gian@gandalf.it

No. 22 - July 5, 1998

1. Editorial: A glimpse of
   common sense?
2. Killers, enzymes and turbulence
3. Littletown, Metropolis...or a continent
4. The value of small numbers
5. "Power" doesn't like the net
6. Banner blues

I see no signs, so far, of a reverse trend on this side of the ocean – and especially in my country (except for the almost "underground" activity of people that fix old computers or explain how it’s possible to be on the net, and do all sorts of other things, with three or five year old equipment). Though some well-known writers, such as Umberto Eco, are discussing the problem of "too much technical development", and a bit of that, sometimes, is reported in the media, fatware remains dominant. Unnecessary and undesirable technical complexity is plaguing our lives in many areas, not only computers.

In the United States we seem to see the beginning of a healthier trend. Some manufacturers are beginning to offer computers that are still much more powerful and complex than necessary – but at least they are less expensive. A small step, and a long way from a real solution; but at least it shows a glimpse of awareness of the problem.

In the overwhelming proliferation of conventions, debates and seminars (on everything, including the internet) often little side conversations are more interesting than the official speeches. I am hearing more frequently comments about the need for lighter solutions. Some people say that even the traditional fatware producers are beginning to re-consider their strategies. I hope they are right.

Just a few little sparks in the dark, so far. But there seem to be e few chinks in the armor of heavyware. We shall see...

In his introductory speech, professor Enrico Valdani concentrated on so-called killer applications; and this was one of the key subjects in the meeting. Warlike jargon is often used in business and it doesn’t worry me; but in this case I find the definition disturbing. Who are the murderers? And who is going to get killed?

A killer application, he said, is something that "asserts itself in such a pervasive and irresistible way as to destroy the established products and services with which it competes". In the history of technological development there have been, sometimes, such developments. A hundred years ago a "killer" was on its way: the motorcar. In some parts of the world it replaced horses, donkeys, mules, oxen, camels, elephants (and humans) as means of transportation. But those animals aren’t extinct. We still use horses for riding and racing; we still breed cows, bulls and calves, though maybe fewer oxen; and we protect other species when they are at risk. It took a while for steam ships to outspeed the clippers, but eventually they prevailed (though there are more sailboats now, for cruising and racing, than there have ever been). A "killer" technology can cause all sorts of trouble... look a the huge effort needed today to move the transportation of goods (and people) back from road to rail. And the energy and environment problems...

Many things in electronics, and in the internet, developed quite differently from what was expected. The World Wide Web was conceived in 1990 as the private tool of a few physicists. Four or five years later, quite unexpectedly, it took over the world. It didn’t kill the internet (in fact it widened its use) though for some new users it’s a layer that makes some of the net’s best values less visible... but that’s a matter of culture and education, it can’t be blamed on the technology.

A few years ago "video on demand" was supposed to be a "killer". It didn’t happen... and it didn’t kill videotape. Tape could be replaced by DVD... but where is it? It was supposed to be on the market early this year but it’s held up by non-technical problems... nobody seems to know for how long. A year ago Network Computers were supposed to replace PCs as the basic internet tools; now some devices of that type are considered viable, but only on centralized intranets. Cable was supposed to be crucial, but now they say the future is wireless. And so on...

The "killer applications" I would welcome are light, cheap computers with very simple operating systems, that could sell for 200 dollars or less. Of course the technologies exist and work very well. The problem is that most people are not aware of these options, and too many applications are written for the heavy systems. There are no technical reasons why this can’t be done; the problems, as usual, are in the market and in the culture.

Now (they said at the Bocconi meeting and they are repeating in several other places) the new fashion is something called "Web TV". I must admit that I am rather skeptical. In most homes the TV set is in the living room, or some other shared space: not very appropriate for internet connection. There are many families with more than one set, but those are the people that own a computer or could afford to buy one. Another obvious problem is that if people think they ca use a TV device instead of a computer they will be deprived of many useful functions: not a good idea in a country, such as mine, that is below standard in the use of information technology. And ... this contraption is likely to encourage a very superficial use of the net.

Will it succeed? I don’t know. It might, for a while, if major TV networks fall into the trap and promote it heavily. If it does, the only benefit I can see is that computer manufacturers and merchants (especially those that wont’ be getting the big orders for web-tv equipment) may be prompted to compete with low-price computers.

In that meeting a second example was given of an upcoming "killer application": e-commerce. Literature about the net is full of quaint horror thrillers, but this is an especially grim one. Here’s a child, not yet toilet-trained in the US, still an embryo over here, that turns into a serial killer. What is supposed to murder? It may, over time, reduce the use of some telephone services (and save companies a lot of money). In some service areas it could weaken those intermediaries (travel agents, insurance brokers) that do only a routine job and give no service or advice to their customers; but that would be euthanasia rather than murder. Generally, I think it will develop alongside traditional commerce, not replace it; and there can be several opportunities for integration, lead generation, etcetera. For instance online book and music stores know that over nine tenths of their traffic lead to purchase in a traditional shop.

Generally, I think we should be very careful with "killer applications". There can be – and there will be – solutions that will totally (and harmlessly) replace existing ones. Is anyone nostalgic about the 5.5" floppy? Is anyone going to miss modems when all phone lines will be digital? I’d be quite delighted if they had a computer (with internet connection) in each hotel room and I could travel without a notebook and just carry in my pocket the standardized equivalent of a jaz disk, or a cd-ram, or even a 5-gygabyte dvd-ram... with room for much more than I can possibly need. And so on... But other "killers" can be quite dangerous, especially if they spread before their effects are fully understood. A bad doctor may use wide-range antibiotics to avoid bothering with a diagnosis, but that kills the intestinal flora and builds resistant bacteria strains.

I think the cultural enzymes of the net should be nurtured and protected. I also believe that its development should continue to be turbulent and unpredictable. Diversity should not be flattened out by repetitive and indistinct "killer" applications. Of course we need shared standards and compatibility. But that can be better achieved with open architectures and interchange than by freezing evolution into some pre-set formula. That could be temporarily profitable for the killers, but maiming for net culture and development.

However, I think one figure in the research that was presented at the Bocconi meeting deserves some comment. According to that study, there are 128.000 people in Italy that "bought something online at least once". I don’t know how accurate that is (it’s better to be doubtful about all statistics) but in any case I don’t understand why the "news" was presented with enthusiasm. It’s a substantially larger number that two or three years ago, but still very small. A bit of analysis can make it even smaller – or much larger.

Let’s start by slicing it up. I know this isn’t very precise, but more detailed arithmetic wouldn’t change the result significantly. If we leave out software download and ordering from abroad of books and other items, we are left with less than 20 percent of those people buying something in Italy from an Italian site. That’s approximately 25.000 people: 0.04 percent of the population – or the size of a small town. Not appealing for anyone thinking in a "mass market" perspective.

But we could look at things in a different way. A really valuable and interesting offer could attract people that don’t have an internet connection but know how to ask a relative, a friend or an office colleague to do the searching and place the order. Now we are talking about four or five million people – maybe more. Not as large a market as New York, London or Paris, but larger than any big city in our country. Of course no individual company or offering could reach all of those people, but it’s a pretty wide arena in which to choose the best customers for a tailored service.

And there’s more. The most attractive market for many companies, large and small, is export. If we include relatives, friends and office colleagues, there’s a market out there of over 200, maybe 300 million people (60 or 70 percent in the United States, most of the rest concentrated in a dozen countries). Now we are dealing with something almost as big as half of Europe – and growing. Lots of competition, of course; but lots of opportunities if we have something really special to offer.

So... why should we care about how many people have bee buying online so far? What matters is what markets can be created by really attractive offers, supported by first class service. The potential is quite interesting. The problem is that very few online "advertisers", so far, have been willing to invest in the effort, consistency and dedication that are needed to succeed in this new environment.

There are supposed to be over ten million people in Europe with an online connection and a mother expecting a gift. A 0.0005 percent result would look rather disappointing if the promotion had been planned on a "mass market" strategy. But a reasonable objective was set, so it’s reported as a success.

We all know that in our country laws and rules are deliberately made complicated, cumbersome and incomprehensible, in order to keep people and companies constantly confused and subjugated by the administration.

Until a short while ago politicians and other people in power had no idea of what the internet is or can become – and didn’t care.

Now their are beginning to realize that it’s there, and it’s not going to go away; but they still don’t understand it and that makes them uncomfortable. They will do everything they can to bring it under control, try to turn it into something they understand, such as one-way broadcast media; and stifle it with all sorts of rules, regulations and inefficiencies.

Is that just one man’s whimsical opinion? No. These things are widely known, but rarely said. I’ve been writing about them quite often, for instance in Cassandra, Aladdin, Ulysses and Polyphemus and recently Nannies, bibs and gags. These and other articles of the same sort met with a lot of approval by people who understand the net – and of course were ignored by the big media and the people in power.

There are several examples of poor legislation concerning the net. After the infamous "Italian crackdown" of 1994 (that was originated by an investigation on unregistered software) and the resulting protest there were other legal proceedings on different issue and some of the investigating magistrates and policemen had understood that the seizure of equipment is illegal and unnecessary. But recently things have been getting worse and several innocent people have been damaged by computer seizures. A recent episode was particularly obnoxious and is causing a great deal of protest. On June 27 and internet server in Bologna, used by thousands of people and a large number of public service organizations, was seized by the police on a court order from Vicenza. All mail and web activities based on that server were suddenly obliterated. The reason for this unbelievable decision was a single message (one of thousands on that site) was a "libel" action on the part of a travel agency, that organizes trips to Turkey, against a message denouncing the persecution of Kurds and suggesting a boycott.

Is this a problem for society but not for commerce? As I said seven months ago I don’t think these are separate worlds. Maybe we won’t get arrested if we order online a book by Lewis Carroll o Vladimir Nabokov. We can be lucky and not have our server seized by ignorant authorities for ridiculous reasons. But an unhealthy, censored, restricted and persecuted net isn’t good for anyone, including business.

There seem to be problems with banners worldwide; even in the United States. Here is an article by Jessie Berts on June 16:

When was the last time you clicked on a Web ad? Not recently, gauging from click-through rates. And the folks paying for those banners are not happy.

Web advertising is the primary source of revenue for many sites (including this one). But Web advertising is going through a rocky stretch right now. Consider:

Only 1% of people who see a banner ad click on it (on average), according to new estimates from NetRatings. That's down from 2% in 1996, according IPRO and DoubleClick research.

Web ad supply exceeds demand. Adauction CEO David Wamsley says 50% to 70% of the ads on the top 500 Web sites go unsold. (The oversupply enabled him to build a thriving business auctioning off unsold Web ads for a fraction of the full price.)

Web ad rates are down. AdKnowledge, the Web ad management company, reports the CPM (the cost per thousand impressions) for May was $37.84, down from $40 a year earlier.

These problems have led some industry thinkers – including venture capitalist William Gurley – to argue that Web advertising is doomed to follow a "pay-per-view" model. In this system, vendors only pay when somebody responds to their ad. For instance, Internet ad seller DoubleClick offers $1.00-per-click price as a "direct marketing" alternative to CPM rates.

The concept isn't new. On TV, for instance, this model is used for late-night infomercials. Stations get paid according to how many people phone into the 800 number.

But pay-per-view hasn't taken over TV and – contrary to the pundits – it won't take over the Web. It shifts all the business risk to the Web site, and that's not a viable long-term model. Think about it. If the product is shoddy and doesn't produce a response, who pays for it? The Web site. If the advertising is poor and doesn't produce a response, who pays for it? The Web site. And that explains why you only see infomercials on less-popular stations during off hours. Those stations can't sell the space any other way. They're desperate.

There's a glut of ad inventory on the Web right now, coupled with a shortage of good ways to measure the audience. As we solve those problems, Web advertisers will do what TV and print advertisers have always done. They will gravitate toward quality, not quantity. Toward the best sites that can deliver the best audience. Sites that switch to pay-per-click exclusively will go out of business.

Pay-per-click will exist on the Web, but only as one of several advertising models. Quality media and quality products won't use it as their primary ad vehicle. For the same reason you are pummeled by call right now! ads on late-night cable shows. But never on prime-time TV.

It’s even worse over here. From what I hear, the banner salesmen aren’t reaching their targets. Large sites aren’t getting the income they expected and some small online organizations are finding it difficult to give away advertising on their sites. There is too much hasty, unplanned, poorly conceived activity on the net. We need a radical change of approach; that begins to be said fairly often, but it’s still rarely done.