Interview with Aza Raskin, Part 2: Making Ubiquity ubiquitous

Continuing BetaNews' interview with the user experience chief of Mozilla Labs, we discuss whether it's possible, even with an estimated quarter-billion Firefox users, to build a semantic functionality network on the strength of volunteerism.

Of the billions of people linked together through the modern Internet, and the tens of thousands of services available to them -- a great many of which are operated by folks who believe they have the capability to serve the Web's principal application -- the number of institutions formed among them that we talk about on a daily basis, that command Web users' everyday attention, and that promise "solutions" to these users' needs, can be counted on two hands.


For a supposedly world-wide web, this is a glaring disparity. Though so much of what has been accomplished in the evolution of modern programming is now done by a group referred to as "the open source community," it is a nebulous entity of multiple factions, any one of which may be capable of fostering truly inventive and groundbreaking ideas. But put them in a room together, and you get a lot of crosstalk.

The Mozilla organization tries to be a sort of hybrid -- an outgrowth of the open source community, grafted onto a leadership body that determines its mission objectives, strategies, and tactics. So far, its most phenomenal success has been the Firefox Web browser. But as a provider of functionality, Firefox is but a framework. The challenge remains to build some level of interwoven, ubiquitous functionality that transcends both corporate and cultural boundaries.

So Mozilla Labs' Ubiquity is almost unlike any other project for building a product thus far conceived. It's the organization's project to consider whether the Web can be used to train a browser such as Firefox to respond to natural language commands, with functions the user expects those commands to imply. The project is being presented directly to its users as an idea that they should conceive, not because the users have this collective skill set that, gathered together, enables them to understand the semantic Web in the way none of them would individually. Rather, it's a mission in search of an objective, in hopes that its creators and innovators will be its users.

It is a weird, wild ride that Aza Raskin has launched. We pick up on where we left off in Part 1 last week, by asking Aza something we'd hoped he'd considered fairly thoroughly: In a world where everyone out there is equally worthy of creating a service that defines how a simple, single-verb function such as "Map" or "Open" or "Choose" or "Evaluate," works...who gets to decide whose definition is best?

SCOTT FULTON, BetaNews: I had a discussion with a Microsoft engineer about 15 years ago, with respect to something they're never going to do from this point forward: enhance the Component Object Model. At that time, the idea was that there would be, in what we now call the Windows System Registry, this lexicon of verbs, and verbs would represent actions and these actions would be things you attach to the context menu. And the exciting thing for this engineer at that time was the idea that, with simplified programming tools, anyone and everyone could generate their own verbs and put them into the context menu. And I said then, "Well, who gets to decide who defines the word 'Open?"' And he said, "It's up to the user to decide who he trusts." And I have a feeling that Ubiquity is embarking along that same precipice: tackling the problem of, when you have multiple contributors who are offering similar or exact functionality, how does the user decide from Web site to Web site, service to service, who he trusts?

AZA RASKIN, user experience chief, Mozilla Labs: That's something Mozilla is in a good place to solve, because we can always think about user experience first and revenue later. That's always the most important thing at the end of the day for us.

To throw your question around a little bit, how do you decide which photo editor to use on the desktop?

SCOTT FULTON: By testing all of them.

AZA RASKIN: And how do you decide which Web mail to use? There's always going to be the case that lots of different people are making similar functionality, and there's always going to be a marketplace, whether it be for actual money or [for the honor of recognition].

One person has started a great discussion in the community about how to do this: How do you take a verb, or a command like "Open," and let multiple people extend that one command, even if they're disparate vendors? It's not an easy problem, but I think it's one that we're going to have to take.

It's another interesting thing that we're trying to explore here with Ubiquity: How do you answer that question of trust, and decide what commands you're going to install? One of the things that Labs has made is something called Ubiquity Herd, which is based off of a project called Herdict, from the Bergman Center for Internet Society. That's all about taking aggregate information about what people are doing, hooking into social networks -- or at least, social graphs -- so that I can say, "I really trust your opinion, and I trust Brendan Eich's opinion, but probably what you guys do will inform what I should do." I think there's a lot of room for really using the power of the Web to empower the social aspect of choice. That's part of the Mozilla mandate, in fact.

SCOTT FULTON: So if we apply this "herding" model to a trust pattern for Ubiquity commands, then, there may be people or authorities who have become respected as authorities who may then be the initial guinea pigs for a new command, and over time the users in the field may come to trust them by saying, "This guy tested it first and it seems to work, thus..." So who determines the order of things? Is it determined by guys like...I don't know, BetaNews, who would be the first to test something? Or is it determined by a vote of the people at large?

AZA RASKIN: If you look at, how did we decide [the order of providers] in the search box for Firefox, there's an order and there's a default. How is that decided? What we didn't do was say, "This is going to be the order, we've worked out a whole bunch of business deals, now, this is going to be the way it is." Instead, there was a rough consensus that Google was the right search engine that got listed first, and then I think Yahoo second...Mozilla isn't really voting as much as there is rough consensus. So I think there's going to be the same way; there are going to be people who become mavens due to meritocracy, so it might not be the people who are first-come, it's going to be the people who provide the most value. Like the Bruce Schneiers of the security world. Those are going to be the people who end up influencing the decisions in the right direction.

I really think that Firefox as a project had solved most of the difficult problems for these sorts of questions moving forward and Ubiquity, because Firefox as a project and Mozilla as a project has figured out how one interfaces [with], and is, a community.

SCOTT FULTON: You talk about the open source community as being the biggest laboratory in the world, but would you advise any members among that community to pay attention to, or research, the history of instances in which the types of questions and problems that Ubiquity is trying to solve, have been asked before? For instance, the Component Object Model project, or maybe CORBA or AppleScript.


SCOTT FULTON: Or PowerShell, or any other instance of trying to solve problems on a command line with commands that people can readily learn?

AZA RASKIN: Sure, I would definitely encourage that. One of the really cool things is, we start to see participation from the [engineering and research] departments around the world, as part of Labs. There are people from Stanford, University of Toronto -- which has Bill Buxton -- who are very tightly involved here, seeing this call to come and participate in the shared exploration of the future of the Web to inform a lot of these decisions.

Next: When does anyone decide upon a "release date?"

Comments are closed.

© 1998-2021 BetaNews, Inc. All Rights Reserved. Privacy Policy - Cookie Policy.