|Index to Section 1|
Introduction to Network Resources
World-wide network of computers (over 10,000,000 at the last count!) in academic, government & commercial institutions interconnected by land lines, microwave & satellite links:
Many of these computers simply act as storage and archival sites for vast quantities of publicly accessible information on every imaginable topic.
Information is multimedia: text, software, graphics, video, audio.
"Virtual tourism" - Take a free trip round the world, but never leave your desk!
"Cyberspace" is the "virtual world" which contains this abstract collection of information; a "Cybernaut" is a virtual traveller in this virtual world.
"Globewide Network Academy" - University degree course units by distance learning.
The trick is to know how to find out what is available, to locate the information you want, and finally to fetch it from the source to your desktop.
Search tools available for PC's (DOS, Windows) and workstations (Unix, VMS).
Sources of interesting information (to molecular modellers!).
Local networks developed within Universities & Government research institutions.
Then local networks were connected together into country-wide networks, to allow exchange of scientific data (e.g. astronomical images), and to share computing power.
In UK we had JANET (Joint Academic Network), in Europe EARN (European Academic Research Network), in USA ARPAnet (Advanced Research Projects Agency network).
Then "gateways" were provided at National Computer Centres (e.g. ULCC, University of London Computer Centre) to interconnect the country and state networks by high-speed satellite links.
Most commercial companies joined later, because of concerns about security - information flow is 2-way!
"Hacking" (unauthorised access to computers on the network that are not designated as publicly accessible) is a big problem now; no-one worried about security in the early days.
No single organisation controls Internet, it is just the sum of its component parts, each part being controlled locally.
Networking "protocols" (rules) had to developed early on, so that everyone agreed on how all the computers should communicate with each other.
Usage expanded from purely scientific data to all kinds of information - e.g. many libraries and several art galleries and museums are now on the Net.
There are many more network users than there are lines available.
Transfer of information between computers is often faster than the computer is able to store it, so transfer is held up, and expensive line time is wasted.
Also if data is sent in one chunk and a transmission error occurs, whole of data has to be re-transmitted - inefficient.
To allow many users to share the same lines, all network traffic must be broken up by the sending computer into small "packets". Also the packets may be further chopped up en route.
Each packet carries its destination "address" and a sequence number, so that packets can be re-constituted in the right order. It is the responsibility of the receiving end to put all the pieces back together again.
In fact each packet may take a completely different route! This is necessary because in real life computers fail and communication links get broken. (The system was actually designed, during the Cold War period, to survive a thermonuclear attack on the USA!).
"Packet switching" - the computers on the route ("routers") automatically switch the packets on to the most cost-effective path. This is dynamic, because it depends on the volume of all the other packets being handled at the time.
If a transmission error occurs in one packet, only that packet needs to be re-sent, minimising waste of line time.
All this is defined by several protocols, the main ones are TCP ("Transmission Control Protocol") and IP ("Internet Protocol"), so the networking system is known as TCP/IP
The network is organised by "domains". There are two types of address on the Internet:
DNS ("Domain Name Server") name, or FQDN ("Fully Qualified Domain Name")
amber.cryst.bbk.ac.uk ............................................(any number of words separated by .'s)
"amber" = my workstation, "cryst" = Crystallography Dept.,
"bbk" = Birkbeck College, "ac" = academic institution, "uk" = UK.
Other examples use
".com.uk" or ".co.uk" for UK commercial organisation,
".de" for Germany,
".ch" for Switzerland,
".za" for South Africa,
".au" for Australia etc.
Note that most US sites do NOT use ".us". Instead ".edu" = educational establishment, ".gov" = government institution, ".org" = non-profit organisation, ".net" = network organisation, ".com" = commercial, ".mil" = military.
126.96.36.199 (4 numbers between 1 & 255 separated by .'s. This is actually a 32 bit number.)
Names and numbers are completely equivalent. A dedicated computer nearby on the Net (the "Nameserver") automatically converts names to IP numbers. You should always try to use the DNS name, because it's easy to make mistakes remembering and typing numbers.
The name or number is unique in the world. Usually the first 3 numbers indicate a department or institution, and the 4th the individual machine or node. See example "hosts" file.
If the computer is new, the DNS name may not have reached the nameserver yet, and then you will have to use the IP number.
You can find out IP numbers with the "nslookup" program (only available on certain machines). This searches a wider domain for the name.
Registration of computer names and numbers is done by the NIC ("Network Information Center"), who can be contacted by e-mail.
Note that some commercial information servers will not allow access from an unregistered computer - the purpose of this is to thwart hackers, and to control export of high-tech material.
The DNS name or IP number only specifies the computer. You will have to supply further information, such as the destination user identifier, password and file name.
ftp (name or number)
.........................................Give user identifier "anonymous"
......................................... and your firstname.lastname@example.org as password.
Note this is for the DOS version, it may be easier to use the
or tar xvf
A "List Server", also confusingly called a "Bulletin Board" (BB), is a way of automatically disseminating comments, queries and replies via e-mail to a distribution list of subscribers, who share a common interest.
Users can "subscribe" and "unsubscribe" on request; this is handled automatically by the ListServ software.
You first have subscribe to the BB server by sending an e-mail message, then any e-mail sent to the BB by yourself or others is automatically re-transmitted to all subscribers on the current list.
This can be a good way of getting answers to questions on a particular subject directly from the "experts"; on the other hand it can become a source of vast quantities of junk e-mail (in which case you may wish to unsubscribe!).
List servers (also called net servers) can also retrieve text files (e.g. software sources) on request, returning them by e-mail.
"Usenet" is similarly divided into interest groups or "newsgroups" (e.g. BioNet newsgroup), but the incoming e-mail messages are accumulated at a central news server, which you have to search with a newsreader program; this means you can be more selective about which messages you choose to look at, and which you can ignore.
Gopher is a program that "burrows" through the Net seeking and retrieving documents, which can be multimedia (graphics, video, audio).
Gopher is basically a menu-driven file retrieval system.
It is more "user-friendly" than ftp and telnet because it avoids the hassle of having to remember DNS names or IP numbers.
You need a "Gopher client" program running on your PC/workstation, and a "Gopher server" program running on the remote machine that you call up.
Gopher provides gateways to network services (FTP, Telnet) and other information systems (WWW, WAIS, Archie, WHOIS).
One problem is actually finding a specific item on the Net. This is solved by Veronica (Very Easy Rodent-Oriented Netwide Index of Computerized Archives), which can do a keyword search of all Gopher menus world-wide
WWW (W3, or "the Web") was developed at the European Laboratory for Particle Physics (CERN) as a means of searching the Net for multimedia information, without needing to know anything about networks!
WWW seems to be the "in thing" at the moment, at least until a new fashion breaks!
The user needs a WWW browser program such as "Mosaic" (from NCSA, Univ of Illinois) running on the PC/workstation under Windows or X-windows.
Jumps across the Net are made by simply clicking on "hypertext" links (same idea as Windows Help); the user can be totally unconcerned about where these links are physically located.
However data transfer can be frustratingly slow, and the user has no idea how much data is coming! It is therefore useful to know that the "abort" button is the rotating globe icon at the top right of the main window!
WWW also has links to telnet, ftp, Archie, Gopher, Veronica, WAIS etc.
There are also others defining various graphic image formats, such as GIF (Graphics Interchange Format) and XBM (X Bitmap).
|Index to Course Material||Index to Section 1|
Last updated 14th Jan '96