Palm claims twice the enterprise users of Pocket PC - Nagel

by Guy Kewney | posted on 12 November 2002

In his first public appearance in Europe since accepting the job of leading Palmsource (the software company) out of Palm Computing (the hardware side) David Nagel talked to NewsWireless Net - and suggests that Microsoft's apparent triumph in the enterprise PDA market is more perception than reality.

Guy Kewney

David Nagel left Apple - where he had expensive products and wanted to focus on the revenue - for Palm. Suddenly, he's a big fan of "look at the number of units sold!" marketing. He arrived in London this week, to explain where the Palm was going, and where it's operating system is going, too. And no surprise; he thinks mobile communicator devices are the way of the future.

<1/> David Nagel

"It's a common experience if you talk to Palm users; they tell you 'I have my life in that thing.' People develop an intimate relationship with these devices, which is why we adopted the slogan 'your world, wherever you are'." These are different from personal computers - people have a different relationship with them, Nagel said.

Here's the text of his presentation:

We haven't been in existence all that long; we first shipped a Palm Pilot in 1997. And despite everytthing Palm Pen company in Japan has argued since then - strenuously - that the name should not be used - people tend to call them that: Palm Pilots.

Demand has grown quite steadily; more than 25m of these devices have shipped, more than 21m of those are still in service. I deduce that from the fact that the bulk of them shipped in the last three years, and we now see that the replacement cycle is 2-3 years.

I have only been with PalmSource for about a year, but Palm made the wise decision to licence to others from the outset; first licensee was Symbol in 1998, doing ruggedised versions of the Palm. Several new licences are due for announcement in the next year, but products aren't available yet so I can't announce their names yet.

We have 245,000 software developers who have downloaded the development kit; the bulk of these are in the US; but it is a world-wide phenomenon. We grow by 1,000 to 2,000 every week. And we have a large R&D centre in Montpelier, France.

Our plan is to become completely independent of Palm in the first quarter of 2003 1Q. So: what do we think will happen?

Well, we think Mobile Digital Devices represent a new stage in the evolution of computing. Each of these transitions - from mainframe to mini, from mini to PC, and so on - has been driven by technology. The largest was the creation of the PC industry - and that growth was based largely on the realisation that it should be opened to third parties to develop them. The concept of shrink-wrap software didn't exist; it created PCs. Initially, there was a contest between Apple, Amiga and others on one hand, and the Wintel alliance between Microsoft and Intel on the other - which has turned into a clone industry, with a strong reference design. And everybody built the same thing.

The winner of that contest is pretty clear. Apple is nearly gone, having just three per cent of the worldwide, market, and Wintel has clearly triumphed.

We think the next kind of product, the family of mobile digital devices, will differ in a couple of ways. PCs have not taken advantage of Moore's law in terms of driving the prices down. They're remained pretty static in price - relatively. We are talking about things in the £200-300 range, not in the £600 range. And our prices are now much lower; down to the $99 Zire we announced this month.

People who buy the $99 products tend to be first-time users, in some cases people who have never used PCs (remember, PCs have achieved only 60% penetration in the US). Other important characteristics will distinguish this from the PC market; we don't think this mobile digital device market will develop as a clone business. We're trying to develop a platform where the software is compatible, but the actual design of the hardware is quite distinct.

Today, 2002, 15m of these devices will be sold; which compares with figures for PCs at around 100m or cellphones at 400m. But this is expected to triple, so that in four years, there will be more like 60m users. And a growing fraction will be in the communications space.

There are interesting differences between Blackberry/Tungsten - data centric - devices, as opposed to voice-centric. Users think of these data centric devices as things they do messaging with. In Europe, the initial wave is more on the phone centric; and perhaps here in Europe, data centric may never be as popular as in the US; where SMS has never caught on. That's possibly because the five US carriers have never got together to arrange a common gateway for SMS.

Let me give some pointers to how large this market may get; our research is based on a wide number of countries, US, Germany, UK, France, China; and covering every sort of mobile device - notebook replacements, game machines, communications, and so on. We divided them up into several categories, and almost 70% of the sample voiced strong interest in products in one or more categories. By far the two most popular were communicators. Everybody imagines that cellphones will transform themselves into one of these devices. The stand alone PDA will - if it does not completely disappear - be much diminished; most devices will be networked in one form or another.

But we haven't hit the boom time, really. The installation of 2.5 and 3G network, is proceeding, if not uniformly, in almost every market, but the market for products in the pocket is pretty flat. There are spikes; for example, the Japanese market always introduces a spike into the figures; and Sony's Clie made them have 80% of the market when they announced the ARM-based Palm machine. Otherwise, the markets are pretty flat. From our point of view, though, I'd like to make the point that we're still doing pretty well.

I can quote figures from the NYT a few weeks ago; front of one of their sections - they have recently added a section showing best-selling PDAs. We're shown as holding 50% of the market, with Sony alone getting 17.8%, while Handspring is getting 13%. By contrast, the biggest Pocket PC seller is Compaq at 7.4% and Toshiba with 4.1%.

Handspring is interesting. It has focused, in the last year, heavily on the communicator devices; and we can expect to see some great things going forward. And Kyocera is 3rd largest; it was first to introduce 3rd generation Palm-powered product.

There are other quite significant players, not all of which are pocket-format. For example, in Europe, Alphasmart is a name you possibly haven't heard of. Their "Dana" is designed as a PC replacement for the primary school market, what we in America call the K-thru-12 segment. It has a character oriented display. They are networked, have wireless capability, and give the kids 30 hours battery life; becoming very popular in schools, because they are a quarter the price of a PC.

There are going to be more. In the next few months, will see new licensees doing products that might best be described as entertainment devices, location based devices, communicators.

There's a misapprehension about the home versus enterprise markets. People all seem to think that Pocket PC has done well in the enterprise; but the enterprise market leader is, still, Palm; it has sold nearly twice as many as the Pocket PC. There was an initial boom in corporate Pocket PC sales, but after initial PPC penetration, it has stabilised. One reason Palm sales are relatively hard to track, is the fact that enterprise purchases tend to be made by individuals, re-imbursed by the company.

Now, we believe our approach to this market is very different from the approach of the "central planned economy" of the Microsoft market, where they are even doing hardware design, starting with mice, then XBox, and now SmartPhone and Phone Edition Pocket PC. There are requirements of design which Microsoft lays down. Our approach is the opposite; we only specify the microprocessors; either the Motorola 68K or the ARM. All we ask is that devices pass the software compatibility test. Over time, we think this is the strength of the platform.

The other way we differ from Microsoft is that we actively work with our licensees to advance the platform. Chip licensees have provided low-level driver stuff; Sony and other top end licensees have provided applications. We think the effect is a more rapid evolution of the platform. It's derived from the thinking that not all the IQ points reside on your team.

Our latest operating system is OS5, which is our first ARM based system. There are many reasons to going for ARM; multiple sources, first (20 licensees world wide, as opposed to Motorola alone) and we think the pricing will be driven down; and performance is between 2 and 30 times the 68K, and most typically in the 5 to 10X range. We now have things we didn't have the horsepower to do; multimedia, higher resolution screen. We can do more, for example, in the areas of security; these devices over the next couple of years will have significantly greater security than PCs. Principally this is because they must have, becasue they are easily lost; I've lost a number of devices over the years! - and as someone pointed out, "it may only cost $250, but it has $25m worth of corporate data on it, so how are you going to protect it?"

In the phone business, there is no mass production; in Windows, it is extremely uniform. Our market segment, mobile communication devices, will come between them. We're introducing the "Mass Customisation" concept - consumer markets are very favourably disposed to branding that meets specific narrow niches. I might give you a parallel with bottled water, where branding is very specific! So over time, we think you will see an increasing range of products. One tranche of products will be entertainment - including multimedia - then there will be information products - and the other large tranche will certainly be the communicator.

Here, there is an enormous range, or a huge amount of uncertainty about what will be the dominant products. Will they be phone based, or data centric, or SmartPhone category? There is an enormous amount of uncertainty - but surely, relentlessly, as we speak they are rolling out and will be an important part of the future.

We've been in comms for a while. The first phone we produced was a Qualcomm phone, 3-4 years ago; the Brick as we called it. Kyocera built a 6035; with a "palm filling" format, as opposed to palm-sized one. Then Handspring came out with a more attractive one, in Treo; and Samsung came out with i330. And now, Kyocera has announced the clamshell design.

One of the fastest areas of growth is growth to support wireless, on the technology side. The work we've done over the last twelve months has been primarily designed to support the growing wireless standards. We plan to support all the technologies. There's a lot of interest in the 802.11 technologies and Bluetooth, and support for both of these is built into Palm OS 5 - and will be extended as we go forward.

Other devices we are supporting are the low-bandwidth, paging networks, like the Transcomm network in the UK. In the US, this is more important than other parts of the world; they are high power transmitters, penetrating buildings; and unlike GSM, can be used in hospitals. They may only be 10Kbits per second throughput, and they do involve high latency, but they are useful.

We're also looking at our position, pricewise, in the market. Prior to shipping first ARM products, Palm was clearly positioned at the low end, while Pocket PC was at $400 up to $1000 range; over the last few months both have moved strikingly to the left of the graph; Palm's Zire has moved below $100, and some of the Sony products are only $500, while Windows is down to $300. Our strategy going forward is to drive prices down; and the design means we can win here. Zire has only a 2 megabyte memory footprint; while the smallest pocket PC has 32 megabytes; so we think in any kind of a price war going forward we will end up with favourable margins.

Part of our strategy is to preserve the margins of our licensees. Microsoft's record on this isn't good; today there's only one PC maker that makes money, and that's Dell. We think that's bad for the consumer, and bad to have that much concentration of power in the hands of one or two companies.

There are opportunities where we are not present; and where we are not, today, is in China. We will fix this problem. The opportunity there is pretty significant. It's hard to know - estimate ranges run from 4m units a year to 7m. At the upper end, that would put it near the US, and around the European size. Remember, it could grow hugely; there are an estimated 225 million Chinese today who are trying to learn English, buying electronic dictionaries, and other portable electronics devices.

Where we are definitely ahead is in applications. I have US data only; at the latest count, we had 15,500 third party applications, and about 430 of those showed up in the last 30 days. The difference between Palm and Microsoft is not only large, but is actually increasing. And the Symbian platform is much much smaller than either of the other two platforms in terms of application count - but no doubt they are doing things to change that.

I mentioned that Palm is found in a large fraction of enterprises, and this has been an enduring statistic for the last four years; it's been a constant answer. People say "at work" principally when we ask: "where do you use your Palm?" What is surprising is that it's selling at a rate of 2:1 compared to Pocket PC - and the RIM Blackberry, which is perceived as being up in front, is actually neck and neck with Pocket PC.

So although Microsoft is assumed to be the dominant player in the enterprise, this isn't the case. For us, a bad scenario would be for CIOs to say: "Use Pocket PC at work." That isn't happening yet. On the contrary, take these two new logos, GM, and Novartis. They are users; they have Palm on approved IT list, and employees buy them. These devices are not large enterprise purchases, but are reimbursed.

In future, we think a key advantage we can exploit is usability.

We are particularly anxious to drive this on. For example, take the example of beaming and your business card. One of the key ways we think we are demonstrably better is in ease of use, here. We'll pick out a couple of examples; just count the number of keystrokes involved. On Palm, you press a button and hold it down, and the business card is beamed. By contrast, Microsoft is more like Windows; you tap and hold, and select and then tap and hold and select: four operations at least. It's a different design philosophy; they are making it as much as possible like a PC.

Ditto on adding something to the address book, on Palm, you just click and start writing; while on Pocket it is about six steps. Symbian is more Palmlike, but there is one extra step. So when you measure the amount of time taken doing something - searching the calendar and so on - as well as the efficiency of the OS, Palm comes out much better. These benchmarks are measures of the relative times to do the same task or, in some cases, the amount of memory to store the same data.

Finally, the total cost of ownership of buying Palm is strongly a function of initial price, capital cost, and a significant amount of saved cost against its rivals, because of the significant amount of softare updates to PocketPC. You can spend about twice as much in administration costs, on a PPC as you do on a Palm device, because of Microsoft's frequent updates.

The future is probably going to be more voice-centric than previously thought. But data centric devices will remain important. What we think will become significant, is the amount of data that gets synchronised over networks, using IP standards. We've announced support for SyncML, but the trend will be the way we improve on the standards to migrate towards continuous network updates, not the "trickle" or "push" technology that rivals are using.

In the near future, we have a few new licensees to announce - in 60 days or so there will be at least two. Some will be exclusive, others, like Samsung, will be people supporting more than one platform. In most cases, we have found that designers have a strong reticence to commit to any single platform; most will adopt multiple platforms. But that's now. Over time we think that will sort out; only very large companies can afford - people like Sony, Samsung can afford - to have development teams on each platform.

This is going to be a very competitive market - and frankly, I hope this will be true for a long time. I hope it doesn't go the way of the PC. I don't just mean in terms of the way they look, all the same style of box. I mean the style of use. If you don't like the way that Windows works, it's really difficult to find anything else.

That said, the PC will have its influence. For example, Dell is going to have an impact on the hand-held computing space. Their differentiation is how you buy things from them and their build-to-order wystem they've really defined to be better than anyone else's. They will bring Pocket PC pricing closer to where the Palm pricing is. Even so, we think Palm vendors will still have a signficant advantage in terms of the cost of hardware.

We've been asked whether Sony will take over the platform. I don't think so. I think they have made it clear that they have a long-term commitment. It's a valuable commitment; for example, if you look at the future of the Internet, IP version 6 is crucial. It's an area we are covering in discussions with Sony, and they are very keen on doing this in more areas than just handhelds; it is a very large company which has interest in other areas other than the hand-held space. They have reasons to think that it is very important. We don't have an agreement with them on that, but we have had discussions with them, as to how we might use their development capabilities in that area.

Having said that, Sony declined to have a board seat, and don't wish to have any more influence over what we do. They were the first company we worked with in this co-operative model, and it is where we got the idea that we could replicate this with other licensees. They understand that the best thing for Sony will be an active, independent software company developing the platform.

I suppose we have some weaknesses; some resources it would be nice to have. I don't have $40bn in cash, for one thing. Then we have to face the fact that we don't build servers, don't build PCs, don't have an office application business. And of course, Microsoft is very strong in the enterprise; even when you talk to IT people who have chosen Palm, and ask: "Will Microsoft be an important player?" - they say yes. Microsoft is a sort of assumption. Everybody assumes, as they used to with IBM, their existence, and we all work around them. I am quite confident that Microsoft loses money on every OS it sells in this market; but it has an enormous amouont of money it can lose.

We differ from Microsoft in several areas. For example the OS in Palm is a much lower power-consuming thing, based on the architecture itself, than Microsoft's OS. And they have a different strategy in the area of synchronisation; by most measures our Hotsync is better than Activesync - in areas like not losing data, and being faster - but they also support continuous sync which we think is a nice model. And you have to remember, Microsoft is good at detecting best practices amongst competitors and building that in; they are enormously talented at doing this! - and it's part of their corporate culture.

To an extent, I suppose you can say that we benefit from being "anybody but Microsoft." Microsoft was rebuffed by the phone-making community; but this doesn't guarantee us a future, because Microsoft is implacable; and you can see this in the way they came back with designs for phones for people who didn't make phones before.

It is certainly true that we have to make our products interoperate with the Microsoft desktop, because it's what is on the desktop. That's something we do pretty well today, and we have to convince the IT folks that we can carry on doing that. It isn't getting any harder - even as recently as the Consent Decree, two weeks ago, makes it relatively easier. The trend is at least in the right direction. And we are working with parts of Microsoft - the Exchange group for example - where the attitude is that Exchange does have to do a good job against rival products, like Oracle.

If you get back into the enterprise architecture, there's a trend towards Web-based structures; Microsoft doesn't own that market, IBM and BEA are far stronger. If you look at Java, it is becoming very important on the enterprise, and on the mobile and comms platforms; it provides a common game architecture, standards for service interactions. We think that in providing a common interface to the network is a very good thing to use Java for. That's why we are leading this PDA Profile (takes MIDP technology further along the path). Obviously Microsoft would just as soon Java disappeared. Also we have tools that support Visual Basic, the Visual Studio development.

Remember, I do know something about the IT culture; I was a CIO for a brief period. There are two things you think about when you get up in the morning; and the first is risk. When things are going well, everybody forgets about you, but when they go wrong, you get noticed - so you ask: How do I reduce my risk? And the other thing is not to get too dependent on one vendor. You find that if you do, they have the prices ratchet up on you. So there are strong "restoring forces" working in that sector of the market. We think we give the IT folks the best of both worlds.

Small things that show where we're going: we have announced a deal with Good; they are porting their client sofware onto the Palm OS; so you will have the sort of push email that was popularised by RIM. The Good folks claim they have superior product over RIM, and we think that's a supportable claim; and we think it's going to be cheaper to upgrade, and easier in use.

We are not going to be an embedded OS vendor. There are a dozen large ones just in US along. The embedded market is tough, low margin, lsot of custom. Having said that, you will see Palm devices in things that are battery operated and portable, personal; not things like set top boxes. But you will see an increasingly diversity of devices.

We think we're on the right path. If you think you want to disagree, look at Creative Labs, Logitech, or Garmin; look at their gross margins, when they do differentiative products. They have 20% operating margins, 50% product margins. Compare that with the PC business where Dell is the only PC supplier to make any profit; Dell has 17% gross margin, 7% operating income. Differentiation is the way to go, not clone-making.