Comprehension
An Invention to go ‘Gooey’ over
Graeme Phillipson, SMH 30th March 1999
|
Two weeks ago in this column, I wrote about Linux and how the Open Source philosophy it represents poses a serious and very welcome threat to the established order of things. That piece was very well received, except that I got dozens of e-mails from all around the world, pointing out that I had made an error in the article. I said that the first graphical user interface (GUI) for Linux was Gnome, whereas that honour should go to an interface called KDE. KDE stands for “Kool Development Environment”, which is- how techies name things. I stand corrected. I now know far more than I need to about Linux and its GUls. But it got me thinking about GUls, and about how we take them for granted nowadays. It was not always so. First, let’s examine the concept, in reverse order of the letters in the acronym. (For it is an acronym, because it is pronounced “gooey”. If it were pronounced “gee-you-eye” it would be an abbreviation.) A GUI is an interface. It is the place at which the computer meets the outside world. And a user interface is the place at which the computer communicates with the user, an actual human being. User interfaces do not come naturally to computers, whose native language is zeroes and ones. The user interface is itself a complex program that converts these binary digits (usually shortened’ to “bits”) to something we can understand. The first user interfaces were cryptic messages printed on a teletype machine, as characters such as these you are reading, no pictures. Things got a bit easier when we moved to video terminals in the 1960s, but the user interface still displayed its teletype origins. These were known as character-based interfaces. They become much more sophisticated over the years, but they were notable for a few specific features. Their key building block was the letter or number - the character. And their screens were 80 characters wide, exactly the same number as on a teletype machine or a punched card. This caused a few problems, because the average line of written text in a business letter or book is about 90 characters. That meant that what you saw on the screen was not the same as what you printed out, which caused all sorts of formatting grief. In the ‘80s, character based interfaces were still the norm on PCs and on most other computers. Then interfaces became graphical, and the world changed. The difference between a GUI and a character-based interface is profound. With a character-based interface, the computer was very much in charge, directing the user to a narrow range of functions. The systems were improvements. The use of standardised toolbars meant that most operational functions were similar, even on different systems. The use of pull-down menus meant that much greater functionality could be built into, each screen. The use of colour and high-resolution graphics meant that the screens were easier to read. But most importantly, GUI’s transferred control to the user. They allowed users to do much more than they previously could. Coupled with sheer power of client/server computing and networking, transferring data between applications became a simple matter of cutting and pasting. GUI’s liberated users. The GUI was invented at Xerox’s famous Palo Alto Research Centre (PARC) in Silicon Valley in the ‘70s. PARC was, and is, a centre for pure research, where many of the major innovations in information technology were first designed or conceived. A team at PARC, led by Larry Tessler, designed the GUI to work with a small office computer that Xerox was developing called the Alto. It worked brilliantly, but in true PARC fashion it was more theory than practice. In December 1979 a team of people from a small company called Apple visited PARC for a demonstration of the Xerox GUI. Apple copied the idea and incorporated it in a new computer it was working on at the time called Lisa. The Lisa failed miserably, but the GUI found its first commercial home , in the Apple Macintosh, released in 1984. By this time the idea of the GUI had caught on, and they were appearing on other computers as well. During the ‘80s a number of GUis were proposed. GEM and VisiON and TopDeck and others are forgotten now, which is probably just as well. The winner was Microsoft Windows, which soon caught up with the Xerox-inspired Macintosh WIMP (Windows, Icon, Mouse, Pull-down menus) interface and has since far surpassed it in popularity. There was an amusing moment in the mid-1980s when Apple had the audacity to sue Microsoft for stealing the ‘look and feel’ of the Macintosh GUI. This case went nowhere when Apple was reminded it had pinched, the idea from Xerox. GUI’s are now pervasive. Some people still use character based interfaces and command lines – they are better for programming and for some data entry tasks but they are in the minority. Now a new interface is with us Internet browser (such as Internet Explorer or Netscape Navigator). The browser is a type of GUI that allows even greater flexibility and power. And with greater power comes greater ease of use - GUIs are much easier to learn and to understand than character-based interfaces, and those skills are much more transferable. A lot of rubbish is being written about how the browser will replace the GUI. It will not - it is a limited interface with limited functionality. How do you do word processing on a browser? Why would you want to? But it is a fabulous supplement to the GUI, allowing great flexibility when thumbing through information (which is what “browsing’ is). We take the GUI for granted. But it has been one of the key inventions of the IT revolution. |
Questions to Answer
|