This was written for a possible book in around 1993/4. Written in Microsoft word, copied to HTML and salvaged by hand. ..Tim BL.
This section was started as a snapshot of how to get on the web, but I don't think a paper book of this sort should have anything except the general principles, leaving it to the Internet forf Dummies etc to be revised every 6 months (KUTGW Margy!) to addresses that.
This is a practical chapter. It shows you how to actually use the web, how to find things, and how to put your own information into the web. We look at specific programs as examples.
[This will have to be split into many chapters I suspect]
Here are practical details of how to get going. The most important step is to pick up a client and install it. After that, you should be able to us the web to find out everything else, including most of the information in this book. Of all the chapters in this book, this will be the most out of date even before the book is printed, so once you are up and running, browse around to see what's new.
Getting on line: getting a client
Am I on the Internet?
This is a brief overview of the ways of connecting to the Internet. See the bibliography for more details and lists of service providers..
At the time of writing, all WWW clients run directly on machines which are "on the Internet". This means, in some way of another, the computer can directly send Internet Protocol packets onto a network which is part of the Internet. This is not (in 1993) the case for most computers at home. Computers on Local Area Networks (LANs) within organisations are often on the Internet because somewhere there is a router which connects the LAN to the big Internet. Personal computers have to have suitable TCP/IP software installed, and then they can directly send packets out.
For example, Macintosh computers can have "Mac TCP/IP" installed, and then they can send IP packets out over the local appletalk to a box which will route them off to the Internet.
If you don't know whether your machine at work is on or could be on the Internet, ask someone who will know. If the answer is "no", it may be time to show them this book or one of the books on the Internet and make some suggestions in the right areas.
If you are not on a network, but have a modem and a telephone, then you can get a temporary, quite slow, but quite genuine Internet connection by dialling in to an Internet service provider. What happens is that packets are sent over the telephone line (normally using "SLIP" or "PPP" protocol) and the machine at the other end relays them to and from the Internet.
If you have some kind of "telnet" or "FTP" program or application (and it works) then that indicates that you are on the Internet.
I'm not on the Internet
If you are not on the Internet, then maybe you can connect somehow to a computer which is. Then, you can run a WWW client on that machine. It is not as good as running it locally. This is because you certainly won't have a flashy window interface, you will have just terminal emulation, and also because it will be more difficult to download files to your machine.
The books on the Internet (see bibliography) give more details about finding places to dial in to for Internet access.
Dial-in Web
If you are on the end of a phone line, you can, as I mentioned above, arrange to get an Internet connection over a dial-up line, or you arrange to connect to a computer which is on the Internet and runs WWW software.
A third possibility may in the future to run what is known as a "split client". This is an arrangement whereby part of the client runs on your computer and part runs at the other end of the phone line. The two halves talk to each other in such a way as to make the best possible use of the limited speed of your modem. At the time of writing, I don't know of any such facilities, but by the time you are reading this it may be available.
I am on the Internet but I don't have WWW.
In this case, all you have to do is to get the WWW software for your machine. This is described below. It may be, though that you cannot install any software on the machine. Maybe it is set up to only allow you to log in through them to other machines, maybe you don't have the right to create any files. In this case, you can use the "telnet" command to log in to a machine which gives a public access to WWW. These machines either need no username, or take the username "www" with no password necessary.
Here is a list. This service is generally provided out of good will, and may be withdrawn by the providers at any time. Remember that you are using other people's resources when you log in to them, and so don't abuse them! Here are the Internet host names
info.cern.ch
ukanaix.cc.ukans.edu
@@@ complete me from the web. And IP numbers.
A more recent list is available on the web itself (it is a Frequently Asked Question) and so you can see whether there is one more convenient for you than those listed here.
Picking up software
Using FTP. Sites to try. Reading READMEs. Checking the web for status.
You have, or you can access somehow, a machine with an Internet connection. You have looked around, but the WWW software doesn't seem to be installed.
You can pick up most of the freely available software directly over the Internet. If you have TCP/IP software, then you probably have an "FTP" file transfer program, and a "telnet" program. You can check up on the status of the software by telnetting to one of the public access sites, or you can have a look at one of the following:
(dump status summaries from the web)
If you have a PC
If you have a Mac
If you have a unix box
If you have VMS
If you have VM
Selecting your home page
Finding a good place to start. Bookmarks and hotlists. Customizing your page: refer forward to authorship.
Remembering where you've been: Hotlists and bookmarks.
Finding things
Searching by subject: W3 virtual library. Indexes vs trees. Examples ... examples..lots of screen dumps. Finding the Vatican library renaissance culture exhibit. Finding the name of the person responsible for data.
As the web gets bigger, it becomes more difficult to find things in it. The WWW interface allows you to browse across links and to query indexes. This means that as you search you rely on the people or programs which made the links and the indexes. There are two big differences between looking for things in hyperspace and looking for them in real space. One is that once a link -- some relationship -- has been discovered, you can navigate across it just as easily no matter where the end points are without having to move from your seat. There is no distance in hyperspace. The other is that, because hyperspace can be navigated by computers, programs can be made to do some of the cataloguing, indexing and linking automatically. Of course people can do a more intelligent job, but they often don't have time to cover vast areas in detail. What we need is a combination of navigational aids made by people and by programs. And that is just what we have.
The Virtual Library
Books have always been classified by subject matter, and if you are starting out from scratch looking for a piece of information there is no better way to begin.
There are many classifications of information by subject. We started one at CERN, and others have started theirown. We keep a list of such top-level subject trees. There are a few different such lists too, which begs the question of a list of lists. The number of top-level subject listings follws from the fact that it is easy to get started making a classification, but difficult to make it complete and detailed. The Virtual Library started by CERN has a feature that different subjects (departments) are in fact stored on different machines and maintained by different volunteers. This reduces the amount of work, and allows each field to be handled by someone who is familiar with it. The seamless nature of the web allows one to browse through the library without being aware of this. It is as though the top specialist collections of the world were all available at your local library.
In practice, information is much easier to organise by other criteria, such as where it came from or the date. Do not be surprised, then, to find that at the lower levels of the subjet tree, the subject categorisation breaks down. At the time of writing, for example, the overview of "High Energy Physics" consists of a list of resources at each institute.
Mesh or mess?
Should be somewhere else...
Knowbots (TM)
Vint Cerf's contraction of a "knowbot" for knowledge robot sums up the idea of an automatic agent which goes around organising information. For the first time, vast amounts of initially unrelated material are now accessible in a consistent way on the web. This information can be accessed by computers as well as people.
@....
However, as the web can be explored by computers,
Contributing
After while, you have read enough of what other people have said to feel that you have something valid to add.
Adding a new node.
Making links
Tour of the Clients
Overview table
For each of the following: how to get it, points forts, points faibles.
Cello for the PC
Mosaic, Samba for the Mac
Mosaic for X/Motif
tkWWW for X
Line mode www for anything
html mode for emacs
Running a server
(Largely from the www server guide?)
When to put up data. Don't be shy of incomplete collections but state the quality people can come to expect.
Easy server
How to set up httpd.
If you just have some files.
If you already have an FTP archive.
Example of what a directory screen looks like with various options to the server.
Using the rule file. Making the document names time lasting over time. Logical names. Common mistakes. Examples.
Telling the server about file types and conversions.
NCSA server
If you already have Gopher.... running scripts. Access Authorisation
Perl servers
Serving Gnu "info" data. Other fancy features. Access control by IP address.
Making a gateway through the corporate firewall. HTTP, WAIS, news, gopher.
Web etiquette for server managers
(basically as on the web now?)
Welcome page -- absolute links, sign it. Alas for your server. Alias for yourself.
Making your own server program
Examples of some csh server
Hypertext Markup Language
SGML and HTML
An example to start with
The elements.
For each element, attributes, Purpose, Examples, what is allowed inside it, How typically it should be presented.
Formal rules: what's allowed and what's not.
Hypertext style for authors
(The style guide as on the web)
For programmers: generating hypertext views of databases
This is a place for a long explanation with examples and pictures as it is only in the book that we have time to put this across. Text from the talk on virtual hypertext.
Making HTML from other things:
From a directory
From Tex
Marking up plain text
What to do with your hypertext
100 fun things to do with your hypertext or anyone else's. Printing it off: as plain text, as TeX.
Indexing it.
Mapping it.
Making C code
Literate programming with HTML
Making HTML from .h files.
Regenerating the .h files.
Examples.
Building a client: The WWW code library
Control flow and data flow diagrams. Run through each module's responsibilities. What you have to provide. What you can override.