This is a speech I gave in mid-1994 to the ToastMasters group at Electronic Arts. You can probably tell that I was frustrated with the way EA did business. I'd also just read Ed Yourdon's Decline and Fall of the American Programmer. There's still a lot of truth in this speech, I think, but there's not much that's going to change things. Organizations either think long-term or short-term, and most American corporations are in the latter category. Do your best to get into one that's in the former.
The French Revolution had "Liberte', Equalite', Fraternite'". EA has "ACTION", and it's going to destroy us. The reason it's going to destroy us is that we've got Achievement, Customer Satisfaction, Teamwork, Integrity, Ownership, and Now!, but we don't have Quality. There's no 'Q' in ACTION. And Quality is THE single critical factor in success.
We make and sell software for a living, and everyone in this room who works for EA depends upon that software for their livelihood. Software — more generally Information Technology — is a critical portion of the American economy. We are the cutting edge of America's economic future. Video games have recently been classified as a vital technology in the latest US-Japan chip pact, in which the Japanese are supposed to increase their chip imports in certain specific areas. We all know that you won't sell hardware without software — good quality software. Without us, US chip companies won't sell those chips to Japan.
It used to be said that "...what's good for General Motors is good for the Country!". Really. Then the Japanese auto makers came along in the early to mid-70s and ate Detroit's lunch. A major reason they did that was a man named W. Edwards Demming. Demming invented something called Total Quality, and in an oversimplified nutshell it means that if you make a product of good quality, everything else takes care of itself. Detroit — in fact American industry in general — laughed at Demming. So he went to Japan, where they virtually canonized him. There's a scene in Back To The Future 3 that illustrates this: "No wonder the machine broke — all these parts are Made in Japan", "Sure, Doc, all the best stuff comes from Japan".
If we only had to compete in the US, we might depend upon the Government to apply protectionist measures for our support. But we have to compete in a global marketplace now. Sega, Nintendo, and Sony are all Japanese companies, and they expect to sell hardware and software in Japan. We want to sell software all around the world — Japan, the rest of the Far East, Europe, the emerging Soviet bloc countries, South America, and here in North America. To do that, we‚ve got to compete with global programmers. I doubt if there's a creative person working in Development at EA who doesn't cost at least $3K per month. There are programmers in India with bachelor degrees (still rare in our industry) and two years of experience who get paid $180 per month. That's 1/16th as much — and they are more than 1/16 as productive as we are — in many cases they're MORE productive in measurable ways than programmers in the US. And they function under tremendous handicaps — computers are expensive to get and fix because it costs a lot to import them, they have unreliable power, and most of their phone lines are incapable of carrying 1200 baud data much of the time. But India has a national program to create an Information industry. We don't — we have the "Information Superhighway".
EA doesn't consider quality of code or product to be mission critical. According to Bing, the quality of a product is related to how much it sells. I think that's bullshit. Quality is a separate, objective, quantifiable element. We ship products with known fatal bugs all the time — probably every quarter. We ship product that we know won't sell well to "make the numbers". We don't code for defect-free software — and it can be done. We don't measure the productivity of our creative people, so even if we try to improve our techniques, we can't tell if we have succeeded or not. We don't track defects in our software in any meaningful way. We don't train our people to do their jobs better. We don't try to match people to the job they can do best. Most especially, we don't concern ourselves with the maintenance of our products. In today's market, any product that we don't think will be have a sequel or be ported is a product we shouldn't be doing. We have created roughly 12 versions of Madden since the first, and every one of them has had fewer features and cost more than they would have if the first one had been written to be maintained in the first place. We are more concerned with the timeliness of delivery of a product then we are with its quality.
As an industry, as a company, we are in mortal danger. So what can we do about it? I can think of three things. First — each person in this room must make a Personal Commitment to Quality. That means telling other people "this product isn't ready to ship yet — it's not good enough". We all know that 9 times out of 10 that probably won't make a difference. But if enough people start doing it, it will make a difference eventually. Second — Improve Yourself! My boss buys books for me to learn how to do my job better. When was the last time you read a book about how to do your job? When was the last time you read a book about how someone else should do their job so you could work better with them? Electronic Arts has an Educational Assistance policy — if you take a class at a local college after getting approval, and you get at least a B, EA will pay for that class. When was the last time you took advantage of that? Third, we must force EA to make a Corporate Commitment to Quality — to put a 'Q' in ACTION, if you will. By sabotage if necessary (and some development groups do this, believe it or not), we must commit EA to producing quality product.
In the mid 1980's, a large corporate group was suffering through a lot of what EA's been through. They'd had a lot of trouble launching product on time, and they were under a lot of pressure. Bureaucrats and corporate politicians put pressure on again and again to make the dates, or the group wouldn't make the money they needed to keep operating. Technical people keep warning them that these things couldn't be done, that they weren't safe. But the products kept getting launched and they keep working OK. One product in particular was late, but needed to be launched on time. The technical people said once again that it wasn't ready, it wouldn't work right, it wasn't safe. Corporate politicians overruled the technicians and launched the product on schedule. And the Challenger exploded. That is the end result of leaving technical decisions in the hands of bureaucrats. That's what happens when political decisions override technical ones.
Why should we concern ourselves with Quality? Why should we put a 'Q' into ACTION? Because if we don't, eventually it means our jobs. Our cushy lifestyles, our cars, our toys. Because if we don't, we destroy a portion of the American economy that's essential to the future economic well-being of not only this region, but of America as a whole. The destruction of the US Steel industry and the US Auto industry have had long-reaching economic effects that are not yet over. Because it's the right thing for us to do.
Because if we don't seize Quality now, and maintain it for the future, our fate is as certain, if not as spectacular, as the Challenger's.
Copyright © 1994 - 2003, Evan Robinson. All Rights Reserved.