You made some interesting comments about software and apps replacing humans in many industries. While that may solve some issues it will create others such as huge numbers of unemployed people, won’t it?
I don’t deny that. It’s a problem but we’re trying to repurpose human capital. If I want to have people that are doctors and lawyers they have to be retrained from people who worked in fields picking tobacco, and that’s a problem. That’s why it’s legitimate to have a government and taxes and things subsidised.
I’m a big believer in free education, that’s one solution to the problem. If you add up all the people in the UK who teach maths, it’ll be a lot of people. If we find a way to teach all that with an iPad teachers could instead spend four years learning to write software. You can pay them to teach maths for fours years, but what if you can do it for free? The point is you can teach maths for free. Once you give the kid an iPad you don’t need to pay for the library or the teacher or whatever. So why not take that existing budget and spend it on teaching teachers to write code, or write book or make films or study nursing?
So if it’s a 10 year plan you could say that over that time you get to the point where all these services are free and automated but all the people who have been displaced get a two year grant to go learn something new. I think that’s rational and enlightened. The alternative is not moving forward.
You also said that software always does what it’s supposed to. But while humans are involved in writing it, that’s simply not the case as humans are fallible. Is there ever going to be bug-free software?
The good thing is that software is a competitive space; there are 600,000 programs for the iPhone. They don’t all succeed. If they are buggy they will fail.
Let’s say there are 600,000 physics teachers. They make mistakes but we don’t audit them, we cannot afford to. But what if there were only six programs that teach physics. Then you could audit them to figure out which was the best one.
So then you’ve got market pressure and dynamics where you can afford to assess the quality of the program. You could have 100 professors analyse just the one course. You cannot afford to have 100 professors analyse each professor. In the manual labour market the entire thesis of model control breaks down. But if you can manufacture one million classroom sessions with a single program you can afford to have multiple people doing quality control and over time it’ll probably be pretty damn good. And if one is better than the other people can switch.
Your thoughts on Microsoft’s Surface announcement were also interesting. What are your thoughts on the Microsoft deal with Nokia?
It was also a disaster. Nokia killed Symbian, which started their free fall, and they did it before they had a decent product from Microsoft. They should have gone to Android. So you ended up with a phone company with no confidence in its ability to do software and a software company with no confidence to do their hardware. They joined together but it’s been discredited and I can’t see any company benefiting from it."
You spoke to CBR last year about the release of Emma, Usher and Alert. How has the uptake of those been?
We’re doing pretty well. With the mobile platform we’re now at 1,000 different organisations and with Usher and Alert we’re working on dozens of pilots. I wouldn’t characterise them as being mature, they are early stage. In their current form they’re no older than three months. In a year we’ll see a lot of decent success there.
There is a more in-depth interview with Michael Saylor here.
You can also read more on his thoughts about the Microsoft Surface launch and his vision of the future.
