Peter Bowditch's Web Site

Advertising policy

ACS NSW Branch

Awards Presentation - Information Technology
North Sydney College of TAFE
Crows Nest NSW
March 1998

 Peter Bowditch presented this speech on behalf of the Australian Computer Society to the
North Sydney TAFE students receiving awards for the 1997 academic year.

Good morning, Ladies and Gentlemen,

When I was asked to talk to you today I of course asked what I was supposed to talk about. I was told "Anything I liked". As I have ten minutes I thought I would just cover the history of computers, where information technology is today, and where the future is taking us. That should allow some time for questions.

Seriously, I am going to look at two hot issues today and talk about where they fit into the history of IT and what we can learn from them about the future.

Before I do that though, I want to just give you a little history lesson about how what we do now is shaped by what we did in the past. When you turn on your PC you get a screen which happens to be 80 characters wide and which hangs around for a while until Windows starts. 80 characters is not a random or arbitrary choice - there were 80 columns on a punch card and the first screens were built to match this. For those old enough to remember, a punch card just fitted into a shirt pocket, which made them really handy for writing lunch orders on. The size of the card, however, was also not random. When Herman Hollerith set out to do the first mechanically-counted US census, he needed some way to store all those little bits of paper so he looked for someone else who needed to store hundreds of thousands of little bits of paper. The size of the punch cards was set the same as the paper money of the time because that made it easy to buy storage drawers. The 80 characters came about because that number could be punched reliably into a card that size.

The card fitting nicely into a shirt pocket was no coincidence either - the standard pocket size was designed to hold paper money. So, here in 1998 we have computer screens and shirts with designs dictated by the size of late 19th century US paper money. It actually gets better than that, because the size of a floppy disk is not random either - it was designed to fit into a shirt pocket.

This is actually a zip disk capable of holding the contents of 12 million punched cards, but its lineage can be traced back for a century.

On to the two big issues.

The first is the Year 2000 Problem. This is not a millennium bug, because it isn't happening just because a new millennium is starting (it is a century change effect) and it's not a bug in the sense understood by real programmers.

Predictions about what will happen range from scenarios where those people who are not killed when the machines go berserk will be starving in the dark to Pollyannas who tell us that everything will be perfect. The truth lies somewhere in between, of course. I tend to be optimistic, although if I were smart I would be predicting catastrophe. That seems to be where the consulting money is, even if the Chicken Littles don't have any solutions.

A lot of the doomsayers tell us that the problem comes from COBOL programmers who wanted to save space. This just indicates the ignorance of these prophets about both COBOL and computer industry history and practice. From the day of its invention, COBOL had the capability to store more than one digit in a single character position, but management decisions were made not to use this feature. We are also told about how hard it is to locate where dates are in programs and data files. Again, the answer to this has been available for a long time but management decisions were made not to apply the techniques and practices. Yes, there were (and always will be) bad and lazy programmers, but people can only work with the tools and rules they are given.

What we have learned for the future is to be more careful about what we do, how we plan and execute software development projects and how we manage the information we hold. An added advantage is that senior management and boards now know that there is an IT department somewhere inside their organisations, and may even have met some of the people who work there.

Here's a small message of hope about how well things might turn out. Many of the credit unions in Australia had their own personal Y2K experience on 17 May, 1995, when the operating system used for their main systems rolled over to a magic date. There was huge potential for disaster as interest calculations went mad and mortgages were automatically foreclosed because of long-overdue payments. Nobody went broke, and the customers didn't even notice.

Time for another history lesson. One of the major players in the invention of COBOL was also the person who first used the word "bug". She was Grace Hopper who, as well as being a pioneer in the computer industry, was also a Commander in the US Navy. Byte Magazine gave her some kind of lifetime service award and the editor of one of Australia's leading computer publications ridiculed them for it. He could not conceive that anyone from the navy could have ever had anything worthwhile to say about computers, especially said so long ago (although he did stop short of pointing out that she was also a woman). This outburst was a bit like a physicist denying knowledge of anyone called Newton, or a doctor who had never heard of Pasteur or Harvey. A small apology appeared in a later issue.

The second big issue of today is the Internet. It's a bit hard amidst all of the hype to remember that the Internet has been around since 1969. The explosion of the World Wide Web over the last four years has brought it to everyone's attention but it's not anything new. The original concept was to network some computers so that information could be shared in a manner which was efficient both from a technology point of view and in human terms. Computers are about information, and there is no information without communication. We just do it a lot better these days with fast modems and good browsers, and it is the growth in the Internet that has given people the incentive to keep making better communications equipment and infrastructure. We can all benefit if we can communicate better.

One aspect of the Internet teaches us about the resilience of a good idea. We all know that everything changes every day in the computer business, so it is revealing to see that the new version of Novell's network operating system will use the communication rules first developed for the Internet all those years ago This fossil called TCP/IP is also the preferred method of communication for Microsoft Windows NT (or Windows 2000 as we are supposed to say now), and is essential for Unix networking. Telephone companies are talking about using it for voice traffic. Like Peter Allen said: "everything old is new again".

It is obviously encouraging for people of my vintage to see that just being old doesn't make something necessarily useless, but it is encouraging for people new to the business too. If you go into it expecting daily change you will not be disappointed, but it will help to know that someone may have learnt what you need to know before and made the mistakes that you don't have to repeat. It isn't really complete chaos out there in computer land. It just looks like it.

Thank you for your attention.


Copyright 1998- Peter Bowditch

Logos and trademarks belong to whoever owns them