A Conversation with PEN PAL’s Sam Wiegand

Over the past couple of years, we’ve been very fortunate to speak with many of the pioneers of this industry. As we noted each time, many of them, perhaps not surprisingly, share a common background at Silicon Valley’s mobile computing incubator, GRiD Systems. Among these have been MobileSoft’s Pete Snell, PenRight!’s David Marino, and Palm Computing’s Jeff Hawkins.

After all this, it’s time to come full circle and talk with a person that not only worked at GRiD, but was one of its founders: Sam Wiegand. Wiegand has seen it all in the mobile computing industry, punctuated by his most recent venture, Menlo Park, California-based PEN PAL Associates.

The Early Computer Industry

Wiegand started in this industry almost before there was an industry. He started: “I got into the computing industry through Honeywell in 1956 and began initially as an assistant, went into sales and spent quite a few years there. I left to join a small startup in a business known as computer output microfilm, which was bursting on the scene in the 1960’s. It was the first time you could generate microfilm directly from a computer file. So we started a company to build indices of information files that people had on computers.”

Unfortunately, that company was undercapitalized and, in a short while, Wiegand found himself moving west. He recalled: “I came out here [California] to start Diablo Systems in the summer of 1969. I wasn’t the key founder and, unfortunately, the company was sold prematurely to Xerox in 1972.”

Industry veterans will recall that Diablo Systems made IBM mainframe compatible disk drives, however their real claim to fame was their innovative daisy wheel printers. Wiegand recalled with pride: “That’s the only press conference I’ve attended where, when we finished demonstrating the printer, the whole room stood up and applauded, at a press conference. Very rare.”

Wiegand stayed with Xerox after the acquisition for about a year and a half, and then set out to join Tandem, serving as vice president from its inception until 1980, when he became associated with GRiD Systems.

Wiegand painted a picture of the times: “In the late 1970’s, everyone was talking about PCs and if you went around in computer circles, there was a lot of interest in it. I was interested in that because I’ve always liked the idea of being involved with new technology, which is not a good thing at all; I wouldn’t recommend it.”

He continued: “We thought we could put a PC in a briefcase—we called them briefcase computers. I thought at that time that I really didn’t want to be working full-time, so I invested in the company and went on the board [of directors]. I am listed as a founder of GRiD and put some of the original money in.”

“One idea was for the computer to be very strong, and work around a systems concept, but it was not a client-server relationship. Instead, you could call in over a telephone line and get information which, around 1980, was still pretty well advanced. In those days, it had around 512 KB of total storage, so you had to have some place else to go to get information.”

The Birth of GRiD

GRiD Systems was incorporated in 1979, and the company completed its first round of financing at a couple hundred thousand dollars in the spring of 1980. Wiegand recalled: “This was a critical time for the company so I decided to join as vice president of sales and marketing and went through the initial launch.”

“After about a year and a half, I retired and moved up to Tahoe. People joke because it was my second retirement. In those days, I had already had a retirement party when I left Tandem.”

With the company’s goal set at building a portable Notebook computer called a briefcase computer, they set about to convince others of the need. In fact, Wiegand described a beat-up old briefcase with a plastic mock-up inside that they used to show around. But what would anyone want with a mobile computer anyway?

Wiegand replied: “We noticed that people like to access information and be able to use that information wherever they are. Even going back to 1980, there were a lot of people working outside of their offices. People wanted something they could carry around with them, but still be able to dial in, get information, and download it.”

“Normally when you download information to something like a smart terminal, all you can do is look at it. Conceptually, we thought if we could download information to a computer (this was before the IBM PC), we thought we could build some software, which we did, which would allow people not only to look at the information, but manipulate it and change it.”

“So GRiD built a whole suite of software packages, including a little worksheet, a little text processor, and a little personal information manager. GRiD built all of that in the first product release. Conceptually, it was a very strong, systems oriented package, which is what appealed to me because I like selling into large accounts. I didn’t visualize then at all the consumer market, the individual buyer.”

Hitting the Target

Wiegand described how the original target price point was set at $5000 (strange how that number never seems to change). However, Wiegand was also slightly frustrated at the trend to think only in terms of price. He noted: “A lot of our pricing sensitivity in this market isn’t related to what the actual value of the thing is. It’s related to the fact that the industry has made a big mistake in telling everybody what everything costs.”

“In the PC business, everybody knows what a memory chip costs, what a disk drive costs, and what a Pentium costs. Therefore I don’t want to pay anything [extra] for it.”

As it turns out, the original GRiD computer, together with its bubble memory and electroluminescent screen, debuted for $8000. However, Wiegand was quick to point out that the original IBM PC, configured for the office, was $5600 in 1981.

“In the late 1970’s, everyone was talking about PCs and if you went around in computer circles, there was a lot of interest in it. I was interested in that because I’ve always liked the idea of being involved with new technology, which is not a good thing at all; I wouldn’t recommend it.”

All in all, Wiegand was pleased: “The initial reception was great. It was a sexy machine. It weighed 10 lbs., fit in half a briefcase, and had a magnesium case. It was a beautifully designed product. It was even in the Museum of Modern Art for years and years, it may still be there. It was a landmark product design.”

Identifying Field Workers

The GRiD had many things going for it, but by 1981, one thing that it didn’t have was IBM compatibility. Wiegand acknowledges that they didn’t realize the importance of this quickly enough, and credits Compaq for being in the right place at the right time. He conceded: “Compaq hit the ball square, they hit that sweetspot. They were IBM compatible and came up with a very good product.”

Realizing that it was difficult to compete in the general purpose business market, GRiD made the bold move of concentrating on the vertical markets. Wiegand recalled: “We switched and adopted the concept of field systems.”

“We were selling systems solutions to people who wanted to provide their sales people—we used to say, if it has the world field in it, it was our market. Field service, field sales, field engineering, field audit, field management, whenever it was in a field, that was a GRiD prospect.”

“We didn’t have an IBM compatible with Lotus, but what we did have appealed to the sales and service managers. We sold to those people to give their people in the field better access and better tools. That became less of a problem after we got the IBM compatible machines, which was finally shipping in the fall of 1984. We were the first in field systems.”

The Dawn of Pen Computing

Wiegand recounted the story of how pen computing got started at GRiD Systems: “The pen computer side of it, which started in 1987, came about because one of the problems, from our standpoint, was that the machines were too big and too heavy, and a lot of the people worked standing up. Jeff Hawkins had gone off to Berkeley to get a Ph.D. in computer sciences or something like that, and he came back and said that he hadn’t really been able to figure out how the brain works, but he finally had a good algorithm for handwriting recognition.”

“He asked us if we could structure a deal where he could complete this handwriting algorithm and, at the same time, be the project manager for a pen-based computer. We thought we had a wonderful market for it, because we were already calling on all these people—we had a blue-ribbon list of customers, banks, insurance companies, all the automobile companies. So we said great, he built it, and Business Week published a big article about it. And we all thought we were off to the races.”

“We introduced this in 1988. We started showing it to our user groups, and we actually announced it in September of 1989. We started shipping at the end of 1989 and that’s when we realized that there was no software to run on it.”

“We had developed the early basis for PenRight!, those APIs were there, and we thought end-users would build their own applications. Which is what we did at Tandem. Tandem users built their own applications.”

However, there were some problems. “It’s not that customers didn’t want to, but nobody knew how to write a pen application. When you write a pen application, you often design your own user interface, which is an exciting thing to do but it’s not easy. The early customers had a lot of trouble coming up with interfaces that satisfied their users. These were not computer users, these were pen and paper users.”

The Challenges of Pen Computing

At first glance, one would think that designing an application based on paper forms would be relatively straightforward. However, Wiegand assured me that this isn’t the case. “When you do that, you leave out all of the advantages of the computer. Paper forms are designed with a tremendous amount of redundancy with the expectation of a lot of errors. If you create a paper form, because of the cost of the paper, they jam way too much on it.”

“But when you go to a pen application, there’s a couple of things you want to happen: one is a very fast response when you touch something, and when you make a mistake, you want it to catch it immediately. So these developers were people who had written COBOL programs for IBM mainframes, they were just stumped.”

“We tried to get users more involved in what we called interactive design where the user and the programmer sit together and, in effect, perform built-in testing as you went along. It was like prototyping, but we were trying to build something that actually worked. However, it was pretty hard. One reason was that C programmers were pretty rare at that time.”

“When we started GRiD in 1980, there was very little C; we wrote everything in Pascal. We switched over to C in 1985 or ’86. So, in 1988, there weren’t that many C programmers. Now, I’m sure that there are a lot more, but I still believe that there is a tendency for them to be working on higher level things than user interfaces for field service people. They work on networking and trying to get all these client-server systems to work together, which isn’t easy.”

Client-Server to Blame

In speaking with corporate developers and managers, Wiegand has found that many departments are simply overwhelmed by the new technologies. He speculated: “I think that’s one of the reasons that pen computing has not attracted the attention from the MIS department that it deserves. They have been so busy with client-server computing, communications, and networking over the past ten years, they’ve gotten way over their heads in many cases.”

Unfortunately, this doesn’t leave much time for other pursuits. “Their technical resources have been very strained. I really think you’ll find that in many large companies, the primary emphasis over the last ten years have been on client-server, online network type systems. The guy out there in the field hasn’t really come to the center of their picture.”

Exit Stage Right

As any investor will tell you, how you exit an investment is as important as how you enter it. Wiegand recalled GRiD’s plans at the time: “We had hoped to go public in the fall of 1988, and there was a big crash in the market. In fact, the crash came in the exact week that we were on the road with Montgomery Securities visiting investors.”

“On Black Tuesday, we were riding through downtown Manhattan in a big white limousine headed for Morgan Stanley. Of course, they cancelled the meeting. So we ended up selling the company to Tandy Corp. We signed the agreement in February of 1989 and the sale formally took place in June of 1989.”

“At the time, everybody was clamoring to get into the laptop market and prices were falling very fast. For us to really compete, we had to grow very quickly. We knew we had to do that. We were counting on a public market to do that, but it didn’t happen.”

“In those days, if you went to a Japanese company and said we want you to build this machine for us, they’d say fine, but you’ve got to order a hundred thousand, or three hundred thousand, or five hundred thousand. We didn’t have the resources to do that, Tandy did. That’s what was attractive to us, their buying power.”

Was this a satisfactory exit? Wiegand was thoughtful: “Given the situation, I’d say it was satisfactory. There were an awful lot of people at GRiD who were very very disappointed. As a result, we lost a lot of people during 1990 and 1991. They had hoped for much more. But the investors were eager to get some liquidity out of the company. We really didn’t have a better alternative.”

Has any of this dulled his enthusiasm for the concept of a pen computer? The answer is not entirely. Wiegand noted: “The market that we saw, I still believe in, which is for a walking worker, or a field worker. It’s a person who spends his time outside of the company premises.”

“Their technical resources have been very strained. I really think you’ll find that in many large companies, the primary emphasis over the last ten years have been on client-server, online network type systems. The guy out there in the field hasn’t really come to the center of their picture.”

Selling into Corporations

“We saw that market so clearly through the GRiD sales force during the 80’s that we knew that market existed. We also knew that it was a custom application market because as soon as we started to sell into that market based on Lotus or something else, it didn’t work.”

“Those large companies had their own operating methods, procedures, and traditions—you couldn’t change those things. You had to build a system around those existing barriers, or they would just throw it into the trunk of the car and you would never see it again.”

However, different people see markets in different ways. Wiegand recalled: “One of the conceptual differences between Jeff [Hawkins] and I is this: Jeff has always had this conviction that if you built the right application, you would sell pen computers by the zillions to the consumer.”

“And that’s why he founded Palm Computing, to build applications that would be so compelling that these machines would just march off the shelf. I didn’t say that market doesn’t exist, I don’t know anything about that market, I’ve never been in consumer marketing. But I know that it’s not the market that GRiD knew anything about.”

Wiegand continued: “What we knew about were custom applications. Jeff [Hawkins] and I were looking at this from two different market viewpoints. In the meantime, the people at GO were looking at a totally different market, which was replacing the desktop and laptop with a pen computer.”

“It was a totally different market orientation. We were replacing pen and paper in our minds for people who had never used a computer. Jeff [Hawkins] was providing something that would make everybody an instant communicator, game player, and whatever else. But GO thought that they would replace the desktops and the laptops with pen computers. They were going to get rid of the keyboard altogether.”

Reception Good, Reality Harsher

Pen computing seemed to be starting off on a good footing and Wiegand was encouraged at every turn: “We had a lot of customers at GRiD that, we felt, would clearly be receptive to this concept—and it was. When we started showing them a plastic prototype in the fall of 1988, the reception was always very very positive.”

But…

“I think there were a couple main breakdowns. One was that writing the applications was a lot harder and more time consuming than people really believed. We signed up Thomas English Muffins, which was our first hard contract for the GRiD Pad around the summer of 1989, and they put three experienced C programmers on that.”

“This was an application that was already running on a little handheld computer, so it wasn’t that they didn’t understand the application. When I left seven or eight months later, it still wasn’t working. That was a good insight into what the difficulty was going to be to get applications up and running.”

“That was one problem. The second problem was that the machines were not reliable. This was true of the GRiD Pad, and it was true of every other machine. We had problems with the digitizer, the battery life, and condensation. We had problems with the screens in that if you could read them in the sun, you couldn’t read them inside.”

“We used to put these pilots out, eight or ten machines, and then pretty soon you would start to hear what happened: it broke, it didn’t work, not enough battery life, I can’t read the screen. All these things that would make people put them aside.”

“We’ve sold thousands of copies of PEN PAL, and almost every one of those people wrote an application. But most of those applications never really got rolled out because they got tested with machines that didn’t work.”

“So an awful lot of work has been done out there that I count as money in the bank because those people are out there. They like the idea, the concept is appealing to them, but they were frustrated with the reliability.”

Thanks Microsoft and GO

Reliability wasn’t the only issue. The battle to win the hearts and souls of mobile computer users also took its toll. Wiegand was direct: “I’d say that if it hadn’t been for the battle between GO and Microsoft, which escalated pen computing to the front pages of all the press, this business would probably be further along.”

“GO and Microsoft were in a PR battle, where they were flying around presenting to every large company, often flying on the same plane to make sure they don’t loose sight of each other. They were telling everybody how great it was going to be, and which was better.”

“People were setting up study committees to decide whether they liked PenPoint or Windows for Pen. More typical of new technology is the attitude: ‘let’s give it a try, but let’s not have a high profile.'”

“If you go back to the early days of Tandem, or even the early days of any technology, you don’t do it in the glare of publicity. You go out there with some early users who will try it and tell you what’s wrong. You fix that, and then you try it again. You do it in the context of a serene partnership, until you get it right. But this thing burst like the sun in the sky and everybody was talking about pen computing.”

Changing Some Lives

Elaborating on the hype surrounding pen computing, Wiegand recalled: “I was at the NCR announcement and it was incredible. The chairman of the board of NCR was there, the president, all the senior marketing people, everyone. I’ll never forget the opening remark, I’m sure this is almost verbatim.”

“He said: ‘This is more important than the announcement of the Xerox 911 copier, this is more important than the announcement of the IBM PC. This is the announcement that will change the way you live.’ I’m not kidding, the president of NCR.”

Well, it probably changed the way he lives.

According to Wiegand, this had the effect of creating all the wrong expectations: “It got tremendous publicity. All of this was going on and we’d go around and say that we have a DOS application builder. They’d say DOS? What are you talking about, you can’t do this with DOS. You’ve got to have object oriented programming, C++, multitasking, and we’d say why? All this guy wants to do is record how many loafs of bread he delivered.”

At this point, there was no stopping Wiegand and his story: “Then Apple came out, and I don’t know why in the world they did it, but they started jumping the Newton when nothing worked. By now, people were so confused about pen computing.”

“So now all of corporate America is sitting around saying ‘what is this? You have to have a new operating system to do pen computing?’ All I’m saying is that part of pen computing’s problem was that it was subjected to far too much scrutiny before it was really ready for scrutiny.”

Wiegand’s conclusions are almost obvious: “A pen computer has to be so rugged that, at the end of Friday afternoon, a person is going to throw it into the trunk of the car, put the golf clubs on top of it, throw the baby stroller on top of that, and not dig it out until Monday morning.”

“In spite of all this, some machines did go out. You can tolerate a lot of problems if you are fixing them in obscurity. But if you’ve got Doonesbury on your back, that’s too much. Nobody can face that.”

The Idea Behind PEN PAL

When Wiegand co-founded PEN PAL Associates, the notion was to address the fundamental concerns of the corporate, vertical market developer. PEN PAL had to be powerful, but it also had to leverage the skills found in most MIS shops.

Wiegand illustrated the power of PEN PAL by comparing it to PenRight!: “They’re different. PenRight! is basically a C language development environment. Our concept is that we could grow the market better and faster if we get rid of the C requirement.”

“And it’s not just C, you have to know the DOS APIs, and you also have to know the PenRight! API, of which there are 270 or something. So, unless you are a pretty strong programmer, you’re not going to be able to build an application very quickly.”

“I’d say that if it hadn’t been for the battle between GO and Microsoft, which escalated pen computing to the front pages of all the press, this business would probably be further along…they were flying around presenting to every large company, often flying on the same plane to make sure they don’t loose sight of each other.”

“It’s one of those things that not only requires a significant amount of skill, but it requires daily use. PEN PAL does not require C experience. If you understand programming in BASIC, COBOL, or RPG, you can pick up PEN PAL very quickly.”

Now that the expectations for pen computing have, in many people’s mind, been completely derailed, how do we get this train back on track? In other words, is there a marketing message we need to broadcast out to developers and implementors telling them that it’s safe to come back into the water?

Wiegand takes this notion and turns it on its head. He expounded: “You’re assuming that things happen because of marketing, and I don’t believe that. I’ve been in marketing all my life, and I don’t believe anything ever happens because of marketing. You may make it appear like something happened, which GO did and which Sculley did, but nothing ever happens that way in actual fact.”

“I’m a much stronger believer that in the long term, if you build a good product and keep improving it and make it successful, you’ll be better off investing in the product than investing in the market.”

“Now that’s not true of consumer markets, but I’ve never been in consumer marketing. If you can sell pet rocks, I’m all in favor of it, but you can’t do that when you’re selling to corporate America. These are all long term relationships.”

“It’s what makes it very difficult to get started, but it’s what creates a very lucrative business long term.”

Sam Wiegand
President
PEN PAL Associates, Inc.
120 Independence Drive
Suite 124
Menlo Park, CA 94025
(415) 462-4888
(415) 462-4899 (fax)