In an in-depth interview with BetaNews, the user experience chief of Mozilla Labs discusses a unique journey of discovery, backtracking through the groundbreaking work of his own father to rediscover the power of the command line.
He is the son of the man said to be the father of the Macintosh. As Aza Raskin told BetaNews, to this day, he still comes across papers or lecture notes or sketches that introduce him a little more to Jef Raskin, a man who passed away way too soon. And he continues his father's legacy, working now with Mozilla Labs to generate more sensible ways for people to use computers. A large part of his job consists of evangelizing a community of developers and users, generating interest in a new project called Ubiquity -- an experiment in endowing Firefox with a versatile, interpreter-driven command line.
In a 1986 interview for the book Programmers at Work, Jef Raskin said he had just started a company called Information Appliance. His goal, he said, was to advocate the ideal of building a computer that was as simple and as intuitive as a refrigerator. Here's an excerpt from Jef Raskin's interview with author Susan Lammers:
Have you ever noticed that there are no Maytag user groups? Nobody needs a mutual support group to run a washing machine. You just put the clothes in, punch the button, and they get clean. To do information processing, I don't want hardware and software; what I really want is an appliance to do my tasks. And what tasks do I do? Surveys show that 85% of all personal computer users use word processing, so I need a word processor, a wonderful word processor -- the best in the world. But I am sort of simple-minded. I can only remember ten or fifteen commands. That's why the system I have has only five. That way I can wake up at 3:00 a.m., get out of bed, go over to the computer, and just type out an idea.
Let's consider what it would be like if a computer company had designed a toaster. You wake up and you want a piece of toast for breakfast. The first thing you do is switch the toaster on. If it was designed by General Electric, you'd put the toast in and off you'd go, but no, this toaster was designed by a computer company. So what happens? First of all, it does a two-minute toaster check. Then you put in the system disk and boot the system. After that you put in your breakfast disk and then you type "Load TOASTED.CODE."
So, what happens next? Up comes the menu. It asks, "What kind of bread are you going to have?" If it is a California program, it'll say croissant, bagel, English muffin, whole wheat, and at the bottom, of course, white bread. They're labeled A, B, C, D, E, so you hit C because you feel like a muffin this morning. Nothing happens because you forgot to hit Return. You'd think the machine would be smart enough to respond to C, but you have to hit Return anyway.
Do you think it does anything now?...Of course not. It's designed by a computer company. It says, "Are you sure?" Now you're ready to throw it through the wall. Are you mad yet? Haven't you been mad at computers for years? But because you spend a couple of thousand dollars on it, you put up with all this stupidity, and so does the rest of the world. Millions of people go through nonsense like this every time they use a computer.
Twenty-two years later, we began our interview with Aza Raskin by asking him whether it had occurred to him that the work he was doing -- making a Web browser respond to commands -- could be interpreted as running against the grain of the visually splendid graphical environment his father pioneered.
AZA RASKIN, user experience chief, Mozilla Labs: That's an interesting way of phrasing it. I'd put it slightly differently: What we're exploring here, and what our community is exploring here, is not something which is terribly new.
Way back when we had the command line, the problem was, it was made by overly parsimonious engineers who were very interested in making things as efficient key-wise as possible. Which ended up giving us names like df and ps -aux [in UNIX], which just aren't very memorable to the average person, including me. [In my experience,] the command line options to TAR are about as much fun as bobbing for apples in a cement mixer. It's a singularly unpleasant experience.
In that headlong rush away from being able to type what you want to do, we lost something. We lost some of the expressiveness that only comes out of language. So moving back there, I think, is the right direction. But I should note that there's sort of a groundswell of people trying to do similar things. If you look at the URL bar as, in fact, a way of typing what you want to do...you used to type rc to go to your news and pine to go to your mail. Now you type nytimes.com to go to your news and gmail to go to your mail. And that's a command line that your grandmother knows how to use. If you look at [Microsoft] Word, and you do a command, "Print," there is a little box in there where you can type what pages you want printed: "1, 3-10." That's a way of typing what you want to do.
Typing what you want to do isn't at all nerdy, right? In fact, Google does this very well, where you type in a request for a movie and it'll return movies. If you type in weather, it'll return weather. Conversation is an incredibly natural way to communicate. So I think the melding of GUIs -- which are great for discoverability but bad for extensibility -- with language, which has an amazing power of expressiveness, is the right way to be moving the Web.
Aza declines the characterization of himself as the "creator" of Ubiquity. Rather, he prefers to be described as the fellow who inspires others to do the hard work of making the code actually happen.
AZA RASKIN: Most companies [in making their] appeal will say, "Hey, we have a community that helps us, or works with us." But Mozilla's fundamentally different. We don't have a community; we actually are part of the community, and it's a very shared space. So there's actually no distinction between people that sit next to us, and people that sit on the other side of the world. In fact, if you look at Ubiquity, it's created from people in other countries, states, and continents...And you can see the proof of this. It's not just some empty talk in our commit log. People all over the world, 24/7, way before it was launched, were having a strong voice in making it. So the story is not, I think, Raskin comes in and takes these ideas; it's really a funneling, getting people to think about the problem the right way, and then it really harnesses this amazing groundswell of innovation from the edges, comes pouring in.
SCOTT FULTON, BetaNews: Well, when you say, thinking about the problem in the right way, is "the right way" in this case, how do we then extend the functionality of the browser using language? Or is "the right way" really more of a comprehensive question, "How do we reincorporate the expressiveness of language into the use and control of our computers?"
AZA RASKIN: That's a very sagacious question. There are multiple levels here. You could look at all of them. I think the first level we're looking at, yes, is saying, how do you incorporate language into the browser? And if you go a level deeper than that, it's saying, how do you empower everyday Web developers to enhance the browser at a fundamental level? How do we take that community of thousands who are making Firefox extensions, and turn that into millions?
That's part of it; the other part of it is to say, how do you solve some of the problems on the Web, which is that everything is dispararate, adding a map takes a lot of clicks...being able to plot things is very difficult to do. You do see data; the best user experience is to have that data displayed in a way that makes sense to you, not necessarily the way that made sense for the purpose of the author of the Web site. So I think we're going to see sort of a shift in thinking: Web 1.0 was all about making static pages that people went to; Web 2.0 is thinking about, how do you empower content providers to mash up data that then get shown to end users; and, dare I say it, Web 3.0 is about letting end users take whatever data they see, anywhere they want, take any service, and bring it to that. So it's gone from very solid, [where] the Web site owns everything, to, the Web site provides functionality for other Web sites, to everything being very user-centric.
The final question you asked, which is, "Is this a way about thinking more generally about computing?" I think if you fast-forward a number of years, that is an interesting thing to ponder.
SCOTT FULTON: If you fast-forward. But will we still be pondering it in those years?
AZA RASKIN: Well, I don't think so. Again, this is the strength of being at Mozilla, and the strength of Labs and our community: that we as Mozilla, as an open source project and as a meritocracy in our community, have a shared resource here. That puts our community in the very unique position to think about this, being unfettered from constraints like thinking of a business model. We can think entirely about the user experience of the Web, and the Web doesn't necessarily mean, "inside the Web browser." The Web means, access to the Web anywhere, at all times. It's that open Web with a capital "W" as opposed to the lower-case web of what you see through your browser. Engendering that innovation from the edges; plus that community aspect means, I think, Ubiquity and Firefox is in a unique position to actually make this happen not too far in the future; and I think the place you'll first see this happening is on mobile, where it's really painful right now to switch between tabs and context, and you really want to just answer your bar trivia questions.
Next: The danger of getting "buttoned out"...
Continued. . . 1 | 2 | Next >>
