INTERVIEW:
Question 1
Could you tell us about commerce on the Internet, the past, present and future
developments.
Answer
If you look in the past, there hasn't been a lot of commerce on the Internet if
you go back two or three years. I think 1996 is the year when commerce and real business,
electronic business started taking off. It's really going to take off in the
business-to-business world first. There's been a lot of talk about business in consumer
commerce but the real growth in that will come a little later. The big focus in 1997-98 is
going to be in the small to medium enterprise working with each other and working with
large customers in business-to-business commerce. It's going to be supported by the kinds
of developments that we and others are working on, on payment infrastructures and security
technologies. That has been the main gating factor thus far. But with the introduction SET
technologies and some of the other security technologies like cryptologues that we've been
developing, the infrastructure is in place to actually drive commerce. And we're finding
that many of our customers are getting ready to move large parts of their commerce onto
the Internet.
Question 2
What are the pros and cons at this point of commerce on the Internet?
Answer
Let's start with the pros. The main pros of doing business on the Internet is the lower
channel cost for incremental revenue. That's the number one pro. You get incremental
revenue at much lower channel cost of doing business than you will using any other medium.
If you take banking: if somebody walks into a bank branch, it costs about US$1.50 for a
transaction. On an ATM it's about 25 cents. On the Internet it's about 2 cents. So that's
the reduction in channel costs we're looking at. The second example is increased customer
satisfaction. Because customers can come in on their time, on their place to do business
with you, their satisfaction goes up. Because they can get access to information when they
want it and they can get access to information quickly their satisfaction goes up. It's
that combination of increased customer satisfaction with lower channel costs of
incremental revenue, that's the real pro of doing business on the Internet.
The cons of doing business on the Internet is primarily that for a large number of our
customers it's a whole new way of doing business. They don't know how to do business on
the Internet. They have to learn how to do it. They have to feel comfortable with the
channel. So that's the main con when you look at it from the customer's perspective. He
sits there and says, I don't quite understand this medium. And that's our job - to help
him with that.
Question 3
What about security?
Answer
There's been a lot of talk about security being a con but it's not really.
The level of security that's available today on the Internet is very, very high. Let me
give you an example. In the United States we've got a joint program with sixteen banks and
we're driving full-service banking on the Internet. These are sixteen banks that have 60
million households—that's over half of the retail banking population in the United
States—that will set up over the next year to year and a half to do full-service
banking on the Internet. Security is not a problem.
Question 4
What about virtual banking? Is it going to become very interesting or is it just an
experiment?
Answer
Virtual banking is picking up already. Deutsche Bank now does full-service Internet
banking. Barclay's Bank is on the Internet, National Westminster Bank is on the Internet.
So many of these banks are on the Internet reaching out to consumers. What is very
exciting as well is that banks are leveraging the Internet to build a forum, to build a
community of small and medium enterprises so that they can do business with each other. So
for example, in Italy we are working with Cariverona. And we're working with them to
create an infrastructure across the Internet where small and medium enterprises can come
together and do business with each other. So I see virtual banking taking off in a big
way.
Question 5
What's the situation with electronic money and digital cash?
Answer
The big story in payment systems is getting a standard, because there's no point having a
payment system if everybody doesn't use it. So we've been working with Mastercard, VISA,
American Express and Microsoft on creating a standard that we call SET, for secure
electronic transactions. This is a major standard for electronic payment over the
Internet. We have the software for it and we're going live with the Danish banking system
before Christmas. An aspect of that is microcash or electronic wallets. So an aspect of
that standard is called SET-Wallet, and we're developing that code as well, and we'll have
that in the June timeframe. And we believe that because of the involvement of the major
players like Mastercard, VISA and American Express, that SET will become the major
standard in the payment infrastructure.
Question 6
One of the biggest problems in the development of the net is the gap between north and
south. Now, what are the possibilities of the south of the world catching up?
Answer
That's a really interesting question because typically when one asks that question, one
thinks of an answer in terms of how soon will the south begin to do the kinds of things
that we do on the Internet. But actually these technologies can be leveraged in some very
interesting ways in the south, which are very different from the way we would leverage
them. In South Africa, you have the European part of the population of South Africa, which
is very close to Europe, and then you've got what is commonly called the Fourth World or
the "unbanked". There's a big social problem in South Africa of how you bring
that group of people into the infrastructure, into the payment system. And something that
we've found that's very interesting in that particular situation is that leveraging
network computing, leveraging Internet technologies, creating kiosks with very easy to use
interfaces and smartcards is a much easier way of bringing that Fourth World into the
banking system than sending them into a bank. They don't want to go into a bank. They're
not comfortable with the whole banking situation. But with an entertaining kiosk and a
smartcard, they're very excited and they enter this whole infrastructure.
Question 7
What about the rest of the population? South Africa has a white population and the rest of
Africa and some other parts of the world seem very far behind.
Answer
There's a part of the business world, which is really the European companies operating in
Africa, and then there are the African companies that are operating in Africa and want to
operate elsewhere as well. Certainly, for the European companies operating in Africa they
see this as a very important way of staying in touch, of keeping contact, of building the
virtual company. This is a critical part for them. For the African company with the global
economy, their focus is on moving out and getting markets outside of Africa. And again
it's network computing and the Internet and technologies like this that help drive that
and help them expand into the rest of the world at very low channel costs. But just as
important as the business side is the social side - education, healthcare, giving people
in sub-Saharan Africa access to information, access to medical care and access to a level
of education that they otherwise could not get.
Question 8
How can they do that if they don't even have a telephone?
Answer
Absolutely. The question is what is the infrastructure that you put in place? It's a very
similar question in South Africa as well. The infrastructure is unlikely to be a wired
infrastructure. A big problem you have in all these places is how to get power to these
people who don't have electricity or telephones. There's a lot of work going on right now
on how to build solar powered stations and how to uplink and downlink on satellite. You
would not be on-line on the Internet, but you will download information from the
satellite. It would be solar-powered and you will access to that information. And every
night you can download more information from the satellite. That's the model we're looking
at.
Question 9
The example you made about banking in South Africa was quite an interesting one. Is there
a way to use the Internet is used to solve their problems, which are different from ours?
Answer
In many cultures around the world you have a great tradition of oral history and history
carried down through dance, not through the written word. The big problem we've found
isn't captured in that, even for the culture itself. The culture itself has a problem
capturing this information. And this is where information technology can play a large role
in helping capture this information and helping spread it and helping make other people
aware of it. So there is a very strong role for technology within the culture itself, not
just bringing them in to our cultures.
Question 10
The subject of oral culture brings us quite naturally to voice recognition. Can you
explain what the situation is? Which are the problems and how can they be solved?
Answer
There are many problems in voice recognition. You should think of it in two different ways
or in two different parts. You have the small vocabulary voice recognition, which is you
interacting with a closed system like a banking system or a travel reservation system. So
it's a small vocabulary and you know what kinds of questions people will ask. The other
side is the large vocabulary, where, say, you're dealing with the legal profession or the
medical professional with a very large vocabulary and is very open in terms of what people
will say and how they will deal with it. Then you have to think about it in terms of
continuous speech. Can you speak continuously the way I'm speaking now or is it discreet
speech. If you look at the current state of the art, for small vocabulary, you can do
continuous speech and you can do that today. For large vocabulary, you've got to do
discrete speech. That is going to change. If we go forward to the next five years, we're
going to see continuous speech for large vocabularies. And it's the combination of the two
things that's going to solve the problem. There are algorithms, there are better
mathematical ways of addressing the problem and more powerful computers to learn these
mathematical solutions. Those two are coming together and we're going to see over the next
five to ten years continuous speech for large vocabulary.
Question 11
Is part of the solution to the problem the building of a friendly interface or is there
something beyond?
Answer
I think speaking to a computer is only one aspect. Handwriting recognition is another
aspect, but in both cases you have taken one form of expression and converted into
computer text. You still have to understand the text. Because if you don't understand the
text, it is useless. So the other part that's very important is the notion of intelligent
agents and artificial intelligence that will take the text and actually convert it into
something meaningful that you then want the computer to do. So that's the combination that
you need.
Question 12
What about intelligent agents? What's the state of the art and where will we progress ?
Answer
Intelligent agents is a very broad area. If you start with something as simple as filters,
where you give some key words and the agent goes away and brings back material that
matches the key words. They get more complicated in terms of they watch what you do with a
keyboard and they build a model of what kinds of things you like. And they begin to use
that model to get more information for you. Then as they get even more complicated, they
begin to understand not just your behaviour model, but what kinds of problems you solve
and then they begin to try and help you solve those problems. Where are we today? We
certainly have the key word examples. We have several examples of the model style, the
middle part that I talked about and they are growing, and we are still several years away
from the real personal assistant who sits next to you and understands you and wants to
help you. That's still far in the future.
Question 13
When Internet started a lot of people said it was a wonderful digital library, you could
speak with people, you could find information, and now it's turned into a supermarket and
into a business. What is your point of view, since you come especially from the commercial
side?
Answer
When the Internet started in 1963 it was really seen as a way of connecting some of the
major universities that were working on defence research projects. From that it expanded
into a network for connecting universities for educational purposes. And for many years it
was really an educational, research scientific network. And then over the last few years
it's had a tremendous amount of boost in the business and the commerce side. And all in
all I think that's excellent, because it is the business side, the commerce side that is
going to drive the level of investment we need in order to create a robust world-wide
infrastructure. And in turn, it is the robust world-wide infrastructure that can then be
leveraged, again, back into education, back into science, back into research, back into
healthcare. So the social benefits we're talking about in the future would not have
happened without this investment. And the investment would not have happened without the
commercial interest and the business interest in the Internet.
|
