail has the rights to email addresses such as mike@ozemail.com.au or even scary.spice@jan.was.here.ozemail.com.au. See also DNS.[Top]

Domain Name Server: See DNS.[Top]

download: To retrieve data (of any type) from the Internet for use on a PC.[Top]

DSL: Digital Subscriber Line. Coming in a variety of formats, including Asymmetric (ADSL), Symmetric (SDSL), High-data-rate (HDSL) and Single-line (SLDSL or SDSL), DSL is another high-bandwidth digital communications standard. Developed by US telcos to oppose cable-based data communications (see cable Internet), DSL offers high-speed digital communications over the phone network. It's much talked about in the US, but in Australia, where the phone companies own the cable networks, it is not such a big issue, and Telstra has yet to offer the service. Like Cable, ADSL requires a special modem to access the network. See also ISDN.[Top]

Dynamic Host Configuration Protocol: See DHCP.[Top]

dynamic Hypertext Markup Language: See HTML.[Top]

ecommerce: electronic commerce. Refers to the process of conducting business online, such as buying and selling goods and services over the Internet, funds transferal within online banks and so on. Security concerns have limited the use of ecommerce in the past, but these issues are slowly being overcome. See encryption.[Top]

email: electronic mail. A service available on the Internet and other networks which allows people to send 'letters' to each other via their computers. These letters are composed in an email application and sent to one or more recipients. The addresses of recipients are usually denoted by name@domain (in some internal networks, the notations might be different, but for Internet mail, this notation is always used). Some email systems also allow you to define groups of people for easy referencing, such as a group called #management, which might send a separate email to every manager in an organisation without requiring their names to be typed in individually. Email can also be CCed (carbon copied) and BCCed (blind carbon copied). This means that the email is also sent to the people in the CC or BCC list, but usually means that the email is not directly intended for them. In most cases, the recipients of a particular email know who else received the email, because that information is listed in the email header. This is where BCCing comes in -- anybody on the BCC list also receives the email, but nobody but the sender knows that they received it because BCC recipients are not listed on the header.

Email is usually not beamed directly to desktops, but stored on mail servers (Internet servers which store mail messages) and are retrieved by clients on request. See also IMAP, POP, SMTP.[Top]

encryption: The use of complex mathematical formulas to make data unreadable without the use of a decoding key. Data sent across the Internet is not secure -- it is not terribly difficult for streams of data to and from your computer to be intercepted and read. Encryption methods overcome this by scrambling data so that, even if it is intercepted, it cannot be read. (Presumably, the intended receiver has the key to decode it.)[Top]

There are two primary encryption methods: symmetric and cryptographic. Symmetric encryption requires the same key to decode the data as to encode it. It is not widely used on the Internet, because it usually requires physical transfer of the key (if you sent the key over the Internet, it could be intercepted and then used). Cryptographic methods, also known as public-key encryption, have different keys for encrypting data (the public key) and decrypting it (the private key). If you want to send an encrypted message to someone, you request their public key over the Internet, which you then use to encrypt the message before sending. Only that person's private key can then be used to decrypt the message, so even if the public key is intercepted by unscrupulous individuals on the Net, the data still cannot be read. Private keys cannot be fabricated by looking at public keys. See also SSL.[Top]

Many countries, including Australia and the US, have strict rules on the export and import of encryption software, and some require that 'universal' keys be provided to government agencies to enable the decryption of all data encrypted using that software. These countries generally use the excuse of 'national interest' when called upon to justify these restrictions.[Top]

extranet: A relatively new term, used to describe an intranet that has been made partially accessible to outsiders. The idea behind the extranet is the integration of the public and private information of the company. Access to this information pool is controlled to keep sensitive information out of the wrong hands. Many companies, for , are linking information on their Web sites to what were once private internal databases, but are now used (with appropriate access restrictions) to provide information to the public about the company and its products. [Top]

File Transfer Protocol: See FTP.

finger: Originally developed for use in certain Unix email systems, finger software was initially used to simply find out if another person was logged on to the system. The email address of the sought person would be entered into the finger software, which would provide information on whether that person was logged on or not. Finger software has developed so that the person being fingered can enter a lot more information about themselves if they wish, which will be provided to the person using the finger software. People can only be fingered if they are using an appropriate system and allow themselves to be.[Top]

firewall: A security method used by Internet-connected companies to prevent unwanted intrusion on their internal networks. The firewall essentially prevents people from accessing the company's network from without, except for the limited services that the company wants to be available to the public. There are a number of firewall techniques, and they can be implemented in both hardware and software.[Top]

FTP: File Transfer Protocol. A standard way of transmitting files over the Internet. It can be used to both download and upload files. Most original FTP programs were text-based, and required a little knowledge of Unix to use; however, now window-based FTP programs are available. HTTP can also be used to transfer files, but is considered less efficient at doing so than FTP. FTP servers usually operate on port 21 (see port), and many have a limit on the number of users at any given time.[Top]

GIF: Graphics Interchange Format. An image encoding and compression method first introduced by CompuServe. GIF, like JPEG, reduces image files to a fraction of their original size, enabling faster transfer over the Web than standard (bitmapped) images. GIF encoding provides better compression than JPEG on images with few colours, but lacks the colour depth, variable compression ratios and high compression rates on images with a lot of colours that JPEG possesses. Most Web browsers are equipped to decode and display GIF files because they are commonly used in Web pages.[Top]

gopher: The predecessor of the modern Web page, gopher was used to organise and display files on gopher servers. It has largely fallen into disuse, and now gopher sites are few and far between.[Top]

HTML: Hypertext Markup Language. HTML is essentially the file format used for Web pages, similar to the way DOC is the file format used for Microsoft Word documents. When viewed, an HTML file can incorporate both text and a variety of objects, such as images. Its most important feature is its linking capacity, which enables words and images to be linked with other documents or other locations within the same document, such that when these links are clicked on, the browser (the most common tool for viewing HTML) will proceed to that address. Because of the wide variety of Web clients out there, HTML was designed so that it can be viewed on a variety of screens running at a variety of resolutions. This places some limitations on the language, but also ensures that it is cross-platform.[Top]

HTML can be written by hand in a standard text editor; however, this requires knowledge of the commands and symbols required to format, embed and link objects on the page. More recently, graphical editors have made it nearly as easy to make a Web page as it is to make a document in a word processor.[Top]

HTML has gone through a number of versions since its first inception, and is currently up to Version 4, dynamic HTML (DHTML), which enables some limited interactivity to be incorporated into the HTML document. You might also see 'server-side include' HTML documents (SHTML) and Active Server Pages (ASPs) floating around occasionally -- these are documents that have been dynamically generated by the Web server in response to a request from the end user. [Top]

HTTP: Hypertext Transfer Protocol. A protocol used in the World Wide Web which defines how clients and servers should react in response to certain commands, and how messages on the Web should be formatted. When you enter a Web address into a browser for , you are actually using HTTP to signal the Web server that you want a specific page. You have probably noticed that most Web addresses start with http:// (although clever browsers will allow you to leave out this part of an address now, and they will fill it in), which simply says that HTTP is the protocol

[Top]

HTTP is currently in Version 1.1, although most Web sites still use Version 1.0. The difference is that in Version 1.0 downloading each element of a Web page involved a separate HTTP connection to the server, which took up a lot of time. HTTP 1.1 allows persistent connections to servers, saving the client from having to renegotiate its connection with the server for each page element. The Internet Engineering Taskforce (IETF) is currently looking at new extensions to the protocol, called Secure HTTP (S-HTTP), which are designed to enable secure data transactions across the Web. See also SSL.[Top]

Hypertext Markup Language: See HTML.[Top]

Hypertext Transfer Protocol: See HTTP.[Top]

IETF: Internet Engineering Task Force. A loose confederation of engineers and other interested parties, the IETF is the main body responsible for setting standards for the Internet. It's divided into a number of working groups which look after different aspects of the Internet.[Top]

IMAP: Internet Message Access Protocol. Similar to POP, IMAP is a standard for retrieving email messages from mail servers. An IMAP compliant email reader, for , should be able to retrieve mail from an IMAP mail server. Like POP, IMAP uses SMTP to transfer data between mail servers and clients. IMAP is currently in Version 4.[Top]

Internet Engineering Task Force: See IETF.[Top]

Internet Relay Chat: See IRC.[Top]

intranet: A private network which uses Internet standards for communicating internally, rather like a mini-Internet. With the rise of the Internet, organisations began to realise that Web pages, SMTP email, FTP and so on were not only useful for the Internet, but also for communicating within their internal networks. In addition, the software often proved to be much cheaper, saving organisations a lot of money on proprietary systems. Intranets are always based on the TCP/IP protocol, and often feature connections to the Internet which can be accessed by the same client software as used for the intranet.[Top]

IP address: Internet Protocol address. The unique identifier of a computer or device on the Internet (or other TCP/IP network). This enables devices to find each other on the Internet and communicate. When you dial into an ISP, you are assigned an IP address which will remain yours for the duration of the session. This IP address usually has to be returned to the ISP at the end of the session. Devices which are permanently connected to the Net generally have a fixed IP address.[Top]

Most IP addresses comprise four numbers between 0 and 255, separated by a dot (for , 203.18.241.19). IP Version 6, which is not widely used on the Internet at the moment, has a much greater number possible addresses. See also TCP/IP, DNS, DHCP.[Top]

IRC: Internet Relay Chat. A popular tool for participating in real-time text chats over the Internet. It enables multiple users to join in a discussion at the same time. Joining an IRC 'channel' requires a special IRC client, which connects to an IRC server.[Top]

ISDN: Integrated Services Digital Network. A means of digital communications between two points (such as a business and an ISP), now integrated into the phone network. Using ISDN, calls are made to a specific location, just like phone calls, except the connection is digital instead of analog. This enables much cleaner voice, videoconferencing or data transfer. The bandwidth provided by a single ISDN connection varies around the world, but in Australia it is 64Kbps, and it is usual for two lines to be sold together as a package (for a combined bandwidth of 128Kbps). Because it offers better data rates and cleaner lines than analog modems, ISDN is often used as a means of getting high-bandwidth connections to the Internet.[Top]

Call rates for ISDN are generally much higher than normal phone calls in Australia. Making data calls with ISDN generally requires an ISDN modem.

ISP: Internet service provider. A business which offers connections to the Net, often in addition to other Internet services, such as Web-site hosting and consulting. An ISP will have some kind of permanent connection to the Internet, which it resells to home users and businesses. Cost generally reflects the quality of the ISP's connection to the Net, as well as the bandwidth demanded by the user.[Top]

Java: A programming language that can be used to produce programs that are (theoretically) cross-platform. A program written in Java can be converted (compiled) into a format that is not specific to any type of computer or processor. This format is called byte code, and it is designed so that it can be converted to machine-specific code as it is run. This is done through what is called a Java Virtual Machine (JVM), a program which essentially has the task of converting the byte code into something the computer it is running on can understand. Theoretically, any type of computer that has a JVM written for it should be able to run Java programs. JVMs are typically distributed with browsers and are invisible to the end user.[Top]

Java is primarily used on the Web to write small programs that liven up Web pages and that can be run on almost any type of client, unlike ActiveX controls, for , which only run on the platform they were written for. Java is also being used on servers, and more recently in the development of office suites that are designed to run on a number of different client machines, without there having to be different versions of the software.[Top]

The greatest criticisms levelled at Java concern its speed and the fact that it isn't quite as cross-platform as it claims to be. The former problem arises from the fact that it is translated as it goes, and the latter is a result of different vendors' JVMs not being as interoperable as they should be.[Top]

JavaScript: A scripting language designed to take care of simple tasks on Web pages. Originally developed by Netscape and allowed by Sun to carry the Java brand, JavaScripts can be embedded in HTML files to provide a simple way of performing minor tasks on Web pages; for , animating images, scheduling advertisements to appear and disappear, or sending messages to the server. It enables these minor tasks to be accomplished without invoking the JVM of the particular machine you are using. The syntax of the JavaScript language is similar to that of Java (hence the name). See also VBScript.[Top]

JPEG: Joint Photographics Experts Group. Can actually refer to this group, but more commonly refers to the image file format and compression scheme it developed. JPEG images can be compressed to a fraction of their normal size, although the greater the compression ratio, the worse the quality of the picture. GIF and JPEG are the most commonly used image formats on the Web.[Top]

MUD: Multi-User Dungeon or Multi-User Dimension. A text-based virtual environment in which a number of people from the Net can interact. The first uses of MUDs were in text-based adventure games, in which players took on characters and wandered around virtual environments, such as dungeons and castles. Commands like Go North, Eat Pie and Kill Ogre controlled the actions of the characters (anybody familiar with the early Infocom adventures such as Zork would be accustomed to the syntax). A large number of players can be wandering around the environment at the same time, and players can usually interact with each other, exchange items, form alliances and so forth. Connecting to MUDs usually requires the use of Telnet. [Top]

multicast: A system of broadcasting over TCP/IP networks, such as the Internet, which makes the best use of available bandwidth. It is used primarily for streaming applications, such as broadcasting video or audio over the Internet. Instead of sending out a separate stream to each client, or sending a stream to every client whether they want it or not, Multicast only replicates the stream at the last possible moment. The server only sends out one stream, which branches as required to reach the clients which have reported an interest in that stream. Anybody who has not reported an interest is effectively 'pruned' from the broadcast, saving bandwidth.[Top]

Multicasting is seen by many as an important technology for the future of the Web. When the bandwidth available on the Internet is enough to handle TV-quality video and CD-quality audio, multicasting is seen as the means by which programs can be broadcast internationally. It is already in use on the MBone (Multicast backbone), a subset of the Internet designed to handle multicasts. Current implementations of Web-based video and audio streaming, such as RealNetworks' RealAudio and RealVideo, do not use multicast methods, but still use unicast (sending a separate stream to each client who wants the data). This is because much of the equipment and software that drives the Internet is not equipped to handle multicasting.[Top]

NNTP: Network News Transport Protocol. The protocol developed to provide a standard way of posting, reading and receiving Usenet messages. Your Usenet news reader will use NNTP to access newsgroups. Like many of the protocols listed here, NNTP is an open standard, allowing it to be used across a number of platforms and products. [Top]

packet: Unlike the phone network, which is circuit-switched, the Internet uses a method of data transfer called packet switching. Devices don't call each other to set up a line of communication; rather, they put chunks of data on the network, with a small address tag telling the network where it wants the data to go. These chunks of data are called packets, and the Internet takes care of delivering them to the appropriate computer.[Top]

Packets will generally contain more than just addresses and data. They will also contain special delivery details, return addresses (so the computer receiving the data knows who to reply to), version information, and a variety of switches telling the network just how the data should be transferred and how urgent the message is. The size of a packet can vary.[Top]

Packets are used because they make the most efficient use of the network. Circuit switching wastes bandwidth when connections are idle, and the last thing we need on the Internet is wasted bandwidth. Because your packets have to compete with everyone else's on the Internet, sometimes they can take a long time to arrive, and sometimes they don't arrive at all. The latter is called packet loss -- it's not fatal, it just means that if your computer is forced to resend the packet if it doesn't receive a response within a given period of time. The receiver of a TCP/IP packet always lets the sender know when they've received a packet.[Top]

peer-to-peer: The alternative to the client/server model, peer-to-peer refers to a network in which devices talk directly to each other, without an intermediary server. An example is a messaging application which sends the message directly to the destination computer, instead of storing it on a server and waiting for the person at the other end to download it from the server.[Top]

ping: A method used to test if there is a line of communication to another on the Internet, and how fast it is. The pinging party simply sends a packet of data to the destination computer, and waits for a reply, timing how long the reply takes. A long wait means a slow connection; no reply means no communications.[Top]

plug-ins: Extensions to Web browsers which enable a larger number of file types to be read. For , the Adobe Acrobat plug-in allows users to read any PDF files they have downloaded from the Net within their Web browser, and the Shockwave plug-in allows them to view Shockwave files embedded in Web pages.[Top]

point of presence: See POP.[Top]

Point to Point Protocol: See PPP.[Top]

POP: point of presence (1) or post office protocol (2).

(1) A location at which an ISP can be dialled at the cost of a local phone call. You need your ISP to have a local POP if you are to avoid long-distance call charges to access that ISP. Some ISPs might have virtual POPs, which means that they don't actually have a physical office in the location specified, but people can still access the service for the cost of a local phone call.

(2) A standard way of signalling one's desire to send or receive email from an email server. Similar to the newer IMAP, POP2 works in conjunction with SMTP (which takes care of the actual transferral of the mail), although the most recent version of POP, POP3, does not require the use of SMTP to transfer mail.[Top]

port: Primarily used to access different services available on a single server. For , a single server might only have a single IP address but both FTP and HTTP services running. So it uses different ports (the default is 80 for HTTP and 21 for FTP) for the two services. Ports are denoted by number following an IP address and a colon; for 203.18.241.20:80. All TCP/IP requests sent to a server require a port number, but when it comes to services such as HTTP and FTP, which have standard port numbers, the browser usually hides these details from the end user.[Top]

PPP: Point-to-Point Protocol. Essentially a standard way of sending TCP/IP packets over the modem connection that connects a computer to an ISP. The modem does not normally 'talk' in TCP/IP, and so it needs PPP to translate between the two languages.[Top]

protocol: A standard way of doing something. Because the Internet is comprised of so many different kinds of machines, Internet protocols were devised to give all these machines a common way of talking to each other. [Top]

proxy server: A server that sits between the client and the destination server which either filters access or tries to fulfil requests itself. The former is a security measure, the latter is generally done in order to improve speed, by caching, for . [Top]

ISPs often ask (or force) people to go through their proxy servers in order to save bandwidth. It might, for , have a certain Web page cached on the ISP's Web proxy server because previous users had recently downloaded the page. If another client requests the page, then instead of sending to the other side of the world and wasting a lot of time and bandwidth retrieving it again, it might simply send them its cached copy of the page.[Top]

Push: Touted as being a technology to revolutionise the way the Internet works, push technology works by sending data to clients over the Internet without waiting for them to request information. This makes push analogous to traditional broadcast media, and a distinct move away from the way people traditionally dealt with the Web and the Internet. It was thought (and still is) by many that this is the way people prefer their information delivered, rather than having to hunt it down themselves.[Top]

Many push applications enable clients to set up profiles detailing the kind of information they would like delivered to their desktops and how often they would like it delivered. Push generally requires a lot of bandwidth, and often works best for clients with a permanent connection to the Internet. The best known push technologies include PointCast and Marimba's Castanet.[Top]

Secure Sockets Layer: See SSL.[Top]

Serial Line Internet Protocol: See SLIP.[Top]

server: A device on a network that delivers data or configuration information to clients, usually upon request. On the Internet there are all sorts of servers in use. A Web server, for , delivers Web pages on request, while an FTP server delivers files over FTP, and an IRC server delivers and receives IRC messages. Servers are usually high-powered computers and often have a dedicated task, although some may have several services operating at once -- for , a single machine might be operating as both a Web server and a proxy server.[Top]

SLIP: Serial Line Internet Protocol. The predecessor to PPP, SLIP is not widely used today. It refers to a standard method of using a modem to connect to the Internet via an ISP.[Top]

SMTP: Simple Mail Transfer Protocol. A standard for sending email messages between devices on the Internet. This refers to both the way servers send mail messages to other servers, as well as the way email clients send mail to servers. Clients generally use IMAP or POP to retrieve email from servers.[Top]

spam: The process of sending unwanted email to a large number of receivers. Also known a electronic junk mail, spam is one of the great evils of the Internet. Email has no limit on the number of receivers of a given mail, and it doesn't cost any more if you send it to lots of people. Unscrupulous characters and businesses with access to email have used this to send emails with advertisements and other materials to massive lists of people who don't want it, but have to download it when they download their mail. Obviously, this wastes everybody's time and money, along with lots of precious Internet bandwidth. As a result, a number of groups have been formed to try to fight habitual spammers, and a number of software packages are available which have the task of trying to filter spam from users' email inboxes.[Top]

SSL: Secure Sockets Layer. A form of public-key encryption (see encryption) devised by Netscape to provide secure electronic transactions over the Net. It's supported in both Netscape Navigator and Internet Explorer. In both, transitions from standard HTTP to SSL can be made transparently to the end user, although by convention Web pages supporting SSL begin with https: instead of http:. See also HTTP.[Top]

TCP/IP: Transmission Control Protocol/Internet Protocol. A set of communications protocols for both private networks and the Internet. The most important protocol on the Internet, TCP/IP defines the way in which computers on the Internet signal each other and send data. Possibly its most important feature is its addressing feature, which allows so many devices to be connected to the one network yet find each other (see IP address). TCP/IP is a packet-based protocol.[Top]

TCP/IP is a default networking protocol on Unix computers, and the software to be able to connect to a TCP/IP network comes with most newer operating systems. Older operating systems (such as Windows 3.1) may need extra software to be able to connect to the Internet. This software is sometimes called a TCP/IP stack.[Top]

Telnet: A text-based terminal emulation program, designed to run over TCP/IP networks such as the Internet. Essentially, it allows users to connect to a server and issue commands as if it were a console on their computer. It is used for a number of purposes on the Internet, from controlling Web servers remotely to access library databases (some libraries in Australia allow you to access their systems just as though you were at one of the in-library search terminals). It is also the primary means for accessing MUDs over the Internet.[Top]

Telstra Internet: Previously known as AARNet (Australian Academic and Research Network) but transferred to Telstra's control in 1995/96 and subsequently renamed Telstra Internet, this network forms the core of Australia's Internet access. This Australia-wide TCP/IP network links the major regional centres as well as several international networks, and forms the largest part of the Internet in Australia.

Most ISPs buy bandwidth on this network, which they then resell to their customers. Some of the larger ISPs have their own nationwide networks as well as international links, although almost all have some kind of link to Telstra Internet because of its importance in connecting sites within Australia.[Top]

Transmission Control Protocol/Internet Protocol: See TCP/IP

upload: To send data from your PC to a server on which it will reside. You could use FTP, for , to upload a file from your local machine to a directory on an FTP server.[Top]

URL: Uniform Resource Locator. The global addresses of devices and data on the Internet. It could refer to a Web site, an FTP server, an email address or even a gopher site. For , http://apcmag.com/newswire is a URL, as is ftp://ftp.cdrom.com.[Top]

The first part of the address refers to the type of service; http://, for , refers to a Web (HTTP) server, while ftp:// refers to an FTP site. The second part of the address has the name (or IP address) and the location of the data desired.[Top]

Usenet: A worldwide bulletin board system, Usenet was one of the first applications of the Internet. There are several thousand such bulletin boards (called newsgroups) on the Internet, each dedicated to a specific topic, such as Windows 95, fast cars or nappies. Posts to a bulletin board can be made by anybody using a news reader, a software application which allows previous posts to the newsgroup to be read and new posts to be made (news readers are integrated into both Netscape Navigator and Internet Explorer). [Top]

Newsgroups are very useful for finding out information on a specific topic, although wading through the masses of posts on some newsgroups can be tiresome. Posts will often receive a response, or a number of responses (which in turn might generate their own responses, forming what is called a thread), thus making newsgroups good places to get in contact with people who know a great deal about the topic you're researching. Like all things on the Internet, however, responses on newsgroups may not always be accurate.[Top]

VBScript: Visual Basic Script. Similar in function to JavaScript, but only implemented in Microsoft's Internet Explorer browser, VBScript enables small programs to be embedded in HTML files which take care of executing limited tasks. The scripting language bears similarities to Microsoft's Visual Basic. See JavaScript.[Top]

W3C: World Wide Web Consortium. Founded in 1994, the W3C is the standards-setting body for the Web, controlling the definition of both HTTP and HTML. It has a large number of international members.[Top]

Web: The part of the Internet with which most people are familiar. The Web comprises those portions of the Internet that use HTTP and HTML to deliver Web pages to end users. Web pages are basically files that can have text, images or applications embedded in them. They can be built to react to user input through CGI, DHTML, Java or ActiveX, and objects can be hypertext linked to other Web pages on the Internet. Using hypertext linking, these Web pages are linked to others around the world, forming a complex network of linked pages stored on a number of Web servers.[Top]

It is a common misconception that the Web and the Internet are the same thing. They're not. Technically, the Web only refers to the network of hyperlinked HTML pages on servers across the Internet. The Internet offers much more than this, including things like Usenet, email, FTP, live chat, online games, and pretty much anything else that can be achieved by computers linked together (the Internet really is just a big network, after all).[Top]

Web phones: Software which sends sound over the Internet in real time, allowing voice conversations to occur through an Internet connection. Essentially seen as a way of using the global Internet as a way to bypass expensive international phone charges, Web phones set up a peer-to-peer connection between two computers running the same software. That software gets input from the user via the PC's soundcard, digitises it, and sends it over the Web to the other machine. At the other end, the signals are converted back to sound, and the person can respond. This way, people can have real-time voice conversations over the Internet, without it costing any more than the cost of their Internet connection. Unfortunately, the Internet is still far too slow to allow decent voice conversations, and most Web phone conversations are plagued by lags and poor audio quality.[Top]

Winsock: Windows socket. The Microsoft Windows implementation of a TCP/IP stack (see TCP/IP).[Top]

World Wide Web: See Web.[Top]

XML: Extensible Markup Language. An enhanced version of HTML, recently ratified by the W3C. It provides greater flexibility when designing Web pages. It's not yet supported in any of the major browsers, although Microsoft intends to support it in its next generation of Web browser.[Top]