The Information Cul-de-sac

Fifty years ago, computers heralded in the Information Revolution. Twenty five years later, the advent of the microprocessor provided the means by it could become of age and computers finally began to infiltrate the very fabric of our society. Today, the Internet promises even more. The vast web of computers spanning the globe is literally changing society, shattering traditional social and economic borders.We are speeding down this so-called information superhighway, driven on by the media and big business. Unfortunately, we don't have a map! We don't seem to know where we are heading or even where we want to go. We don't even seem to care! But what if the highway turns out to be going in the wrong direction or, even worse, to be a cul-de-sac! It seems to me that we have, indeed, been heading in totally the wrong direction for several years now. Maybe this was inevitible, even necessary, so that we can ultimately find the right road, but I am not so sure. Let me try to outline some of the trends which have caused me to think in this way!

I will start with the thorny question of freedom of speech, a topic which has received much attention and generated much debate of late. I do not have much to add, except to point out that freedom of speech can be dangerous if combined with freedom from responsibility. The Internet provides a freely available platform on which to publicise ones views, which is all well and good. Unfortunately, it currently fails to provide a mechanism whereby people can be held responsible/accountable for those views. Of course, I do not mean by this that states should be able to prosecute or persecute individuals or groups for their beliefs. That is exactly what (in America) the constitution and the first ammendment were intended to prevent. However, being able to persecute, harass or slander another person, is still considered immoral and is prohibited by other laws. The basis of these laws, however, is that the author or publisher can be held responsible and ultimately punished, if need be. Complete annonymity, however, as provided by the Internet, means no recourse to the law and hence no comeback whatsoever. This, to my mind, is downright dangerous and should not be tolerated. So, freedom of speech yes, but only if we can be sure who is doing the talking. Note too, that knowing the identity of an individual might also defuse some of the problems with copyright protection.

Another hotly debated topic is the security of electronic transactions. I suppose business had to come to the Internet. For the most part I don't object to it, in fact, I suppose I benefit a lot from the information that places and businesses like Microsoft, Borland, CNN, and so on, make available. The little adverts at the top of many web pages, sometimes take a bit of time to load and can become quite intrusive. Again, I suppose it had to come and, as long as it doesn't get too obtrusive, we just have to live with it. Anyway, back to security. Purchasing things via the net requires some form of electronic funds transfer. Banks have only recently begun to trust such transactions over their own secure networks, perhaps understandable given the large sums involved, but why should there be any difficulty for small private transactions? Sure, messages containing financial information can be intercepted, but only by individuals with considerable technical expertise. Incidently, the fact that this can be done so easily (and the fact that people can remain annonymous so easily) is due to the poor design of the networking and operating system software which was not really intended for such applications. Anyway, plugging these holes will leave potentially much bigger ones still wide open.Why? Because shopping relies on trust. As a customer you rely on the sales agent to act in good faith. When shopping in the high street there is usually no problem. Chances are, the shop will be known to you and will be there tomorrow. Mail order shopping is much more risky, simply because, unless the company is well known, there are no such guarentees. On the Internet the situation is even more fluid, so it seems to me that the slight risk of your credit card number being intercepted is miniscule in comparision. In fact, there is much more chance that the salesperson in your local shop will make use your details!

One last point while on the question of security. We tend to assume that our private communications are private. Unfortunately, email does go astray. I have certainly had several random items reach me! Of course, reading anyones files and email is not difficult if you have their password. Looking over someones shoulder in a crowded cybercafe or installing little programs which automatically record them is all too easy.

On a more technical note. The web has adopted, or rather evolved, a standard way of representing information known as HTML, for Hypertext Markup Language. This is actually a derivative of SGML (Standard Generalised Markup Language) which embodied, perhaps for the first time, the idea that document content and its "semantics" should be seperated. What this means is that, for example, a document's title, its author, their address and so on, should all be clearly identifiable as such, not just by an intelligent human, but by a machine. To this end, additional "tags" (special sequences of symbols) are added to the document in order to mark these items accordingly. These allow a machine to scan the text and automatically pick out any desired information. HTML is a relatively simple and natural extension of this which allows us to specify other documents which contain information related to particular sections of our current document. These are hypertext links and our computer can automatically recognise them, and be made to fetch the other document, if we so desire (all we do is "click on the link".) Great!

Unfortunately, there are two little details which bug me. First, these markup languages seperate semantics from content, but they do not say anything about presentation. This is deliberate, the idea being that, when it comes to viewing information, people have individual preferences. For example, some, because of poor eyesight or simply bad viewing conditions, may find a large font size is helpful. Others may not be able to see text displayed in certain colours, such as red, and would thus appreciate it being shown in green or blue instead. Moreover, certain display equipment may not be able to show particular forms of material properly, for example, italic fonts. It thus makes sense to build in some latitude so that people can adjust the presentation to their individual circumstances. But authors also want and indeed need, some control over the way their document is presented. It would be ludicrous for the display system to shuffle the sequence of words so that, for example, there was exactly eighty characters on every line, since clearly the message would be lost! Similarly, the layout of complex graphics and other elements is important to the overall impact and meaning of a document. Authors must be able to specify these things, even if the individual user can choose to ignore them. Only very recently have significant layout facilities begun to be incorporated into the HTML standard.

Why have we had to wait so long for something which is so obviously necessary? I don't know. Even more amazing to me, is why we appear to have gone backwards in terms of the concept of what a document is and how it is handled. It has taken years to develop the idea of OLE (object linking and embedding) as an open model for documents containing any form of information. Yet, suddenly we return to the old days! Worse still, when it comes to creating and editing HTML documents, we return to simple text editors in place of the WYSIWYG word processors we have got used to. Why are we expected to learn HTML's cryptic commands, to insert <P> everywhere we want a paragraph and to surround italicised text with <i> and </i>. Adding images and links is even more laborious. Unbelievably to me, no-one seemed to notice or even care. In fact, many teachers were happily explaining the advantages of spending classroom time to get kids to learn HTML coding. Surely this is just a return to the 1960's form of computing, when people bent to the machine. Didn't we learn that machines are supposed to bend to us? Incredibly, only very recently have WYSIWYG HTML editors began to appear.

Want to know something else which is really stupid? Well, you can now specify the font that you want on your web page. Great, however, it is WYSIWYGBPNWTG! That is, "What you see is what you get, but probably not what they get," unless they happen to have the same fonts installed on their machines as you do. So, it really doesn't make sense to include fancy fonts on your webpage, because the person browsing it from across the world, probably won't be able to see or print them. The only solution is to use graphics, but that means longer downloads and the search engines can't index it (unless you add alternative text!)

Ok, last but one (for now!) The hottest thing right now is a programming language called Java. This is supposed to be the best thing since sliced bread. It runs on every type of machine, Unix, PC and Mac. It runs across the Internet. It sings and dances (well almost.) Great, eh? Well yes, but... basically it is C or C++ with some of the more stupid idiosyncrases removed. Unfortunately, not all of them have been taken out. If nothing else, it has sacrificed a relatively clean syntax (such as that of Pascal, for that of C), just to gain commercial advantage. We want our programs to be beautifully laid out so that they are easy to read, but then we design compilers to accept text laid out any-old-how. We missed a golden opportunity to get rid of those stupid curly brackets (or begin/end's in Pascal) once and forever. I may be wrong, but I think we have actually lost something good too. Pascal's enumerated types may not have got used as much as they should have, but they did make programs easier to understand, which is surely what we want. I could go on, but let me not bore you now, suffice it to say that I like Pascal and Delphi best! I guess we will just have to wait for the next universal language before we get it right (yes, there will be life after java!)

Finally, the software industry has begun to release products for "public" beta testing, a trend I find worrying. At university, we teach our computer engineering students that software must be properly designed, it must be bug-free, right-first-time. Civil engineers, we tell them, don't get several attempts at building a bridge. You wouldn't dream of exposing passengers to planes that you weren't absolutely sure would fly. Why, we say, should software be any different? They used to listen, but now they see giants like Microsoft and Netscape doing exactly the opposite. And not "just" with programs, but even with the operating system itself! What does the future hold?

So, that's some of the reasons that it seems to me that we might be heading in the wrong direction on the information superhighway or even that it may turn out to be a cul-de-sac. I hope I am wrong, that things will turn out ok after all. Technology and the Internet can make the world a better place, but let's think about how we can and should be using them.


Copyright (c) 1996, David Davenport