Jakob Nielsen on how Apple blew it, how Linux will blow it, and the Next Big Thing
And no Dancing Bears
On life after WAP...
Wireless content is still the same kind of gold rush as the web, and people aren't thinking whether it makes sense to sell X [via wireless data] just as it didn't make sense to sell X, such as pet food, on the Web. I still believe in e-commerce, but not for everything.
For example banking makes much more sense to do when you're sitting at your desk with a connection and your paperwork just there. Why do it when you're sitting in a taxi on the way to a meeting?
Even if WAP had a better network connection it would have failed. It had a triple whammy of problems: poor content, latencies and content not suitable for the format, as well as a bad UI.
It could work for things where you have a very very strong interest in an area - such as reference, or people betting on horse races, or getting football scores: odd things that are very time critical.
The more we can remove the UI, the more smooth the UI can be. For example with GPS the device will know where it is, so you can have Call-Me-A-Cab applications, which should be one click.
On SMS: good UI or evil UI?...
SMS is a good UI. [Sound f/x: incredulous reporter gasping] Sure the text entry is awkwardly designed, and the notion that you send a message to a phone number not to a person: I would agree with all of that.
But the actual feature set is really simple. You send a message, and you receive a message.
It is such a compelling idea - a wrong but attractive idea. When you talk to so many Internet marketing managers they say "I want a bigger block of users brainpower: I want x plus five, not x of a users brain."
But in trying to enlarge the footprint, they add less value. They add services and dilute the mission. They say, "I am going to be the Internet for you" but a weather forecast is irrelevant.
It doesn't just understand one of the fundamental of that what the web is incredibly good at, narrowcasting.
On web services...
But the web browser is a completely improvished environment! There's only one thing it does well and that is browsing articles. Think of the other things... like monitoring auctions, real time monitoring, the news, email.
The Internet is comprised of many things - discussion, video on demand in the future, music - and it's not really integrated. Email is the lost cause cause right now - it's like it was 20 years ago.
There are integrating trends but because of historical inertia they're not optimal. If you look at Microsoft Office, it's a very poorly done product. Each application, Word or Excel, has a million features, but it's still the individual applications. File Open is different and works differently in each application.
Now Microsoft never sat down and said "let's design Microsoft Office."
It's quite hard to criticize Bill Gates for taking that strategy with the many billions he's making from it. They can basically say either we're going to scrap what we've done for 20 years, and ask 'what does it take to support business professionals in large companies and connected home users?' Or they can say 'why can't we rev it one more time, add twenty new features that we can advertise and it will guarantee us sales'.
It's just like in Hollywood - it's become the Monster From The Lagoon. The software quality is poor and the experience we all have is bad, spending half and hour getting it to print. That's just bad coding.
On how Apple blew it...
It is not just Microsoft's fault. Apple has decided to stop innovating.
Fundamentally they decided to get back to the safe little area around the fireplace, and they're just making little tweaks to what they did 17 years ago, and putting new colour on the machine.
Which is a shame, because they had great projects ten years ago.
There were a lot of different projects; the challenge was how to integrate that into one design. There were object orientated interfaces - removing applications and replacing them with component architectures - and instead of a Finder where you keep drilling down there were 'piles', semi-structured sets of documents. All this was going on in the early 1990s.
Look at the Apple Newton, for example. If they had continued the Newton they could be in front now.
It was so obvious that the Mac UI wouldn't scale, but they stopped and fired their research department and instead take Jobs' old system and port it.
On Linux desktops...
Will Linux desktops innovate? No. I don't think of that as being the solution: because it's open source.
It doesn't lend itself to coming up with new paradigms. The one thing it's very good thing at is designing software for other hackers, for other nerds, really.
That's their skill and that's their strength - there's a thousand nerds to look at it. If something doesn't work it's going to be a debate on the mailing lists and it's going to be fixed.
But that's a bad method for complex decision management or business professionals or this next generation of home users, because that requires a very different project management approach, a clear vision.
They're great programmers and that's very nice, and it generates good stuff for that environment, but it's a little sandbox.
For example they're so proud once they've ported [sic] PowerPoint. But that doesn't give us a new way of doing presentations.
To do that you 've got to follow business people around all day and study them and ask them what they need.
Microsoft did that and finally got a feature I like out of that: where you get a preview of the next slide while you're giving a presentation. Everyone who's every given a talk will tell you that: I have to print an extra copy of the presentation off - even with my 1Ghz computer I have no extra benefit, because I can only see what the audience can see.
But why did it take them years?
On why Jakob has so many software patents...
That was from my work at Sun. Whatever I've invented for my company [founded in 1998] I've not filed patents for.
As long as we have patents, everyone needs to get their own portfolio purely for self defense. A prior art database isn't enough. It doesn't protect things they did get patents for.
But there's a case for saying that the patent period should be shorted for Internet innovation.
On the Next Big Thing
We're at bursting point. Email is truly getting at the breaking point.
Email will soon get to a point where it just doesn't work. If you get more than 200 or 300 emails a day you will get no work done - and email will be your life.
What do you have on a PC? Two things - one is the OS to manage the computer resources, and the UI, and the goal of the UI is to produce hot copy: sassy reports. But it's really office automation, that's the underlying metaphor: big fat printouts to the laser printer.
But we will change those two around. The goal now is to optimize users' time and users not optimize resources, as computer resources are cheap now. The output is a communication.
On being a cult figure...
The T-Shirt part is a little odd. I'm wondering whether that could be a sideline for me, as all the T-Shirts with me on them are unofficial! But for now I'm sticking to books and reports to make money.
Actually it's good because it's a tool to make that fight for simplicity happen. It adds spice. In the struggle between complexity and simple, complexity has a lot on its side: they can show the dancing bear and I say keep simple, so I need something on my side.
[The Jakob roadshow is touring Europe and the East Coast in the next month, with Nielsen joined by colleagues Don Norman and Tog of Apple fame. Details here]®
Sponsored: RAID: End of an era?