Dialing up the next transition

When it comes to having a vision, Jeff Hawkins has an impeccable résumé. A co-founder of both Palm and Handspring, he helped establish two industry giants that have helped transform how most people think about information organization.

Hawkins has now changed his vision for portable computing devices. The conventional computing order, he believes, is about to get flipped--desktops and notebooks will become accessories to combo cell phone organizers.

What's more, Hawkins no longer expects fast growth from the handheld market, which he thinks is destined to become a commodity business.

That's why Hawkins has actively redirected Handspring's bread-and-butter business away from handhelds toward what he calls "communicators" that combine the capabilities of cell phones and organizing devices.

These sorts of transitions, where a company shifts focus midstream, are tricky to successfully navigate. But Hawkins says the Mountain View, Calif.-based company can and should change its course before events force its hand.

The move may prove discomforting to fans of Handspring's Visor organizers, because the company has stopped actively supporting some of the technological innovations it helped popularize, such as its Springboard expansion slots and its Graffiti handwriting recognition in favor of keyboards. Hawkins is nonetheless confident that Handspring can hold the attention of enthusiasts while it figures out the best way to navigate the carrier-centric world of the cell phone industry.

Hawkins recently spoke with CNET News.com, teasing us with news that he has been using an unannounced Handspring Treo product for weeks, and giving us his thoughts on emerging technologies and applications for converging handheld and cellular devices. Might these prove to be hints about what to expect in upcoming Handspring devices?

Q: Knowing what you know now, what would you have done differently in transitioning from a handheld company to a communicator company?
A: I anticipated that the transition would be very hard, but that doesn't make it any more pleasurable. This is a very disruptive thing for a company to do, and we looked at analogies before we went into it. I looked at Intel early on, and I remember when they went from a memory business to a microprocessor business. It was terribly disruptive. Their cash cow basically disappeared, and their new business hadn't taken off yet--that's kind of where we are. So there is precedence for doing this.

Nothing yet has convinced me that getting out of the organizer business and getting into the communicator business was not the right thing to do. I think that the organizer business is going to get very uninteresting. It hasn't been growing. I think it is going to be a commodity business.

I could always go back and look at the tactical decisions that I made that didn't work out, but I still believe that this is the future of personal computing.

I'm not a technology guy. I understand technologies, but I don't really get excited about technologies. I get excited about applications?It's all about applications and functionality.

Communicator-like devices will become the center of personal computing, while desktops and laptops will become more like workstations and will be considered accessories to these devices.

From a business and product standpoint, what will change?
In the past, when we built an organizer, we had a completely different set of channels and customers. We designed the product for the end user and distributed it through general retail or through other channels. A communicator product like the Treo is designed for the user, because (a user has) to like it, and for the carrier, because they have to approve it. If the carriers don't like the product for whatever reason--if it doesn't meet their business needs or doesn't tie into their infrastructure--they might just decide to not carry it.

From a product development point of view, it actually takes longer to launch a cell phone product, because once we're done making it, we have to give it to carriers who put it through their own product-testing series. That testing time, which is considered to be part of the development process, can vary in length.

How do the interests of carriers differ from the interests of consumers? Are carriers less prone to adding new features?
It's not that the carriers don't like new features. We might say, 'Here is the best way to design an e-mail solution,' and the carrier might say, 'Well, we don't like that way because it doesn't tie into the infrastructure that we've already deployed.' They want all of their cell phones to have the same e-mail apparatus (even though what they have may not be nearly as good as the one we come up with) because they want to have a consistent story across their product family. It's not that we want more features and they want fewer features; it's just that they have certain requirements.

We used to have much more control of our destiny. Now we have to focus on our enterprise sales. It's a diplomatic process--in how we approach various issues and design problems, and in how we figure out which features are essential.

Handspring is morphing from a consumer product company to an almost intellectual property company. Do you have a reference design for your new products?
We actually design the whole product and then go to the handset makers. It's not a reference design. We have not gone down the reference design path yet.

As the performance of the wide area networks improves and their costs come down, the advantages that Wi-Fi brings are sort of boxed-in.

Maybe that will change in the future. We're not a large company. We can't be the best manufacturer of cell phones, but we're really good at developing user interface, software and industrial designs.

Handspring has made a dramatic shift in its product focus. What are some of the other product changes that the company is making?
It's a little counterintuitive. The people who created Graffiti and stylus text input are now developing products with keyboards, while we're building cell phones with keyboards. We think that many cell phones in the future will have keyboards.

Why is that?
We learned this process. Our thinking was, "If we're going to add the capability of e-mailing and messaging to a product, a little keyboard is better." It's faster and more accurate. Once we had designed a keyboard for the Treo, we began to realize that while keyboards are good for text, they could be really powerful for the basic phone. It was a very simple idea: You should be dialing by name and not by number.

We've designed our Treo products so that you can type in someone's initials and get all their phone numbers listed. So, in essentially three or four taps on a keyboard, you can look up any of several thousand phone numbers. It's quicker than speed dialing.

You think this will be a trend?
Keyboards will show up in well more than 50 percent of cell phones in the future. You can make them smaller and more functional.

How do you do that? By changing the layout?
I don't want to be too detailed here. In terms of key shapes, layouts, colors, it's how you backlight the keys and how you overwrite their functions. There are a lot of varieties.

When do you think combination cell phone/organizer devices will cease being considered high-end, and what do you think will cause that change?
When we reach a certain volume of units sold, component pricing will drop dramatically and will feed on itself. When we did the Palm Pilot, we struggled to make the lowest-end Palm Pilot with 32K of memory--and we struggled to sell it for $299.

In fact, we really couldn't make any money at $299. So we sold a higher-end model for $369. That was about seven years ago. Now that product--or a better product--can be bought for $79, while a high-end product on sale for $300 comes with a beautiful color display, a fast processor and lots of memory.

You should be dialing by name and not by number.

The same thing is going to happen here. In the beginning, these devices will be expensive, attract a small audience and won't sell rapidly. It's inevitable that the functionality of these things will drive broader adoption.

Seven years ago, memory and screens were probably the most expensive components. What about now?
Cellular radios are the most expensive part, but even now, people are working on some very dramatic shifts. If you look inside a communicator, you'll find two processors: one for the radio and one for the device. Chipmakers are developing a single processor architecture for radios that will be powerful enough to handle organizer capabilities. That will save some money.

As with any technology, as volume builds, people figure out clever ways to drive costs down. In just four years--maybe as soon as two--from now, you'll be able to buy one of these things for $99.

Do you foresee more features like cameras being built into cell phones? How about technologies like Bluetooth?
I'm not a technology guy. I understand technologies, but I don't really get excited about technologies. I get excited about applications. I've been cautious about Bluetooth for many years, because I ask myself what a user would do with it. It reminds me of infrared. Infrared was in a bunch of products, but no one ever used it. And we didn't put it into the Palm Pilot until we had a great application for it.

We were the first people to make infrared popular. You could beam card information from one device to another. It worked 100 percent of the time, and essentially it became the first real successful consumer application of infrared in a computing device. It's all about applications and functionality, not technology.

Can Wi-Fi and cellular coexist in a device?
I'm a big believer in Wi-Fi and free local-area wireless, and I believe that those technologies will become more prevalent. Some cellular carriers have been making investments in Wi-Fi, and some companies have been talking about introducing products that have both Wi-Fi and cellular radios.

Now, if you have a good communicator product with fairly high data rates--which will be getting better in the short term--and you're paying virtually nothing for data, you should ask yourself what benefit you're getting for having Wi-Fi capability. You could get 10mbps, but for most of the things you do on a handset, you don't need that kind of performance. You can already do any kind of Web search--you can do a hell of a lot. If you're thinking, however, about downloading audio and video, that's something I probably wouldn't do over a wide area network.

What about voice communications over Wi-Fi?
What's the advantage? For $40 or $45 a month, I get unlimited voice and data now. As the performance of wide area networks improves and their costs come down, the advantages that Wi-Fi brings are sort of boxed-in. Then there is a sort of complexity issue, because if I have two radios in a product, I have to know how to switch between them or how they work together. If I start my call on a Wi-Fi network and walk outside and get into my car, do I have to hang up and call again on the cellular network? Can I switch it off?

So it's two separate worlds, and the Wi-Fi thing is going to take a while to sort itself out. You're seeing the experiments already, but no one except the equipment vendors is making any money off of it right now, in terms of service providers. So I don't believe that's going to be a major player.

What you're going to see is cellular data networks get really good, and we're talking up to 300kbps as early as in the next year and a half. That's going to be a lot for most people. It's not that Wi-Fi is not going to exist; it's just a question of whether it needs to co-exist with cellular service in a device. There are also technical problems with Wi-Fi. Today they tend to have much higher power consumption, so battery life with Wi-Fi is much shorter than it is on a wide area network.