Made of Jade: Aoraki's Gil Simpson

In a matter of weeks, a relationship between Christchurch's Aoraki Corporation and Unisys that has spanned 20 years will end - signalling a challenging new era for Aoraki. The relationship began when a 30-year-old programmer called Gil Simpson walked into the Christchurch office of the mighty Burroughs Corporation (which later merged with Sperry to become Unisys) and asked for some free time on one of their mainframes to work on a new software idea. The idea evolved into LINC (Logic Information Network Compiler) a development environment Simpson, as CEO of Aoraki, sold to Burroughs for a nominal $1 in exchange for the income his company would gain from developing and supporting the product. Simpson's focus is now on Jade, the end-to-end development, database and distributed processing environment that Victoria University's Communications chair Howard Frederick believes to be 'the new C++ - but nobody knows it yet' and which Intel New Zealand chief Scott Gilmour says is 'world-beating technology'. Both men worry that Jade needs a bigger push to meet its destiny. In this full transcript of an interview by Russell Brown for the April issue of Unlimited magazine, Simpson traverses that issue and much more.

In a matter of weeks, a relationship between Christchurch's Aoraki Corporation and Unisys that has spanned 20 years will end - signalling a challenging new era for Aoraki. The relationship began when a 30-year-old programmer called Gil Simpson walked into the Christchurch office of the mighty Burroughs Corporation (which later merged with Sperry to become Unisys) and asked for some free time on one of their mainframes to work on a new software idea. The idea evolved into LINC (Logic Information Network Compiler) a development environment Simpson, as CEO of Aoraki, sold to Burroughs for a nominal $1 in exchange for the income his company would gain from developing and supporting the product. Simpson's focus is now on Jade, the end-to-end development, database and distributed processing environment that Victoria University's Communications chair Howard Frederick believes to be "the new C++ - but nobody knows it yet" and which Intel New Zealand chief Scott Gilmour says is "world-beating technology". Both men worry that Jade needs a bigger push to meet its destiny. In this full transcript of an interview by Russell Brown for the April issue of Unlimited magazine, Simpson traverses that issue and much more. The LINC relationship with Unisys ends soon. It means it's nothing but Jade now. How much difference will that make to you? I suppose it used to be in a sense nothing but LINC - but probably a better way to put it is that back then it was nothing but Burroughs. Whereas with Jade, we're not tied into that sort of exclusive relationship. For the LINC story, we're down to 20 people now, and at our peak we had around 130. We've gradually come down over the years. And I really appreciate the effort put in by the rearguard - everyone wants to work on the new stuff. But it's been good, profitable revenue for the company - not extraordinary, because you're just selling someone's time and there's a limit to how much margin you can really charge on that. Unisys actually formally cancelled the contract in September 1993, and at that stage they confidently expected to be gone by January 1995 at the latest. And I think quite frankly that Unisys are still ill-prepared for LINC, but that's their issue. Did you shed a tear over it at any point? It was your baby, after all. I wouldn't say I shed a tear I don't think Unisys have made the right business decisions. I wanted to move the LINC base to Jade in an orderly way through Unisys. I offered to put Jade on their proprietary mainframes as an example. I think LINC's been a magnificent product and people out there have made a large investment in it. And I thought it was important for Unisys to be able to offer the preservation of that investment. But they thought otherwise. They took the product away in 1983 and embarrassingly had to send it back, so I don't think they'll take it back this time. There's too many jobs on the line. What did you take out of the LINC experience? I think we took out of it the issues of certification and documentation. The issues of having a field test programme. We've had times when Jade has fallen over for customers, but nowhere near to the extent that we had with LINC. We've learned a lot. And when we see things going wrong, we think déjà vu and we can very quickly move to remedy those problems. It's embarrassing, some of the problems we've had with Jade. Things you'd think we would have thought of. Identical problems to those we had with LINC. But at least we've been able to react quickly and confidently. The second thing is that I know how long it takes to take a product into the world market. I've long since got over the idea that because you've got a great product, therefore it should sell. I know we've got a great product but by no means does that mean it will sell. There's a lot of work to do. Is there a good name abroad for Aoraki and Cardinal as a result of the LINC association? Is that going to help? I think so. It gives us a degree of credibility. People will find this amazing, but there's an aspect of that that detracts from it - because computer companies usually never have a successful follow-on product. Most try, but they usually fail. It's like saying, what's behind Oracle? Say Oracle come out with a new product and not just a new version of Oracle. Jade is in no way an evolution of LINC, it's quite a radical step. In database terms it's like from flat-files to relational database. It's interesting that you should mention Oracle, because I've been thinking quite a lot about Oracle as I've been talking to people from Aoraki. I noticed the plaque stating your commitment to quality in your main building and I'm interested in how you go about achieving that. I've witnessed the Oracle culture first-hand once, when Ellison stood up and unexpectedly declared their development product, Sedona, wouldn't be shipping - and it was like a public execution. His style is quite brutal, playing off groups within the company against each other. What's the culture here? If you're going to have a quality culture you've got to have an honest culture. And if that ends up appearing to be brutal because it gets into the public domain, then I would say, so be it. You can't have quality without honesty. You can't walk around with your head in the bloody clouds ignoring it. But hopefully, that honesty's contained within the company. I don't think it is admired necessarily by the market. The market likes to have confidence that you can manage things. But if it does get into the public domain, and you've got a problem, them you've just got to be upfront and honest as fast as you can. And if that means pulling a product, then so be it. How much does it help in leading an IT company to have actually cut code yourself? I think it helps a great deal. It concerns me that there are chief information officers out there who have no idea how to program in any of the so-called modern languages - Java or C++. So they have little understanding of what a product like Jade does, they've no idea what LINC is or how it operates. That's pretty scary, I think. I don't think that can necessarily be corrected, but I'm not sure how it will unfold. Our industry is going to continue to change, and those changes will be phenomenal over the next five years - and the CIO's role is going to equally continue to change. And they are not immune from the changes in the industry. The outputs are not there in our industry. If we were a performing industry delivering to our customers' expectations, then there would be issue. But we're not achievers. I'd classify the computer industry as the worst bunch of underachievers in the history of man - with the possible exception of the military. Only the military has wasted more money and more people's time and talents than the computer software industry. So how does the computer software industry keep on getting away with it? Well, it's sort of like they've got the keys of the cave where the secret rites are performed. I mean, how can the user take control? The user is not empowered to take control, they don't have the language to take control, they don't have conceptual models to work with so they can understand systems. They've got an operating system with more instructions in it than The Bible CIOs aren't exempt either. They recommended to their companies that they make these purchases and spend this money with consultants and buy these packages which have not been the success they were expected to be. I've been intrigued at the way, say, SAP has been able to walk in and order companies to completely change the way their businesses are run in order to use the software. Yeah. Change your business to the way we think it should run. Lose your differentiation, and if you want to keep your differentiation then just pour money into this big, black hole over here while we re-architecture SAP. It's staggering. You buy a package which isn't about how your business is run and then try fundamentally reconstruct the ship while it's at sea. It's just unbelievable, the stupidity of the engineering that goes on. You don't take a 747 apart while it's at 37,000 feet, but that's what people do with packages. The architecture's just not there. Packages are driven by a fear of programming. Programming is scary, and packages give you a feeling of comfort, because you pre-buy all this code. Don't drink salt water if you're thirsty at sea - that's what these people do. Was it the case that you had a moment of clarity about how things really ought to be done? There is clearly a philosophy about how things should be done embodied in Jade, isn't there? My point of clarity came in about 1973. That was the first major point and there were a few after that. But basically, I adopted a view that programmers as a bunch of people were basically okay, and that it had to be the way we were programming that was wrong. The way we were programming was wrong - and that's what LINC came out of. LINC was about the ability to conceptualise large-scale systems. Because back in the mid-70s, people were starting to consider building these very major applications to replace their perceived legacy systems then - which are now the systems they're trying to make Year 2000-compliant. We call them heritage systems now - we had to change the name! But I felt these systems wouldn't work, because nobody could conceptualise them. We just didn't have the way of seeing systems. We had fiasco after fiasco. In this part of the world, the CS-90 from Wespac was the most publicly evident problem - where they publicly admitted writing off $800m Australian. But that's unfair on Westpac, because there were lots of projects like that." We're reached a state now where many people, and particularly those in government, are actively suspicious of IT because so many projects have gone wrong. They have every right to be suspicious. What I find staggering is that they don't even look at the new technology of Jade. They don't look adequately at products like LINC. I wouldn't say LINC has had the penetration into New Zealand government that it warranted. Where it was used, it was used very successfully compared to other investments. That's obviously disappointing and it's something we're trying to work on. Should national and local government look first to use a locally developed product? Definitely. I think they should have a policy of why shouldn't they use a local product. And I don't see that kind of attitude among the bureaucracy. They seem to think it's presumptuous of us to even assume that we could supply such an august group of people with the scale and complexity of systems they have to run. I think there's also a hell of a difference between coming to Christchurch to look at a system and going to Canada or the United Kingdom. I've seen that in public health, going back to when the Auckland Area Health Board bought a system from the UK. And the guy went to jail for that on corruption charges. Although we'd run second on that we never got the chance to come back in. I don't think the system had a very successful outcome, ever. But it was from the UK and you got a trip there to have a look at it - which is a hell of a lot better than coming to Christchurch! It seems that it's almost an aspect of your company's credibility that when people do come here they're well-treated. Yes it is. It's fundamental. If you have come half way around the world to have a look at this product, you have an expectation that this has to be something different. And if we don't respond to that expectation, we're undermining the product's credibility. They get in to Auckland and then they fly to Christchurch - the back of bloody beyond from their point of view. They get off the place and they're saying, by Christ, this had better bloody be worth it. So their first impressions are very important. They need to confirm it was right for them to make this decision. It also seems like an expression of confidence on the part of the company itself - the fact that you have the corporate wine cellar and the rest - as if you're saying 'we're a real company'. Yeah - one with our own little quirks and sets of interests. It's a personality, isn't it? It's okay to be different. If you're going to have an innovative climate, then you've got to have an environment which says it's okay to be different. We've got a lot of things that are a bit different, things that other companies might look sideways at. And perhaps rightfully so. In terms of your own role, I wonder if the CEOs of IT companies play a stronger role in the branding of the company than in most other industries. I think that's true, because we're a creative industry. We're a creative technology industry. My view of the computer industry is of a model that's very close to the movie industry. You've got your big distributors - Warner Brothers, Paramount Pictures - and your key directors, your staff. And in a broad sense I think the software industry fits much better with that model than with any other corporate model. I you were to look up on me as a director with a film company, then you'd have a better view of what this company is all about. I've got some great actors, some great producers, all the elements A lot of good New Zealand directors go to Hollywood. Why are you in Christchurch and not in Silicon Valley, where it might be a bit easier to get the movie made? Fair question. In some ways, what I feel is that in New Zealand, and in Christchurch in particular, it's about the ability to attract and retain. The one thing we need in the software industry is continuity. If I've got a star, I can't afford to lose that person - I need them all the way through. Within the company, there are people who support that star, and they're just as important. The mobility of staff in the US, in a place like Silicon Valley, is extremely high, and it's totally unsatisfactory for software development. And that's why bugger-all comes out of it, frankly. All these companies are facing dilemmas of one sort of another. I think we're achieving very high productive goals here. You simply couldn't produce Jade in the US with the size team we've got, because you couldn't hold them together for long enough. I've got all the original Jade team together here, and most of those people before that had 10 years of LINC before that." You've been hiring world wide in the past couple of years - what attracts people here? Reputation, I suppose. That's important. And then coming to the location and finding it's an okay place to live. The key element we look for is experience, people who've done things. The door's always open for that. You've gone on record on issues like the compliance cost of doing business in New Zealand. What could the government do to make life easier for businesses like yours? It sounds like government, and at the end of the day I suppose it is. But it's really more an aspect of the bureaucracy. Inland Revenue has a policy unit. We had capitalisation of software, which hasn't been adequately dealt with. But my whole finance group is busy now doing transfer pricing, which would have cost this company well over $200,000, possibly even double that, in the last 12 to 15 months, just to comply with our obligations. And if we don't do that, we are guilty until proven innocent. We get a choice - do all this, set it all, up, and then the IRD has to prove that we're wrong. If we don't do it, the IRD can issue an assessment. We have to pay half of that and we have to prove our innocence. And as we do all that, back at IRD there's still this policy unit - already developing the next policy in the next area. So once we stagger through capitalisation of software, which was a major impact on this business, then transfer pricing, there'll be some new policy. There are people in Wellington whose job it is to just create bloody change for the sake of it. What's your view on the changes to education announced this year by the government - splitting off tertiary education with the apparent aim of modernising it? Anything that enhances transparency and competitiveness in education is good. There's a point where they're going to break through this academic freedom stuff - which is the bomb shelter that most of these people seem to take cover in. There is a role for academic freedom, but it's only that, it's only part of education. To use those things as reasons why you can't have transparency, accountability, contestability and competitiveness is a different debate altogether. I do have a tertiary focus in my thinking. I can see that with secondary, it's really a matter of the PPTA and when the PPTA's gatekeeper role is gone we'll see changes - very good changes, probably. I do like the fact that we now seem to have a minister of tertiary education. Do you have much time for the Michael Porter-style ideas of industry clusters, perhaps feeding off the top of the universities? Is that worth pursuing? Yes. I think Porter's got a lot of very good things to say. There's not a lot of things I'd disagree with Porter over. I don't think he places enough emphasis on things like having the right work environment and the right work culture. Perhaps that's because he sees things in a macro sense. The reasons that economists have difficulty predicting whether the economy's going to grow or not is that 50% of that is to do with people's attitudes. If the attitude of the people is one of confidence, then they will take risks and they will do things that will cause the economy to grow. If they're unsure, they won't take those risks. This thing of attitude is fundamental, and I think you could probably run the economic models far better if you took a case that said, well, given these economic circumstances, what's going to happen if there's either a good or a bad attitude? I think Porter's ideas on clustering and that sort of thing are quite valid. And certainly the bureaucracy will buy into it - but they won't buy New Zealand software. They haven't got the attitude. I'm interested that you should mention workplace culture. I've been quite aware of a culture at Aoraki - your staff seem well-rounded, positive people. They're not all well-rounded we have a few here of what I call the unemployables. But I think what's important is that if people feel good about being here and that it's natural for them. The fact that I might be able to look at a few of them and say they would struggle in another environment is just my view, and it's healthy if they just see themselves as being happy and enjoying themselves. We work quite hard to make sure people have en environment they're comfortable in. Do you think you'd be as receptive as Hugo Simpson was when you walked into Burroughs and pitched LINC, if someone walked in here tomorrow and said 'I've got a great idea, cut me some slack and we'll see what we can do'? I would say put it in writing. Did he say that? No I'd say put it in writing, and then I'd look at that and if it had a reasonable touch of credibility I'd have a look at it. But the first step is to get it in writing. If they can't express it in writing, then it's a dream You can't just walk in here with a good idea. You've got to have made it reality in some way. There are thousands of good ideas, and they're like good intentions. They're never going to happen." Many people regard it as a major problem in New Zealand that banks are so risk-averse, and there are few other sources of capital. Yeah, you're right - we need more venture capital activity in New Zealand, but forget the banks. We are wrong to look to the banks. The banks have long since moved away from that. Maybe they'll one day own shares in a venture capital company. I think the signs of venture capital are emerging, and the more the merrier. People have to understand that their money's at risk, and that good venture capitalists will make a lot and the bad ones won't It's a speciality area and I'd like to see more of it across Australasia. How do you bring that about? Make it attractive for investors to do it. It's going to be foreign money, so rather than the government borrowing the money offshore as they'd done in the old days and dish it out through some bloody good science fund or some business development board, now we're in a position where venture capitalists or funds in the US would look to make an investment through an operation in New Zealand. I think that's good. But they haven't yet. Well, they might. I've heard a few rumours around that kind of area. Is there anything for us in the Israeli and Irish experiences, where there was a public sector role in bringing in venture capital in the first place? I'm not sure. I'd like to restrict public sector stuff to things like good science funds. I think in my heart of hearts there is a role for that. Things that NASA does in the US, they're no different from a good science fund - they give a contract to Boeing or Lockheed." NASA supported half the Internet for a while there. Yeah. So there's a role for good science-type funding. It can be controlled, developed and enhanced. I'm reluctant about concepts of any start-up grants. I think that allowing someone who's been successfully running a business for 20 years to give free advice to someone who's just starting out is what's needed. If the education people want to contribute, then they can teach people at school and university that if they want to do something they should go and ask someone. I don't remember being ever taught that at school. They could run a great game with kids - give them a massive amount of work and let them figure out that if 10 of them all did a bit each there wasn't much there at all. And nobody would be a loser. Let's look into the future. What's the next five years for the industry? Hardware is going to become even more embarrassingly cheap. In software, we have to get off the ERP bandwagon because it's coming to the end of the line. People have pumped a lot of money into that and they're not going to pump any more. This industry will need to move on to its next gold-mining exercise, which I suspect will be electronic commerce. I think we're going to have a fundamental revolution in commerce, caused by electronic commerce. Electronic commerce is about interactive information systems. Systems talking to each other interactively. And it's about integration of your systems with those of your suppliers and customers. It's going to require major changes to systems. Large portions of them will have to become real-time so they can interact, one-to-another. Things like capacity planning and material planning will have to become real-time activities. There's going to be a front-office revolution which will equal the back-office revolution of the 1960s, which only affected the people involved and not the community at large. The front office revolution will affect the community at large and will be far more spectacular in that sense. That sounds exciting. But what does it mean? Your information systems are going to become your front office and your customer is going to deal directly with your information systems. I see call centres as a transient phase. It's ridiculous to call someone on the phone who then talks to a computer system. There's no reason you shouldn't talk directly over the Internet to that system. In some cases you might get personal assistance at the same time. So the biggest revolution since the 1960s is going to occur and it's going to be far-reaching. You ain't seen nothing yet. We couldn't have this electronic revolution to the extent we've had without something ultimately blowing up or giving way or imploding or whatever." What's going to be the public perception of this? Are people going to be aware of a revolution? They'll just see new ways of doing things suddenly become available. They probably won't think of the ramifications. But it will mean phenomenal changes. I've got this term - and I'm going to trademark it for Australia and New Zealand, I think - and it's "Y21K" - 21st century compliant. Don't worry about Y2K, it's a non-issue. I wouldn't even worry about being Y2K - you might as well close up your business and liquidate it. That's the challenge and we're going to be right there. Our organisation is interested in firms that want to be 21st-century compliant. When you started work on Jade was it informed by what you're saying now? No. I don't think people should get carried away with the visionary stuff. The genesis of Jade is a design I did in 1984, for a mainframe with 1000 processors, each with their own memory. That was the hardware issue - but my problem was, how do you run a program on a machine like that? We know how to run a program on one processor, or on four processors sharing one memory, but the reason we don't have more than that is that it gets less efficient, to the point where you've got nine processors and it runs slower than if you had one. So how could you make 1000 processors work? The problem of distributed processing - that's what Jade came from. And then what happened was the PC arrived and the LAN arrived. And if you can imagine the LAN as the backplane of a computer, each with these processors with their own local memory hung off them, suddenly you've got this machine. And then the Internet arrived, which was brought about by the telcos. Don't let any bastard ever tell you it was the computer industry. It was the telcos who delivered commodity pricing of connection to a data network. I mean, Bill Gates, Mr Superhighway himself, had to revolutionise his own company in six months to catch up! Even Gates wasn't aware of the Internet. It was the telcos who did it - and the electronics boys at the end of the day who delivered us that, as they've delivered everything else. My view on Gates is that he's a marketing guru, not a technology guru. I've got a lot of admiration for him. I believe he's the cleverest marketer in our industry. The way he spun his company round when he saw the market moving He's not egotistical. He must have quite a high degree of humility, that guy. I know that with little things I've done in this company, I've thought this isn't doing my ego much good, but it's got to be done. You must have had a few wrong runs in your business - the brief venture into consumer CD-Roms was an aberration, for example. Yes. We like to look upon these things as experiences (laughs). At least I'm not as bad as Carter Holt Harvey who called pouring $47 million into SAP "an investment". I'm not that stupid. But rather than taking wrong turns, it's more been a case of us continuing to go up trails that were leading into the desert. You notice that there are fewer and fewer trees out the window and things are getting fairly stunted and it's taken us too long to notice that we weren't going anywhere. It's a turn-back concept. Our business venture taking Cardinal into the US was probably the worst example of us not pulling out soon enough. That cost us a lot of money, and I'd blame my own ego in that case. We got too far in and I don't want to make that mistake again. We should have pulled out at least two years before we did. We pulled out in February 1995, and we should have been out in 1993. It's hard to escape the fact that many of the major software vendors are rushing to develop for Linux right now - but you still seem to be putting your energy into proprietary OSes - NT, AIX and the Unisys flavour of Unix. Is there likely to be a Linux strategy emerging for Jade? Yes, I think I might have alluded to it - it's like Java, we're in a position to watch the horses as they come into the straight, so to speak and if the Linux activity carries on - we're certainly taking an interest in Linux, we've had Linux running We haven't made the strategic decision to release or implement it, but we're very Linux-aware. What's your view on open-source software in general? Do you think it's going to find its place in the enterprise? I think that it will have some place, but I wonder how much of that is a reaction against the Microsoft world, which is quite monopolistic. I think a lot of it is driven by that emotion. The issue that people want with software is support, and support means people and people cost money. You're obviously going to have to take Jade to the world in order to fulfill its promise. How are you going to do that? Through our own efforts, largely, in Australia and New Zealand, with the support of partners. And then hopefully through those relationships that we form in Australia and New Zealand and to a lesser extent in the UK, then we'll achieve it in a global sense. We'll try and find people that want to do business with us. In talking about your company and Jade with other people in the industry, one thing that has cropped up several times is the concern that you're not going to be able to do it on your own - that you're going to have to sell part of the product to a larger organisation or stage an IPO. What's your response to that? My response would be that, no, you can't do it all on your own, but how the structure might be in the future that will evolve. Clearly we're working on developing relationships and developing partnerships with people. For the product to be global it will involve relationships and partnerships. Trevor Eagle suggested to me you were pretty good at playing your cards close to your chest and you'd have a strategy up your sleeve. Is that the case? I tend to try and run a series of strategies, really, because you don't know who wants to do business with you, if I could answer it like that. At the end of the day it comes back to who wants to do business with you. It takes time. I'm looking forward to the day that somebody decides it's a good thing.

Join the newsletter!

Error: Please check your email address.
Show Comments
[]