Some questions for Google Earth

Creating a low-end Mac market — with Linux

[photo]

Dave Liu of gOS

Needless to say, this and other remarks along the same lines did not please Microsoft. A great deal of history followed, including the "browser wars", the sale of Netscape to AOL, the federal lawsuit against Microsoft, the dot-com crash, Y2K and much more. Forgotten in the shuffle was Marc's original ambition, which was to establish the browser as a platform and a development environment, and in the process to commoditize operating systems to what they had long been called in any case: "bags of device drivers".

Now it's 2008 and Google is busy treating the browser as a platform, and is generally agnostic toward operating systems. (Its own services are mostly deployed on Linux-based systems, but its applications are either browser-based or applications that run on multiple platforms. (Google Earth is the ideal example. Picassa is not.)

But the browser is mostly where Google likes to run user-side apps as Web services. In fact Google now provides most or all of your basic desktop application suite -- mail, office (documents, spreadsheets, presentations), calendar and instant messaging -- inside your browser. It's up to the user which bag of device drivers runs between browser and iron. May the best bag win.

Thus it was perhaps inevitable that somebody would come along and make a bare-bones -- or bare-browser -- box that's optimized to run Google's browser-based apps, fulfilling The Andreessen Prophesy.

That somebody is Dave Liu, the 21 year old CEO of Good OS LLC, maker of gOS, an Ubuntu-based distro tweaked to run Web apps as if they were desktop ones. gOS might have been Yet Another Linux Distro had it made news last November when the $199 Everex TC2502 Green gPC, running gOS, sold out at Wal-Mart. We caught up with Dave in the midst of the buzz that followed. Here's the interview.

DL: Most of my studies at UCLA and my work outside of the school have been centered on Web 2.0. I'm actually fairly new to open source. I saw a lot of great Web 2.0 applications that weren't taken seriously because they just didn't have a platform where people could reach out and get to know them. You look at TechCrunch, Mashable and these Web 2.0 blogs and there's a very small community that actually talks about and uses these applications.

When I met the Enlightenment guys, doing open source, I realized that if you were to put all these Web 2.0 applications on one serious platform -- like a Linux one -- you could show the world that Web 2.0 is a viable way, and perhaps the next way, to use a computer. So that was the main idea. It wasn't so much to create another distribution. We're resting on the shoulders of giants here. Not just Linux but Ubuntu and their communities, and Enlightenment and their community. In fact, the Enlightenment team is our team here now. We have seven core members of the Enlightenment community. They're our whole development staff.

DL: I'd see our role as something like Kubuntu's has been. Derivative, but taking Linux in some new directions.

DL: There really is a growing subset of people in the open source community who care about consumers. People that want to make the Linux experience palatable for average Joes. And that's a big shift. And that's who we're appealing to. We felt that Linux as a native OS project is a great platform, but still hasn't gone that final step to really connect it to consumers and to differentiate it from everyone else in the consumer space. To do that you really have to bring a consumer aspect, and we saw that in Web 2.0 -- in Google apps and YouTube.

LJ: When you talk about Web 2.0 apps, you mean ones that work in a browser as a kind of Web service?

DL: Pretty much. We're basically talking about software that runs in a browser and is based on Linux. It's a paradigm shift away from the way you would typically use software. Instead of compiling and selling it, and having it run in the system, you're running everything in the browser. And there's a lot of Web 2.0 software out there that hasn't gotten into the eyes of the public.

DL: Even with the Google Docs -- spreadsheet, calendar -- we were surprised at how people either had never heard of these things or had never tried them. We were among the first to put all these things into one coherent package. So people could realize, going from one icon to another, that Google and Web 2.0 really are their computer. Not only that, but by computing in the cloud, users really are able to take their computer with them without taking their computer. As long as they can log on at a cafe or a friend's house, they have a computer of their own.

LJ: You don't generally think of a desktops as being something that lives in a browser. Are you abstracting the apps and their icons out of the browser and putting them on the desktop?

DL: Yes. That's the idea. Users are accustomed to seeing applications in a dock or in a start menu. We want to stick with the easiest ad most familiar models for desktop computing, even if the programs are executed elsewhere. So yes, we're using a lot of Firefox shortcuts. From a tech view, its a lot of browser short-cuts on a dock. All the main ones are in a dock, and a few more are in the start menu.

LJ: How about for documents you want to keep on your own machine to work on when you're off-line? Calendars, for example.

DL: There is still a gap. We're seeing a lot of big companies investing in Web 2.0 apps. Google is the first big one. Even Microsoft is doing it with Office Live. There are still a lot things that people prefer on the offline side, though. So we've packaged in a lot of things there: Open Office 2.2, Thunderbird and Gimp, for example.

LJ: Or a presentation that somebody might create online then save off to give offline on their machine or transported by thumb drive to another machine.

Yes. One of the great things is, in the past few years, Google has done a good job of bridging with the community. So they just packaged Open Office into the Google Pack. Google Gears too. So I think there's a good chance in the future of seeing better and better integration between the on-line and offline models.

Yes. I should make clear that gOS is not the Google Operating System, though I kind of wish it were. Earlier on in the project, before we were sure we were going to work with a hardware partner, my team and I applied to an obscure trademark usage forum on the Google website. I said "Hey, we're an open source group. We want to make an operating system that makes it extremely easy to use Google Apps, and want to use that trademark. So in two or three weeks we got a letter back from them and they said "Yeah, go ahead, as long as you have a disclaimer saying this is not a Google product..." So from the very beginning we had that kind of green light. We didn't think it was going to be a big spotlight project. But later on when we started working with Everex, we found that they had their own standard toolbar deal with Google. And they showed Google a sneak peak of our screen shots. So in that sense there was a standard agreement on both sides. Nothing more than that. We keep in touch with Google, though, on both the gPC and the gOS.

The g name stands for good. We set out to make good products. For example, we know that Microsoft is a tough beast to work with. So we set out right from the beginning to make ours consumer friendly, hardware-friendly, with all the values of community, Web 2.0, open source, and Google, all wrapped up into one.

Right. Walmart.com sold out in the first two days. But it came back up pretty soon after that, in plenty of time for the holidays.

It's all those things. A simple alternative. Something that works out of the box, online, with the help of the cloud. Yes, we've had a lot of people tell us they're excited about it because they want to buy it for their moms or their dads, who don't need too much power and want something simple.

LJ: But it's still a Linux box, so the geek who's giving the gift can still ssh into it and help out if need be, right?

It's definitely hackable. It's also good for someone who knows how to do customizing in general. Or to work as-is. And yes, you can wipe it clean and install Windows if you like, but that's not what it's for, and we'd rather people didn't. But in terms of component, assembly and the software combination, it's a good deal.

The more important thing is where this is headed. We're going to see more companies invested in Web applications. What we're trying to do is fit into that next curve.

LJ: What is your dev community like? Have you grown your own, in addition to the Enlightenment folks?

Our core team is about seven or eight people. We've added one or two in the last month. Once we got in the news, people starting hearing about us, and we have developers coming in from different communities. Some Ubuntu folks are helping out, of course. So we have a decent community going. But it's still early. We're still trying to create a structure so people who want to help can do that. But it's still a controlled chaotic space that we're trying to sort out.

DL: I think that as time progresses, the younger generation of hackers and developers includes more Mac fans. They like that look and feel, that end user experience. What they have there is a consumer mind-set, which they want to bring over to the Linux platform, to make Linux more mainstream. Look at things like Compiz Fusion, Beryl, all that stuff. I had a chance to talk to Quinn Storm, the lead developer on Compiz and Beryl. She thought people in the Linux community could make something that end users could enjoy. After that conversation with her I realized that this was a small but growing community with a lot of promise.

LJ: Well, from an easy-to-use UI perspective, Apple has left the low end open. Do you see Linux making a move there?

DL: Yes, one of the big trends that created this opportunity for Linux was that when Vista came out it killed a lot of low-end hardware companies. On the low low end Vista can't run, or it's a terrible experience. So there's an opportunity at the low end because Microsoft vacated it as well.

DL: Right now we're aiming for the low end hardware, plus appealing to people who are cost-sensitive or want a second or third PC, and could use a PC experience that's all in the cloud. Our main goal is simple: Improve gOS, so average Joe end users can get the full computing experience at the low end right out of the box. So we're doing a of things with the UI, on pre-loading more or different apps, to just optimize the end user feel or experience. I also want to create the same benefits that Mac creates, in the sense that you have a real nice OS in terms of the look and feel, and making things just easy and fun to use — but in a way that more people could buy into because Macs cost a thousand or two thousand dollars per computer. That's on the product side. There are a lot of international opportunities as well. We're also looking to form more partnerships with hardware companies making, say, cameras or printers. That's probably our best step forward right now. We can't guarantee that a computer will work with anything, but we can grow the list.

LJ: How about laptops? Generic desktops are all the same. But laptops are all different, by design, through OEM partnerships with Microsoft. What are your plans there?

DL: One of the things that will make our laptops viable is software that offers seamless syncing online and off. You are going to see gOS on some flavor of laptop very soon, if not by the time you read this.

DL: It's still early and a lot of the relationships we could have aren't detailed out yet. Right now we're working with Everex, a single hardware partner; but the goal is to expand to working with other companies as well. As a software company, we really appreciate Everex and expect we'll be working with then exclusively over the short term. But our main strategy right now is growing our team so we can work with more companies on various fronts.

LJ: I would think that Dell, Lenovo, HP and others would be looking at a Linux offering in the cost range that you're working in, at some point. Does that concern you?

DL: I think Dell was one of the big brands to launch a product, which is good; but among the smaller PC companies, Everex is still one of the top companies. They sell at Best Buy, Circuit City, Wal-Mart, yet are small enough to be motivated to experiment and take chances with a company like ours. The larger hardware companies, such as Dell and HP, have a lot at stake with Vista, and with the Microsoft relationship. Seems to us a company like Everex is less locked down that way. So we see companies like Everex taking the first steps that need to be taken in this market segment.

DL: We talked about Web 2.0 and open source in China. Things are exciting because we're seeing the same kind of Web 2.0 websites launch in china and do well. It's exciting because China is in the very early stages. There are huge opportunities there, especially in relation to Linux.

LJ: People have been waiting for this segment to open up for a long time, and I'm not just talking about the low end of the PC marketplace. I'm talking about the browser as the environment for all kinds of applications. Because this is exactly what Netscape talked about doing way back in the 1995. One of the reasons that Microsoft came after them was because Netscape had the audacity to say the real desktop on the Net will be the browser.

DL: They were way advanced. I think it's going to be real exciting to see things unfold here. A lot of people have been saying Web 2.0 is a bubble but I don't think so.

DL: Yeah, it might be, but the advances are still there. Meanwhile, it takes hardware some time to catch up to software. Hardware companies at some point are going to give Web 2.0 a serious look. They'll see the competition is in the cloud. Then a lot of these start-ups that were laughed at will find themselves front and center for what's next in computing.