In an intense discussion about ‘the interface of the future’, John Lilly (then CEO Mozilla Corporation, now working for another company); Aza Raskin (then Head of User Experience at Mozilla Labs, now running his own company) and Dan Mills (then Lead Developer of Weave, now leading the Account Manager project at Mozilla Labs) debate what happens between intent and result. If the interface goes away, how do you learn what to look for?

http://www.youtube.com/watch?v=4gygQ2GgN10

00:01,671 --> 02:15,276
AR: Let’s jump a little bit forward because I think when you know where we are going, it is easier to figure out how we chart our course there. So really simple example, starting with the shovel. A shovel has two bits. It has the thing that does the work, that is the blade. And it has the interface or the handle.
What we often see is that most projects is that people work on that blade, making it titanium, making it diamant encrusted. Spending all their time making an incredible backend technology but forgetting about the handle so you and up with a beautiful blade with a tiny short handle and you can’t really use it.
Let’s go a step further. What would the ultimate interface be, for a shovel. The very best one.
Some people might say you want an automatic digging machine, where you might get to ride it like a tractor trailor, other people might call that bloatware.
But what is the ultimate expression of a shovel? It’s a hole, it is having a hole already dug.
Instead of needing the tool, you would just have the result. So I think if we go way way into the future, the place we cannot get now, when we think of interfaces, what you really want, is just the result already done.
I think that is mighty spaceship thinking, what we can pull from that is the closer we get from no interface, the closer we get to not having to think about the interface between us and the machine and us and the artifact to having just the result that we want, the better.
For instance instead of going to our browser and typing in a search query, to get a search result, we should just be able to think or say and interact … “I wonder whether the [?] won this season?” or “Won the last season”, and you get the answer immediately.
To most people, that is not going to feel like an interface. And that is the sort of interface we want. Because as soon as you need to stop and realise and think about an interface, you’ve already lost.

02:15,276 --> 03:09,056
JL: I think that in digging the hole, you learn much about what you are trying to do. I think you are presupposing an intentionality that is not always there.
So I think when you look for knowledge, when you construct knowledge, when you construct a future space station, you don’t always know the goal from the outset, where you are trying to get.
The process of getting there often creates new knowledge. So I argue at c-base the point is not actually (I’ve known you for thirty minutes, so I don’ know if this is actually right) but the point is not to get to the space station, but it is all the things and all the aha’s and all the insights you have along the way.
And you can only get those insights, by struggling. By working through it. So I think that if you presuppose that you understand what you want always and that you are very intentional about everything you that you do, and that you understand precisely the results that you want, than I agree with you. But I think that human beings most of the times are not like that.

03:18,484 --> 03:25,633
AR: So if I understand what you are saying, than you are saying that imperfect tools are better?

03:25,633 --> 03:48,717
JL: No that is not what I said. I said, that you construct knowledge, through the use of tools. You are articulating one type of interface that is getting to the result without use of tools. And I am saying you miss a type of knowledge creation by thinking that way.

03:48,717 --> 04:03,437
AR: Rather what I think I am saying that if your goal is to dig a trench to add a landline, some place, that there are a whole bunch of problems there that go way beyond the physical act of digging of the hole.

04:03,437 --> 04:12,262
JL: I agree with you. If you know exactly the result you are trying to get to, and you don’t believe that you need to create knowledge along the way, than you are good with no interface.

04:12,262 --> 04:21,456
AR: Really what I am saying that any time you spend fiddling with your tool, instead of doing the job or the task that you set out to do, is wasted time that we should be trying to minimize.

04:21,456 --> 04:23,593
JL: And I think that is an over-broad generalization.

04:23,593 --> 04:26,797
DM: What if you don’t know what the task you are setting to do is?

04:26,797 --> 04:34,877
JL: Or what if what you think you know what you are doing but you are wrong? I think that in many cases you are correct, but I think it is an over-broad statement though.

04:34,877 --> 04:40,728
AR: Well, in that case you are going to use a tool that isn’t particularly apropos; discover that it is wrong and make a new tool.

04:40,728 --> 04:46,254
DM: But in your case, there is no tool. There is only intent and result.

04:46,254 --> 04:56,425
JL: That is right. Intent and result. And there is a process between intent and result. And sometimes the process changes both intent and result.

04:56,425 --> 05:05,295
AR: I think I’d argue rather that if your intent is wrong the result will be wrong, and that is the iterative loop you want.

05:05,482 --> 05:19,507
JL: I don’t think that is obvious either. I don’t think the result is wrong if the intent is wrong either. you often think you are headed in one direction, and you end up in a different place entirely. It is an ok result.

05:19,507 --> 05:25,358
AR: I won’t disagree with you there, I just think it is not our tools that should be…

05:25,358 --> 06:24,057
JL: I can be specific. I think right now on the web, we are heading to this non-serendipitous web. What people really want, they want the news they want, just the news. They just want what they are interested in, what their filter is, what their lens is.
But I will tell you that it misses serendipity. When you read through a paper, that you did not expect, but that you wanted to learn and needed to learn. When you walk through Berlin, you notice up a whole bunch of things that you did not think you needed to know, but now you do.
I think that serendipity and the process of knowledge construction… I think that we are talking about different types of activities, but I think that … I’ll tell you that I have learned a lot about myself when I’ve been digging in my yard with a shovel.
It is not that I say we should look for imperfect tools, but I think that connecting intent and result too tightly will not always be the right thing.

This discussion took place in C-base, Berlin in October 2008