What did people do on the internet before the Web was invented?

Playing a game using telnet (photo by maczydeco - CC-BY)

The World Wide Web was invented in 1989, but it was not publicized until 1991 and did not really start to catch on until the release of the Mosaic web browser in 1993.

But the internet has been around much longer than that. It grew from ambitions expressed in the early 1960s and implemented in embryonic form as ARPANET in 1969. By 1983 the core protocols of today’s internet such as TCP/IP were in place. So how did people use the internet until the web came along?

There was electronic mail, of course, which had already been around since the early 1960s. It functioned much as today’s email systems do, except that there were no web clients and spam had not yet become a problem. However, even by 1993 email addresses were not as standardised as they are today. In addition to the modern form (user @ domain.name) there were UUCP addresses (computer ! path ! names ! user @ uucp-host), the convoluted X400 addresses that only a committee could love (G=given; S=surname; O=organization; OU=dept; PRMD=uninett; ADMD=uninett; C=country), FidoNet numeric addresses (1:234/567.8), CompuServe octal addresses (12345.6701 @ compuserve.com) and others.

Business and social discussions took place via Usenet News, a bulletin board system that is still in use today and which is also available on the web via the Google Groups interface. There were numerous newsgroups, mostly unmoderated, in the following categories:

  • comp: computer-related topics
  • news: discussion related to Usenet
  • sci: scientific topics
  • rec: recreational activities
  • soc: social issues
  • talk: topics of current interest
  • misc: all other topics

Although Usenet was only lightly controlled, there was demand for topics beyond those which it included, and the “alt” hierarchy was created for these. Anyone could set up any group they liked under “alt”, but they couldn’t force ISPs to carry it. Some ISPs prided themselves on providing as complete a Usenet service as possible; others preferred to stick with the mainstream newsgroups.

Text-based chat used the Internet Relay Chat system (IRC), which is still going strong today.

There were several different ways to retrieve information. The oldest of these still in use is FTP, the File Transfer Protocol, first specified in 1971. This was used mostly to retrieve technical documentation and software installation packages from FTP servers, but there were also a few FTP servers that hosted a wider range of material (lists of hiking trails, ham radio logs, pictures of cats, etc).

Rising to prominence just before the WWW was Gopher, which delivered information by the page much as today’s web does. Each Gopher page consisted of a hierarchical menu, or text, or both. By clicking through the menus (or more commonly by typing the menu option number) the user could navigate to any page. The pages could be static or dynamic, and the dynamically-generated pages offered a range of services that offered a glimpse of what would later come to the web. There were Gopher pages to display the current phase of the moon, generate ASCII graphics, provide currency exchange rate calculations, and so on. There are still a few Gopher servers operating, mostly for reasons of nostalgia.

With all this content on the internet, how would the user find it? The obvious answer was by search engine, and the Archie search engine indexed Gopher pages and FTP files. An improved search engine, called Veronica, later became available and is still used on the internet (just in case you need to access the nostalgic Gopher pages).

Another internet protocol that’s still around, but was more widely used in the past, is the “finger” protocol. Given an address, the “finger” application could be used to obtain a corresponding nugget of information. It was originally devised to retrieve user information (sort of like a very basic “home page” for a user) but its use broadened over time. My ISP still offers a “finger” address by which subscribers can obtain service status information when its website is down.

A software application could be connected to the internet and accessed by users around the world using “telnet”, another protocol that lives on today (although more secure versions such as “ssh” are usually preferred). I remember being amazed in 1992 at being able to log in to the US Library of Congress and displaying their catalog data. Computers could also interact with other computers on the internet to exchange information, using Remote Procedure Calls, much as they do today using protocols such as SOAP and AJAX.

So the internet was a useful and lively place, long before the birth of the world wide web. Once the web came along, it rapidly grew because it was easier to use, and also because it was easier to write content for. The explosive growth of websites eventually replaced many of the older services, but others live on because (for some users) they provide something beyond what is provided by the “one size fits all” web.

Oh, and spammers generally can’t be bothered with the older protocols. So if you want a website that spammers won’t try to hack, your best bet may be to set it up as a Gopher site instead!

Here’s an example of a Gopher server that was still functioning as at April 2010 (copy the address into the Firefox address bar):

  • gopher://gopher.floodgap.com/

and here’s a web interface to an Archie search engine.

Related questions:

  Need research? Quezi's researchers can answer your questions at uclue.com

Written by | 5,325 views | Tags: , , , ,

1 Comment

  • Jim Goltz says:

    Don’t forget WAIS! The Wide Area Information Service was a library-oriented protocol that allowed for free-text search, author/title search, and lots of other nifty things that the Web lacks today. The protocol was nothing short of bizarre, however.

RSS feed for comments on this post.

Privacy Policy | Acknowledgements