Google+ contributor and Mac pioneer talks with CNET (Q&A)
A core feature of the new Google+ service is Circles, which makes it drag-and-drop easy to build "circles" of friends in the browser. Meet Andy Hertzfeld, the man behind Circles--and the original Mac.
Thirty years ago, Andy Hertzfeld was a young computer engineer working at Apple Computer on the first Macintosh under the leadership of Steve Jobs. As Jobs had repeatedly promised the small team, their creation would change the world, and he was right.
Today, Hertzfeld's passion for technology and his experiences at Apple have been baked into a new product that Google unveiled to the public this week. Called Google+, it's a suite of features for helping people communicate across the Web with friends, family, and co-workers. If that sounds like Facebook, it should. Google+ is openly described as Google's latest attempt to break in to the social space in a big way, potentially chipping away at--if not dismantling--Facebook's dominance.
Hertzfeld's role on Google+ was primarily building Circles. Simply put, Circles grabs all the contacts in your Google profile and allows you to drag and drop them into circles labeled friends, family, sports, and so on, similar to taking a batch of songs in iTunes and dragging them into playlists. You can then send updates and photos to one or more circles of contacts.
CNET interviewed Hertzfeld by phone to ask more about the ambitions of Google+, the thinking that went into Cirlces, his next projects at Google, and some broader industry trends.
Q: For those not familiar with your role in the early days of the personal computer, can you summarize your work on the Mac?
Hertzfeld: I was one of the first people on the original Macintosh team that started working on the Mac in February 1981. I first started doing low-level stuff like the BIOS system and all the device drivers, and later developed the user interface toolbox, which was the code for the windows menus, scrollbars, and all that, and left Apple a couple months after the Mac [was released in 1984], so I developed the original thing. After that, I continued to do stuff for the Mac, like Switcher and everything, but that was a very positive formative experience for me.
And you've collected those stories into a book and a Web site. You want to give a plug for those?
Hertzfeld: I'll plug my book, which is called "Revolution in the Valley." Actually, I don't need to plug it anymore because it just recently went out of print, but there's a Web site, folklore.org, where you can read all the stories in the book and even more for free.
After Apple, you went to a few companies: Radius, General Magic, Eazel.
Hertzfeld: I actually helped start three different companies: Radius, which addressed the limitations of the Macintosh in 1986, and then General Magic, where I worked from 1990 to 1996, where you could say we developed the iPhone of the '90s, kind of. We were developing what we called Personal Intelligent Communicators...We correctly identified mobile as the next big thing after PCs; we were just about 10 or 15 years too soon. And then I got very enthusiastic about open source in early 1998, which led me to found a company called Eazel in 1999, to try to help make open-source software easier to use, and so I worked there for a couple years. Unfortunately...Eazel was a creature of the bubble. We just didn't know it. And when the bubble burst in 2001, we couldn't raise our second round of funding, so we had to scuttle the company.
You went to Google in '05. What were some of the projects you worked on at Google prior to Google+?
Hertzfeld: I worked on a smattering of different things, including the Photo Picker...it was in Gmail for many years, but I think it finally got switched out by something newer. But probably the single thing that (was) sort of my project was Google News Timeline, which launched around April 2009, and it's still available. (It) is sort of a zoomable timeline that you can project all kinds of information on.
Which brings us to Google. What exactly is your role?
Hertzfeld: My role's been evolving, but really, the Circle Editor was my baby. I wrote a prototype for it singlehandedly and then sort of led its development through to the present time. I was not just the interface designer but the main implementer as well, writing code.
Circle Editor is the page where you drag the little tiles that represent people into circles. And, of course, I had a lot of help, both with the interface design and the implementation.
More recently, I started three or four months ago helping out with other areas of Emerald Sea (the internal Google code name for Google+), and more recently, in the last month or so, I've taken a broader role, trying to make all of it be as great as I can make it.
One thing I just want to make clear is, I feel a little bad that I've actually gotten too much credit as the designer of Google+...I feel more comfortable saying I'm the designer of the Circle Editor, which...even though I had help from other people, I really was the driving force behind that. I was not, really, for the entire Google+, and I'd just like to make sure that the superb UI team that we have here does get credit for their work.
So you focused on Circles, but now you're going to branch out into what, Sparks and Hangout?
Hertzfeld: Yes. All of it, really. Just helping to refine the UI, improve it, adding important features that people want, etc.
I'll come back to Circles in a second, but can you just give me kind of a high-level view of what G+ is? What was the intent for Google+, and what are you hoping to do with it?
Hertzfeld: Sure. Everything on the Web can be improved by knowledge of your social connections, so Google+ is an effort to...add a social layer to Google, to YouTube, to Google Search, to every Google property. And so the core of Google+ are some APIs and applications that let you organize [and] maintain your identity through a profile and organize the people you know. Really, one of the main contributions we're making compared to the status quo is giving you much finer grain control over your sharing. You know, instead of just having an undifferentiated mass of people, we make it really easy and at the forefront of the UI to send some things to some people, other things to others.
Let's talk about Circles. It works really well, and I think the early returns have been very positive. I think somebody called them "delicious," which is pretty rare for a Google product, especially when talking about an interface rather than some algorithm behind it. Tell me about how Circles came about and how your past experiences at Apple, at Eazel, at General Magic may have played into it.
Hertzfeld: Well, really, at Apple, I learned that the best thing I can do is put a smile on the user's face, and actually, I want to make their jaw drop when they use what I do. I want to astonish and delight people, and I've tried to do that throughout my career with everything I work on--just make the user happy. And even more than happy, get them excited and pleased with what they're doing. And so with Circles, one of the things I was trying to do is take advantage of the new capabilities of the Web browser. I had a feeling of a lot of fertile territory, because the Web browsers pretty recently, in the last year or so, got a whole range of new dynamic capabilities that people weren't exploiting yet. So, as you probably know, the Circle Editor is heavy on animation, with rotations and scales and fades that really were very difficult to do in Web browsers up until 2010 or so, so part of it was just my technical hunger to try out new things. But really, that's driven by the main desire just to put a smile on the user's face and make a vivid, exciting experience for them.
I started [Circle Editor] work in February of 2010, and then by May, they'd conceived of this broader Emerald Sea project that my prototype became one of the central pillars of, and so since May, I wrote the prototype, which was quite different than the Circle Editor that actually shipped. For one thing, at the very beginning, we weren't even calling them circles, which of course influenced the graphic design.
Once we decided to call these groups of people "circles," I tried to make things very circular, but anyway, starting in May or June, I got a bunch of people to help me, and by September, we had something resembling pretty much the Circle Editor you see today. But I spent another six months refining it, adding more animation, trying to make it as delightful as my imagination would allow me to make it.
Google promotional video for Circles.
When I used it, it felt a little like creating playlists in iTunes, where you have this big bucket of content--in this case it's names or contacts--and you create folders and drag names into them to create a circle. Is that a fair metaphor, or was there something else that we maybe have used in the past that you were drawing from?
Hertzfeld: Yeah, kind of. I wasn't thinking so much of playlists, but I know that from my Mac experience, drag and drop has a visceral experience for the user that, given the right feedback, can be just very satisfying, and so that's really what drove me. And in the course of developing it, I solved some problems that really the original Mac had 30 years ago involving dragging with multiple selections...when you drag with things multiply selected, the things are spread out, not near the cursor, and it becomes awkward [to figure out] what goes where when you're only dropping one of them, really, over the drop target; the rest of them are spread out. So I came up with this gather animation when, as you start to drag, while they're dragging, they slide together into a cluster. And we even put a little paper clip--and prior to that we had a little rubber band; we changed it to a paper clip to get them together--but since they're all together, it becomes very clear they're all going into the drop target. That's just an example of...how we improved the state of the art that was originally established with work I did 30 years ago.
Once it's solved, it seems so simple and obvious.
Hertzfeld: That's right. I've never seen anyone else do something like that.
What's been the reaction from the executive team there?
Hertzfeld: I've consistently gotten pretty positive feedback every step of the way, but also, they sort of beat me up with things they didn't like. Both Larry (Page) and Sergey (Brin), I think, even today are unhappy with the capacity. Right now, because of performance reasons, we can't handle millions of people in your circle, but I think they would like to have no limitations at all.
You said you're going to move on to working on some of the other aspects of Emerald Sea. Is that including Hangouts, the video chat feature? At CNET, we think Circles is very cool, but Hangouts looks like it could be a real killer app down the road, as far as finally bringing the promise of videoconferencing into reality.
Hertzfeld: I agree. It's just great that all you need is a Web browser, no configuration, no planning, and you can interact in a very natural way.
What type of things do you think you'll be fixing on that? Right now, we love how the speaker's head kinda floats to the larger part of the screen and back. Is there anything else that we can see coming on that?
Hertzfeld: I can't talk too much about future plans, but I think the experience can be refined a little. But it's already very good, I think.
So let me ask you about some general industry trends. You've been around, and you once told me that software development's really been lagging behind hardware development. Do you still think that's true, given what's happened with Facebook, Twitter, Google, iTunes, and app stores in general? Is software getting back into the driver's seat, or has it already been there for a while?
Hertzfeld: I think one of the big things that happened since I talked to you in 2004 is, the hardware industry hit a little bit of a thermal wall. Up until right around 2004, you saw the processor clocks increasing by 50 percent per year--something like that--and that's had to stop, mainly because of thermal considerations...and then mobile created a situation where it's not just how fast things are, it's how fast they are per milliwatt. It doesn't do any good for it to be really fast, if the battery only lasts an hour or two. So I think you've seen the relentless pace of Moore's Law slow down a bit. And then one place where a hardware problem becomes a software problem--this is getting a little technical, but because of the thermal considerations they can still cache more silicon on the chip so they went multicore--but I think we've only scratched the surface of being able to use the multiprocessing capabilities, which are now even in the mobile devices. But to the broader point, software's always been in the driver's seat because the hardware is useless without the software. Software is what delivers the potential of the hardware to the user, and that's even more so these days.
Another factor that I probably mentioned the last time we talked that's changed a lot is, Microsoft, for a long time, was able to hold back innovation in the industry because they were in such a powerful position, and the innovation was not in their interest, because they were afraid they'd lose their iron grip on things. The good news is, in the last six years, they have lost their grip, and that's one of the reasons you're seeing so much more innovation. Another thing that really helps is, Moore's Law has driven down the raw cost of the hardware such that it's not a barrier. College kids can start services that scale to dozens of millions or even hundreds of millions of people, and so there was sort of a financing bottleneck to innovating that's also disappeared in the last few years or so, and that's also leading to the wave of innovation we're experiencing today.
You mentioned Microsoft losing its grip, and you're a big proponent of open source--obviously, you mentioned your company Eazel. How does that play out with Google?
Hertzfeld: Google is probably the largest user of open-source software in the world. I don't think Google and its approach would be possible if Google had to buy software licenses for each of its innumerable machines. Google also gives a lot back to open source; sort of the default at Google is to try to make something open source--if possible, like Chrome, like Android--and so open source is one of the things that really enables Google to be Google.
One last question here: I want to read a quote of yours from 2005 and to reflect on where we are now. You said, "The ubiquitous connectivity profoundly influences how we use our computers. We're 10 years down the road--we're just in the middle of the transition. Essentially, the hegemony of the PC is over. Now the center of every user's world will be in a network repository projected into many different devices. How those ecologies interact and work out, that's the story of the next 5, 10 years."
Hertzfeld: Like usual, I was a little optimistic on the time frame, but I think I'll stand by that quote; I think it's prescient. It sounds quite a bit like what Steve Jobs said at his recent event, where the PC is now deprecated to being just another device instead of the center of your world. The center of your world, from Steve's point of view, is now iCloud, and so my quote, even though it was six years prior, says basically the same thing, because back then, it was pretty obvious that that was going to happen. That was really before mobile took off to the extent that it would, but I knew that for decades--that [it would happen] eventually, I mean, if you just extrapolate the trends of computers getting more and more personal. Of course, you have one with you all the time in your pocket, so all of that now, I would say the mobile revolution is in full swing, although there'll be many more form factors in five years than there even are today. It seemed like really it was the iPhone, announced in January 2007, that I think people, when they view back 10 or 20 years from now, will see as the turning point when the old world cracked open--and so now it has cracked open, and we're trying to harvest all the benefits from that.
Thanks, Andy. It's great to talk to you. You sound excited, as I imagine you were back in the early '80s.
Hertzfeld: I think it really is an exciting time in the industry, you know, with mobile, social, local--we can do more for the users than ever before.