Please feel free to forward this newsletter to as many friends as possible. I have a personal goal of having one million readers. I can achieve it with your help
the Software View: Underlying software laws
Welcome back, gentle reader. In the 1970's, the state of Wisconsin passed a law requiring the manual flushing of urinals in all buildings. Wisconsin state law requires that anyone sending, receiving or transmitting a telegraph must be helped on a first come-first serve basis. If your ram is out and "running at large" you will be forced to pay a five dollar fine. Other outlandish laws include protecting state inmates' rights to never eat margarine as it is illegal to serve a butter substitute in state prisons, and to never work on Sundays or holidays. In Washington D.C., it is illegal to use your parking place to dry laundry. In Georgia, a funeral home can lose its license if one uses obscene language around a corpse.
For those of you with Web access and a Netscape Navigator browser, please click here:
http://www.softwareview.com/
Scroll down the page and you will notice a link entitled, "Daily view weblog". The daily news page is also known as a "web log". It is en vogue and the fashion of these days to call it that. Click on the link, click "reload" on your browser or clear your browser cache to ensure that you always receive the freshest, hottest daily news concerning JavaTM, Linux, XML, and the software industry! The link never changes, but I will be updating the HTML file page behind it every day. Please, do take a gander at it every day.
Also, gentle readers, the Software View is an Associate Internet World Wide Web site of Amazon.com. I'd like to extend my sincere, heartfelt gratitude and thanks for your patronage. I'm offering links to books, et cetera that you can purchase from my web site. I'd greatly appreciate it if you would purchase software industry books from my web site. Help support my newsletter and web site by purchasing items from Amazon.com from my web site. Here is the URL (Uniform Resource Locator):
Click here
Now, dear readers, on with this week's episode of the Software View!
You have to read it, because it's the law! In this issue, I will be discussing software laws that undergird, underpin our industry. Like the laws of nature, the seasons of the software industry rise and fall and operate according to these laws. If you choose to flout them, you do so at your own peril.Moore's Law: "The number of transistors per square inch on integrated circuits had doubled every year since the integrated circuit was invented or the pace of microprocessor technology change is such that the amount of data storage that a microprocessor can hold doubles every year or at least every 18 months." A related corollary is Machrone's Law: "Gordon Moore just plain got it right ... I should also mention that Moore's Law has also given rise to Machrone's Law, which was true for many years, which is that the machine you want always costs $5,000." A related corollary is Rock's Law: "A very small addendum to Moore's Law is Rock's Law which says that the cost of capital equipment to build semiconductors will double every four years." And finally, a related corollary is Parkinson's Law of Data: "Data expands to fill the space available for storage." Niklaus Wirth, the inventor of Pascal and other programming languages, has propounded two new Parkinson's Laws for software: "Software expands to fill the available memory," and "Software is getting slower more rapidly than hardware gets faster." Moore was born in San Francisco, California, on January 3, 1929. He co-founded Intel in 1968 with Robert Noyce. He was named Intel CEO in 1975. He was elected Intel chairman and CEO in 1979. He served as both until 1987. Before Intel, he was director of research and development for the Fairchild Semiconductor division of Fairchild Camera and Instrument, a company he co-founded in 1957. He received a Bachelor's in chemistry from the University of California at Berkeley and a Ph.D. from the California Institute of Technology. He received the National Medal of Technology in 1990 from then-President George Bush. In 1965, Gordon E. Moore, co-founder of Fairchild Semiconductor and Intel and Intel's Chairman Emeritus, was preparing a speech and made the memorable observation that was canonized as Moore's Law. Intel's web site states, "In 1965, Gordon Moore was preparing a speech and made a memorable observation. When he started to graph data about the growth in memory chip performance, he realized there was a striking trend. Each new chip contained roughly twice as much capacity as its predecessor, and each chip was released within 18-24 months of the previous chip. If this trend continued, he reasoned, computing power would rise exponentially over relatively brief periods of time. Moore's observation, now known as Moore's Law, described a trend that has continued and is still remarkably accurate. It is the basis for many planners' performance forecasts. In 26 years the number of transistors on a chip has increased more than 3,200 times, from 2,300 on the 4004 in 1971 to 7.5 million on the Pentium II processor. Moore was a brilliant, hard-working engineer and trusted colleague, in charge of research and development at Fairchild, and, along with Bob Noyce, one of the eight founders of the legendary Fairchild Semiconductor. Moore was Intel's employee number two. FOLDOC writes, "Moore's Law has been (mis)interpreted to mean many things over the years. In particular, microprocessor performance has increased faster than the number of transistors per chip. The number of MIPS has, on average, doubled every 1.8 years for the past 25 years, or every 1.6 years for the last 10 years. While more recent processors have had wider data paths, which would correspond to an increase in transistor count, their performance has also increased due to increased clock rates. Chip density in transistors per unit area has increased less quickly - a factor of only 146 between the 4004 (12 mm2) and the Pentium Pro (196 mm2) (doubling every 3.3 years). Feature size has decreased from 10 to 0.35 microns which would give over 800 times as many transistors per unit. However, the automatic layout required to cope with the increased complexity is less efficient than the hand layout used for early processors." PC Webopedia writes, "Most experts, including Moore himself, expect Moore's Law to hold for at least another two decades." CNET reports, "Moore's Law is coming into direct conflict with the law of nature. In September 1997, Moore stated that the industry's ability to shrink a microprocessor through improved manufacturing processes is going to start butting up against the finite size of atomic particles." In September 1997, announcements by Intel of 2-bit flash memory and by IBM of chip circuitry of copper rather than aluminum suggested a return of the original version of Moore's Law. Moore states, "We had a very big program to get as many new designs to put into our microprocessors as we could in the 1979-80 time period. We got more than 2,000 designs. One of them was the IBM PC. We didn't recognize it as being anything very different than the other 1,999 designs at the time. It took awhile to appreciate the impact of that." Computerworld writes, "There is nothing natural about Moore's Law. It emerged from the work of people developing dramatic new technologies, products and processes. And there is nothing sacred about the 18-to-24-month time period between generations of microchips. The relentless pace of chip development has to do with many different factors. Among them: having access to a lot of capital, moving research and development close to the manufacturing process, creating open organizations where innovation can flourish, accepting change as a constant and, of course, having a healthy dose of luck. Indeed, these are the factors that have shaped Silicon Valley over the last four decades." But there is another, more fundamental driver of Moore's Law, according to the man himself. "More than anything, once something like this gets established, it becomes more or less a self-fulfilling prophecy. The Semiconductor Industry Association puts out the technology road map, which continues this generation [turnover] every three years. Everyone in the industry recognizes that if they don't stay on essentially that curve they will fall behind. So it sort of drives itself. Until we get to the point where, for some reason or another, no matter how hard we try we can't do it anymore." What will stop the silicon freight train? Moore, who in 1965 first identified the relationship between time and the number of transistors on a microchip, has two fundamental concerns about staying on the curve. First, there are business limitations. As chip densities rise, the cost of production rises almost exponentially. Second, there are real physical limitations. Problems arise as chip design gets down to the atomic level. Barring a replacement technology, Moore's Law has until about 2010 before it comes up against a brick wall, Moore says. What then? Will product development begin to slow? Will the cost of successive generations of technology rise instead of fall, as it has over the past three decades? A lot depends on technology innovations over the next few years. But it wouldn't hurt for business leaders to start thinking about the consequences of the end of Moore's Law. Moore states, "In one respect it has become a self-fulfilling prophecy. People know they have to stay on that curve to remain competitive, so they put the effort in to make it happen." He now believes that people refer to Moore's Law as anything in semiconductors that when planted on semilog paper results in a straight line. "I used to think (Moore's Law) was a historical curiosity. As I continued to work on it, I thought it was a self-fulfilling prophecy," says Randy Isaac, head of basic science at IBM. "Now I view it more as a self-consistent economic cycle." A self-what? Isaac explains that, basically, there are a lot of factors - expectations, money and many different pieces of technology - that feed and play off one another to perpetuate Moore's Law. "It just hangs together," he says. John Lazzaro writes, "Hoeneisen and Mead wrote a paper in 1972, "Fundamental limitations in microelectronics -- I. MOS technology", that made technical projections of just how small MOS transistors could be made and still function and why. From what I understand, it was pretty influential in terms of underpinning the predictions of Moore's Law. And incidentally, working on this problem led Carver Mead to believe that "the real problem won't be putting one million transistors on a chip, but what to do with them", which shifted his research focus from device fabrication to IC design, which resulted in the Mead and Conway textbook on VLSI design which had a big influence in the chip design community in the late 70's. Unfortunately the original paper isn't online, but Carver wrote a paper a few years ago taking another stab at projecting the future of MOS scaling that is online (Click here). You might also check out Berkeley's group on MOS device scaling (Click here)." Igor Fodor writes, "What was a stroke of genius in the case of Moore's Law is that Gordon Moore has noticed the regularity and scalability of the DRAM structure. Another stroke of genius of Gordon Moore (and Bill Gates) was to realize that given the learning curves, only the first one down the curve can make money. Therefore they did their best to become all-pervasive industry standards (achieved, of course, only with the help of IBM). The other important observation of both of them was to practice incremental improvement of their products - no radical inventions/innovations. In this way they can stay in business and have control over their customers and at the same time, also over their competition. This led Steve Steinberg to write about Schumpeterian - predictable technological developments in 'Wired' of Jan. 1998. The question remains what happens once the learning curve of Moore's Law is exhausted (DRAMs are slowly approaching the end of their commercial usefulness, i.e. price vs. manufacturing costs: we have now already 64Mb and working on 256Mb. 1Gb might be achieved around 2000-2002? And then? Similarly, the CPUs are moving now from 32-bit to 64-bit devices, with clock frequencies 300-400 MHz. One can double it to 600-800 MHz, or perhaps even 1 GHz. Is that the end or shall we have one day perhaps 128-bit devices? After Gordon Moore, Robert Noyce and Andy Grove, what is the vision of Craig Barrett, the new Intel CEO? The burden of legacy: backward compatibility, which lays on both Intel and Microsoft. This gives rise to increased complexity and acts eventually as a brake (on performance). It has forced Intel already twice to change the CPU architecture. First from 'pure' CISC to hybrid CISC/RISC and then to VLIW (very long instruction word), jointly developed with HP. And as the new fab lines are getting prohibitively expensive there are efforts to share the costs: Siemens with IBM and with Motorola, and Intel with HP." We know that digital technology progression occurs at a predictable rate: chips double in performance every 18 months; storage every 12 months; bandwidth every 9 months - or close enough, anyway. I know a billionaire venture capitalist who says his secret of success has been simply "to project Moore's Law into the future and to imagine what new products and services it would bring." Moore's Law has been a billion-dollar idea to this investor.
Metcalfe's Law: "The value of a network is n squared, with n being the number of nodes on the network." It works on any network, PCs to telephones. Here's how it works: one telephone is useless. There isn't anyone to call. Add another telephone to the network and the value of the network increases. Add a third phone and the value increases again. The very ubiquity of e-mail is what makes it such a valuable tool. Same with fax machines, the phone network, and the inter-continental railway system. Metcalfe's law of the telecosm makes sense, is simple, and, in the ultimate test, feels true. A related corollary is Lipman's Law: "The privacy and security of a given network is "n" raised to the minus 2." Benjamin Lipman writes, "While every extra node on the network increases the value of the network for all users, it also makes it more vulnerable. Take the case of adding a second node to a network. Under Metcalfe’s Law the value of the network jumps from 1 to 4. At the same time, however, the privacy and security decreases from 1 to 1/4. Adding the third node increases the value to 9 and decreases the integrity of the network to 1/9 and so on. Another way to think about it is that for any given "n" the value of the network multiplied by the security and privacy will always equal 1. In other words, any increase in protection will result in a decrease in value, and vice versa. Like the law upon which it is built, this corollary makes sense, is simple and, yes, feels true. Remember the old party telephone lines where many houses shared the same telephone number? Or how about that extra extension and a nosy kid brother or sister? The more people who have access to the information, the less private and secure it is." Dr. Robert M. Metcalfe began his publishing career in March 1992 by joining International Data Group as publisher and chief executive officer of Infoworld Publishing Company. He has served as IDG's vice president of technology since 1993. Just before joining IDG, Dr. Metcalfe was a visiting fellow at the University of Cambridge, England, and was writing for Computerworld, Communications Week, Digital Media, Network Computing, and Technology Review. He now writes a weekly column for Infoworld, where he has served as executive correspondent since 1994. Dr. Metcalfe is a graduate of the Massachusetts Institute of Technology and received his Ph.D. in computer science from Harvard in 1973. He taught part-time at Stanford for eight years, ending in 1983 as consulting associate professor of electrical engineering. He is a member of MIT's Board of Trustees. In 1972, Dr. Metcalfe went to the Xerox Palo Alto Research Center, where he invented Ethernet, the Local-Area Networking technology that now connects almost 20 million computers worldwide. In 1979, he founded 3COM Corporation, a computer networking company which over the next 11 years he helped to grow to $400 million in sales. At 3COM he held various positions including chairman, chief executive officer, president, division general manager, and, most memorably, vice president of sales and marketing. In 1988, he was awarded IEEE's Alexander Graham Bell Medal for his work on the invention, standardization, and commercialization of local area networks. If ever there was an heir apparent to Moore's Law. A law that could succeed and eclipse the power and influence of Moore's Law. A law that could serve as a touchstone for innovation for the software industry from 2010 and beyond. It would have to be Metcalfe's Law. George Gilder writes, "Metcalfe's Law germinated in his mind in 1970 as he read a paper by Norman Abramson of the University of Hawaii describing AlohaNet, a packet radio system used for data communications among the Hawaiian Islands. A student of computer science searching for thesis ideas, Metcalfe believed that by using a form of advanced mathematics called queuing theory he could drastically improve the performance of AlohaNet without damaging its essential elegance and simplicity. What Metcalfe, then a graduate student at Harvard, eventually discovered would bring such networks up toward 90 percent of capacity. In this era of networking, he is the author of what I will call Metcalfe's law of the telecosm, showing the magic of interconnections: connect any number, "n," of machines - whether computers, phones or even cars - and you get "n" squared potential value. A fax machine is only valuable if lots of other people have fax machines, and the only videophone in the world is just about worthless. Think of phones without networks or cars without roads. Conversely, imagine the benefits of linking up tens of millions of computers and sense the exponential power of the telecosm. Meanwhile, the law of the telecosm is launching a similar spiral of performance in transmission media, ultimately increasing their bandwidth, also by a factor of millions. Bandwidth is a replacement for switches. If you can put enough detailed addressing, routing, prioritization and other information on the packets, you don't have to worry about channeling the data through ATM switches. The emergence of dumb, passive all-optical networks with bandwidths some ten- thousandfold larger than existing fiber optics will obviate much of the pressure on switches." As it matures beyond its infancy, the Internet and the World Wide Web will continue to grow and its growth will be influenced by Metcalfe's Law. What this means in layman's terms is quite simple: the more people who are connected to a network, the more valuable it becomes - and increased value attracts even more people to the network. You create a virtuous cycle, with one dynamic feeding off the other. 24 million people (11 percent of the total population) have used the Internet in the past 3 months. 18 million people (8 percent of the total population) have used the World Wide Web. By the year 2000, more than 50 percent of the United States population will utilize the web. By the year 2000, analysts predict commerce and trade via the Internet will approach $50 billion. Apparently, Wall Street believes in Metcalfe's Law. Look at the valuations it has showered on Internet stocks like Netscape, AOL, Yahoo!, Amazon.com, and Broadcast.com. Also, further proof of why I believe that Metcalfe's Law will soon take center stage is that you find no one brags about the speed of the microprocessor in their personal computer any more ("My Pentium is faster than your 386."). People now brag about how fast their connection to the Internet is ("My direct T1 Internet connection is faster than your 28.8 modem.").
Grosch's Law: "The cost of a computer grows as the square root of its speed." Which many of us have always interpreted as a defense in the mainframe days that larger computers were better. Dr. Herb Grosch does not disagree with that. But he also says his law was about consuming of computers by software people. That is, he was an early observer of the "Andy Grove giveth and Bill Gates taketh away" principle, where the software people consume the processing power faster than it can be made. Grosch is remembered for saying that bigger computers are better, but what Grosch wrote in 1953 was "that there is a fundamental rule, which I modestly call Grosch's law, giving added economy only as the square root of the increase in speed -- that is, to do a calculation 10 times as cheaply you must do it 100 times as fast." Grosch's Law is not just about discredited expressions of mainframe scale economies. In explaining why economy only goes up with the square root of speed, Grosch wrote (after updating for political correctness), "No matter how clever the hardware boys and girls are, the software boys and girls will piss it away." You will recognize this as a much earlier form of the oh-so-true, "Grove giveth, but Gates taketh away."
Grove's Laws: "Only the paranoid survive." "What can be done, will be done." "Bandwidth doubles every hundred years." Andrew S. Grove was born in Budapest, Hungary, in 1936. Born as Andras Grof in Hungary in 1936, during World War II, Grove and his mother lived under forged identities to evade the Nazis and his father, George, a Hungarian dairyman, had been snatched away in the middle of the night to spend much of the war on a cruel work brigade. In November of 1956 when, as a 20-year-old chemistry student, he completed a harrowing 12-mile cross-country escape from his native Hungary. His homeland was being overrun by columns of Soviet tanks, the aftermath of the brutal suppression of a student-led uprising against the Soviet Union. "There were minefields and stuff like that," Grove says matter-of-factly, recalling his cross-border journey. He joined a legion of Middle European refugees fleeing the communist state. "You have to picture the scene there," Grove says, shifting his slim frame in a chair at Intel's Santa Clara, California headquarters. "At any given time, there were about 100,000 Hungarians, mostly my age, milling around, looking for food, looking for places to stay," Grove says. "We overran Vienna almost overnight." The International Rescue Committee helped him escape to America. There then followed what Grove remembers as surrealistic acts of kindness. The IRC immediately sent Grove to the dentist after his arrival in the United States. "I thought, 'Why are they doing this? It must be a fraud or something,'" Grove remembers. "It was the first time in my life that I had had my teeth cleaned." His teeth were not his only concern. A childhood bout with scarlet fever had left Grove with a 50 percent hearing loss. His case worker told Grove to find a suitable hearing aid, and the IRC would foot the bill. "I ended up trying out everything," Grove recalls. "I chose one that was, by far, the most expensive, around $500-$600. That was back in 1957. It had the best sound, but it was so expensive," says Grove, who has since had reconstructive surgery to restore his hearing. "I was very hesitant to tell my caseworker, but they just picked up the bill. They didn't say anything." Then there was Grove's admission into the City College of New York in 1957. "I just walked in off the street," Grove says. "I had no papers, no nothing. I spoke broken English. A man sat me down and worked with me to figure out what I needed. On the basis of just what I said, I got admitted. It was a long journey with many steps," Grove says, sitting inside the modern headquarters of what is now a multinational, multibillion-dollar Silicon Valley giant. But it is those first steps and the kindness of strangers that stand out for Grove. He graduated at the head of his engineering class from the City College of New York in 1960 with a Bachelor of Chemical Engineering degree and received his Ph.D. from the University of California, Berkeley in 1963. Upon graduation, he joined the Research and Development Laboratory of Fairchild Semiconductor and became Assistant Director of Research and Development in 1967. In July 1968, Dr. Grove participated in the founding of Intel Corporation. He was Intel employee number four. In 1979 he was named its President, and in 1987 he was named Chief Executive Officer. In May 1997, he was named Chairman and CEO. Dr. Grove has written over 40 technical papers and holds several patents on semiconductor devices and technology. For six years he taught a graduate course in semiconductor device physics at the University of California, Berkeley. He currently is a lecturer at the Stanford University Graduate School of Business, teaching a course titled "Strategy and Action in the Information Processing Industry." His book, High Output Management (Random House, 1983) has been translated into 11 languages. His latest book, Only the Paranoid Survive, was published by Doubleday in September of 1996. In December, 1997, Dr. Grove was named Time Magazine’s "Man of the Year." Time Magazine writes, "TIME chooses as its 1997 Man of the Year Andrew Steven Grove, chairman and CEO of Intel, the person most responsible for the amazing growth in the power and innovative potential of microchips. His character traits are emblematic of this amazing century: a paranoia bred from his having been a refugee from the Nazis and then the Communists; an entrepreneurial optimism instilled as an immigrant to a land brimming with freedom and opportunity; and a sharpness tinged with arrogance that comes from being a brilliant mind on the front line of a revolution. Grove's mission is his product, and he shuns the philosophical mantle and higher callings often adopted by titans of an earlier era. Ask him to ruminate on issues like the role of technology in our society, and his pixie face contorts into a frozen smile with impatient eyes. "Technology happens," he clips. "It's not good, it's not bad. Is steel good or bad?" The steel in his own character comes through at such moments. He has a courageous passion alloyed with an engineer's analytic coldness, whether it be in battling his prostate cancer or in guiding Intel's death-defying climb to dominate the market for the world's most important product." Thanks to Dr. Grove, Intel has managed to thrive in the constantly changing computing industry. A strategic inflection point occurs when change is so powerful that it fundamentally alters the way business is done. Amy Wohl writes, "Andy Grove's Only the Paranoid Survive (Doubleday, New York, 1996) is all about identifying and acting on changes in the business environment. These inflection points represent big opportunities to those who can recognize and take advantage of them. The trick is (1) recognizing them and (2) acting on them at the right time. The recognizing is the hardest part -- you need to look for changes in who's the competition, who's your partner, and who understands the market. Grove understands how the ability to do this successfully in the past took Intel from the memory chip to the microprocessor business, but deflected an unsuccessful try at the RISC chip business. He even understands that thinking about the Internet -- and Internet appliances -- with an "old head" may be hard to do, but may represent an inflection point and Intel's next Big Opportunity." Ewan Sutherland wrote, "If the reader takes away only one lesson from Only the Paranoid Survive, it should be that: We must discipline ourselves to overcome our tendency to do too little too late. (page 133) This is what makes the crucial difference when a business reaches what Andy Grove calls a strategic inflexion point. The catch is that there is no way of knowing in advance whether it will turn down or up!" According to Grove, only the paranoid can avert calamity, because they are constantly looking over their shoulder to see who’s getting close and who’s doing something so radically different they might change the industry. It’s true that every company will be affected by a strategic inflection point (SIP), but one can see why only the paranoid might survive. Even Gordon Moore stated, "I suppose you never see the blow coming that ends up knocking you out. We (Intel) have a healthy case of paranoia. We see potential challenges behind every rock. It is always a concern that as we get big we can't move as fast as we did when we were smaller." Andy Grove writes, "I have a rule, one that was honed by more than thirty years in high tech. It is simple. "What can be done, will be done." Like a natural force, technology is impossible to hold back. It finds its way no matter what obstacles people put in its place. The beauty of this rule is that it can be used to look into the future. All we need to do is remember what already can be done: All information can be expressed digitally. All information can be transported in digital form. All information can be stored in digital form. If all this can be done, the rule says, it will be. What this means is that digital creation and display of information will predominate over other forms of communication -- telephony, broadcasting -- at the workplace as well as in our personal lives. Digital display will also subsume all other forms of storing information -- libraries (personal and public), professional records, photo albums. All this won't be easy: Schumpeter's creative destruction -- on Internet time. But we have no choice. We must face it. For whatever can be done, will be done." Andy Grove's opinions of JavaTM from Upside magazine. "Whenever we do something or other in Java, they (Microsoft) are unhappy with us. Platform independence is not necessarily a loss. Platform independence is a potential gain. The majority of the world's software has been written for platforms other than Intel-based computer systems. Yeah. I would love for that to be modernized and run on Intel microprocessors. Java is a means to do that. We can lose in Java, but we can also gain in Java. We could gain in a major way by unifying legacy systems and mainframe applications, and allowing them to move over more and more cycles to run on our microprocessors. That causes us to work with the likes of IBM. The portability of Windows NT has never been an issue for us. We are not sure Windows NT is portable."
Bill Gates: "A computer on every desktop and in every home running Microsoft software." "Information at your fingertips." How saddening it must be for Bill Gates to witness that his dream of information at your fingertips was delivered to the world, not by some version of Microsoft Windows, but via a coterie of the Internet, the World Wide Web, the Netscape Internet browser, and Sun Microsystems' Java technology.
The Magnet Effect. Jesse Berst invented this phrase to describe a new way of doing business. The old way: Build a product, then look for an audience. The new way: Build an audience, then look for more things to sell them. This phenomenon is occurring in the real world (Disneyland, Hard Rock Café, Niketown) but it is especially powerful in the virtual realm. Yahoo! and Netcenter didn't take off until they figured out the Magnet Effect. Until they realized they had to amass enough content to attract people. And enough services to keep them around.
Nicholas Petreley coined the first law of computer trade journalism: "No technology exists until Microsoft invents it." Michelle Murdock writes, "A few months back, many in the press sat up and took notice of thin clients for the first time when Microsoft bought WebTV. Now, fresh attention is being paid to Java and multiuser computing as Microsoft adds extensions to the language and goes for full-court press coverage with its promotions of Hydra and Windows-based terminals. Many reporters are giving a big thumbs up to the concepts of distributed objects and streamlined clients now that they bear the imprimatur (or in the case of Java, the sticky fingerprints) of Redmond's giant. Reporters who formerly invented imaginary "deficiencies" in the network-centric model are now crowing its praises (or Microsoft's version of it, anyway) with abandon."
Joy's Law: "Most of the bright people don't work for you--no matter who you are. You need a strategy that allows for innovation occurring elsewhere." In the past six years, Microsoft has quietly assembled 245 of the brightest researchers from around the globe, drawing from top universities and corporate rivals. Gates plans to hire another 400 researchers in the next three years, and build research centers around the world to complement the one in Redmond and another he and Myhrvold are setting up in Cambridge, England. The goal is as ambitious as you might expect. Bill Gates is building his brain trust and trying to lock up all the intellectual talent on the planet. A cerebral monopoly to match his Windows desktop personal computer operating system monopoly. Countless tenured professors, newly minted MBA students, and experts in their fields have followed the siren call of Microsoft's lucrative stock options. Says Paul Saffo, a director at the Institute for the Future in Menlo Park, Calif.: "Microsoft is like this intellectual roach motel--big brains go in and you don't see anything come out." Microsoft should pay heed to Joy's Law. Ideas that change the world aren't created by hordes, armies of programmers. They are conceived by visionary individuals like Bill Joy and the venerable Sun programmer James Gosling, the inventor of Java. Known as one of the great minds in software, the software industry's other famous Bill, William N. Joy is the co-founder and current Vice President for Research at Sun Microsystems, Inc. Joy, the creator of the C shell, was the principal designer of the University of California, Berkeley, version of the UNIX operating system whose networking protocols and implementations helped spawn the Internet, co-designed the SPARC microprocessor architecture, and developed the Network File System (NFS), which allows multi-vendor remote access. In the 1980's he spearheaded Sun's evangelism of the "open systems" model of computing which allows different groups to contribute design by making the specifications of its components freely available. He negotiated Sun's contract with Netscape for Java, drove the technical team that specified Java Applets and the release 1 version of the Java Programming Language, created and spearheaded Sun's recently released Jini technology project, and, recently, was named a director at Novell. Joy holds a B.S. in Electrical Engineering from the University of Michigan and a M.S. in Electrical Engineering and Computer Science from U.C. Berkeley. He holds a Lifetime Achievement Award from the USENIX Association and the Grace Murray Hopper Award from the Association for Computing Machinery. George Gilder writes, "According to Joy, the key to software innovation is smart programmers. Smart programmers are hundreds of times more productive than ordinary programmers. To the Justice Department, Microsoft's overwhelming OS market share and its teeming armies of programmers seem a barrier to entry for other software competitors. To Joy, Microsoft's size and dominance could become a barrier to entry for Microsoft, blocking it from the key new markets of the late 1990s." Microsoft's very size, dominance, and need for backward compatibility with MS-DOS and Windows 3.1 will always make its software desktop-centric and prevent it from dominating future markets like electronic commerce and online content. Microsoft software programs are too slow and complex. I hardly use one percent of the 60,000 features of Microsoft Word. Gilder writes, "Year after year, Joy lamented the prolix inelegance of the triumphant waves of Microsoft programs sweeping through the industry: "As we add more and more of these features to older systems," he said, "the complexity gets multiplicative. I have 10 different packages that interact in 10-to-the-10th different ways. I get all sorts of surprises, and yet because these things don't play together well the power is only additive. I get this feature and that feature but the combinations don't work. What I'd really like to see is a system (Java) where the complexity goes up in a linear way but the power goes up exponentially." In software, complexity has long been rising exponentially, while power has been rising additively." Joy states, "We've been getting a free ride with Moore's Law. We can write worse and worse software, and the machines just get faster and faster and cheaper and cheaper -- and they cover our tracks. The common programming languages of C and C++ basically beached us. These languages are like whales. Microsoft maintains this monstrous C program -- Windows NT -- that is built out of materials that are very difficult to work with. Windows NT 4.0 is 16.5 million lines of code that will never be debugged. It is infinitely complex. It is like having an elephant living in your apartment. The thing is just monstrous. Windows NT for consumers is an oxymoron because Windows NT is basically mainframe software with all these windows and very little architecture. It is a mess. Many people were happy with the cars they bought from Detroit before Honda came along. I'd like to think that Java is more like when the Japanese came along with quality cars. With Java-based programming, instead of having one big system with infinitely complex buggy software, we can get a federation of machines working together to solve problems. The individual components are simpler. Yeah, but those options are no better than what we had 15 years ago. Windows 98 is basically the same architecture that the Mac had in 1984. In a world of millions of devices, what you want to be able to do is send new bits of code and have them interlink. Ideally, the code would have flexible linkage -- flexible linkage is, in fact, the hardest bit of the job. C and all the programs related to it don't solve the programming problems of this world. They did not anticipate a world of millions of devices. Yes. Java is state of the art. I don't know how to do better. We have in mind that all these things -- bolts, shirts, watches, smart cards, rings, physics, chemistry, instructions -- all of this knowledge can get represented as components. I think Unix is a great system -- especially for running data centers -- because it is very mature, very reliable, very scalable. But when I want to go out and populate small devices, I think Java. That is why "Windows Everywhere" is such a flawed idea. You don't want the same user interface on a small device as on a desktop. They can't both be right. Java makes it easy for more people to write smaller programs that can be combined to make larger pieces. Think of it as letting you do Lego software. It has a network effect. Something like Windows or the Macintosh doesn't really have a network effect. They have pieces of code that must grow linearly into one big mess. Any program that is written by hundreds of programmers will inherently be too hard for most people to understand. It just has too many features. It is going to be so complicated. They are trying to make an operating system that gives you an infinite number of choices, that anticipates everything you might want to do. They are trying to plan the whole world. For our own good. But we are better off being a little more decentralized and a little less complicated. I view the Internet as the real opportunity, not the PC. The Internet is a bigger phenomenon in my mind. I want our stuff to run on PCs, but they are just a clumsy way to connect to the Net right now. The PC is not a particularly interesting device--it is pretty horrible from almost every vantage point. I guess they are cheap--you can buy them in parking lots across the country. But they don't represent a particularly great use of technology. Microsoft owns all the application categories, anyway, and they have no real interest in doing much innovation, so the whole PC space is just kind of boring. I mean, what's the big new thing now in PCs? I guess people are going to get Doom chips for Christmas, and we'll have a little animation. Big deal. The PC market has been really lucky--it has had a long smooth sail. But one of these days the dogs aren't going to eat the dog food. It happened already! The business/commercial customers didn't buy Pentium MMX's. Maybe Windows 98 will be the same way. People might just say, "Yeah it works, but I really don't see enough advantages to making the change." Windows 98 has lots of problems. It really doesn't run well on existing hardware. It's going to be a real nightmare. Maybe a lot of people will just decide they don't want it. There is no guarantee that Windows 98 is going to work better. That's the myth of progress. But things don't always get better. We spent five years designing Java, and the fact that it worked out is wonderful. With most projects where you sit down and give yourself five years' development, a language will come out of it that no one wants to use. Microsoft has got a huge code base, and Windows is never going to focus on the Internet; it's always going to be a disk system. Though they'll try and layer stuff on top of it, they just can't. Networking is an afterthought for Windows. I'm sick of mediocre software. It's a lot of fun trying to improve the state of the art."
Arthur's Law: "Do it right, or don't do it." Arthur van Hoff, a Dutch programming wizard, is Chief Technical Officer and Co-founder for Marimba, Inc. Prior to co-founding Marimba, he was a Senior Staff Engineer at JavaSoft until early 1996. During his tenure at JavaSoft, Mr. van Hoff authored the Java compiler, the beta version of the HotJava browser, and helped design much of the Java language and the Java applet API's. Mr. van Hoff earned a degree in computer science in Holland and a Masters degree at Strathclyde University, in Glasgow, Scotland. A noted speaker and author, he has won several awards for his contributions to the creation of Java, including PC Magazine's award for Technical Excellence. In late 1997, Mr. van Hoff was honored by Inter@ctive Week as one of the Top 25 Unsung Heroes of the Internet. van Hoff states, "Java was developed by James Gosling at Sun, and was designed for consumer devices, televisions, and stuff like that -- toasters and microwaves. When I joined, this was like two years into the life of Java -- they were using Java for interactive television at the time. That really didn't go very well. Interactive television wasn't taking off, and we decided, when the project kind of got canceled, and the interactive television business went on without Java, we were left with this great language. So with like 15 people left in a project, we sat down and wrote a business plan and said OK, well, let's stick it in a browser and let's get going. My biggest contribution at that point was to clean up the language to the point where it became really simple, you know, really well-defined. When I joined the project, it had gone through many iterations and there were features that were less than useful. And so we went through a very rigorous exercise of cleaning up the language, and I spent a lot of time with Bill Joy and James Gosling and Skye Steel in that process, and one of the things I did to make that sort of real was I rewrote the compiler in Java, and that's the compiler people are using nowadays. That was a really good exercise because I didn't want to have any sort of backwards compatibility with features we weren't going to make public. So I forced everybody to rewrite their code to the new standard, which was a really good exercise." James Gosling gets credited as the creator of Java for putting the right things into the development platform. Arthur van Hoff should get more credit for what was left out. "The more ways there are to do something, the more ways there are to do it wrong," van Hoff says - and that concept animated his approach to the embryonic language he worked on at Sun, then known as Oak. "We took out enumerators, macros, operator overloading and pointers," he adds. "If there was one way to do things right, we put that in." Van Hoff knew all too well the traps into which programmers were willing to fall. "My goal was to create a platform with no sharp edges, one you could thrash around in without hurting yourself," he says.
Amdahl's Law: "The speed of a system is determined by its slowest components." or a chain is only as strong as its weakest link. Gene Amdahl is the father of the mainframe computer and inventor of the parallel processor. The Web is "distance insensitive." In other words, the fact that sites are close to each other does not necessarily mean traffic will move between them faster than between far-apart sites. Singapore and Phnom Penh, for example, are about 750 miles apart. A data trip from one to the other and back takes 1,100 milliseconds. But from Singapore to Los Angeles, which are maybe a dozen times farther apart, it is just 400 milliseconds. This has a lot to do with capacity - the fattest "pipes" of access are to the top-level servers, the backbone computers of the Net, the ones that function like trunk lines, providing long-distance interconnections. If you and the computer hosting the page you want to reach are both near a big pipe, signals will travel between you very quickly, even if you're very far away from each other. But even if you are close together, the signal will travel slowly if you've got a small line connecting you. The "last mile" of the Internet is the slowest - the 56Kbps connection of your modem to your Internet service provider.
Netscape's Mike Homer's free test: "If someone gives you a product for free, will you use it and continue to use it?."
The Barksdale effect: Jim Barksdale, the former CEO of Netscape, gets his jollies figuring how to make large, complex systems deliver value to customers. He managed operations at FedEx. He ran McCaw Cellular, which became AT&T Wireless Systems. And he ran Netscape. Now he's rich and living in Aspen, Colorado, whence he aims to finance new companies that use technology to manage physical processes. He's betting that the next opportunity in the technological revolution will be in providing seamless, flawless fulfillment and distribution of products. He's invested in HomeGrocer.com, which is building a system to deliver groceries to your home; Tellme Networks, which is building a system to provide unique services via the telephone; and Respond.com, which is building a system to help Web users find all kinds of products and services. In this next e-wave, every company, big or small, new or old, will try to build physical systems that provide better service to customers. The Web has forever changed the way companies and customers (whether they be consumers or other businesses) buy and sell to each other, learn about each other, and communicate. The best companies will now build systems to pick and pack products that are shipped individually, keep track of those shipments, and make sure the stuff gets delivered to our houses or businesses. They'll integrate those systems with the manufacturing, distribution, and computer networks already in place. It is figuring out how to use technology to move stuff around efficiently.
The Law of Unintended Consequences: Sometimes, the most remarkable events in life are unintentional. Such is the case with Linux, which started out as a student's hobby but has quietly become in the past eight years one of the world's fastest-growing operating systems. Linux was started in 1991 by Linus Torvalds, then a student at Finland's University of Helsinki. Since then, tens of thousands of volunteer coders have debugged, fixed and enhanced Linux. But the story of Linux really begins in June 1979 at the Usenix meeting in Toronto, according to Peter Salus, editorial director at Specialized Systems Consultants Inc. in Seattle, which publishes the Linux Journal. "The lawyer from AT&T Corp. got up and announced the new pricing structure for AT&T Unix System V," Salus says. "The discounted educational fee was $7,500, and the full commercial fee was $40,000 per CPU. You can imagine what the feeling among the guys there was." One of those guys was Andrew Tanenbaum, a professor at a university in Amsterdam. "He couldn't ask a free university to pay that kind of money, but he wanted his students to work with Unix," Salus says. So Tanenbaum wrote Minix, a small version of Unix that would run on a minimally configured desktop system. Torvalds began using Minix after becoming frustrated with getting computer time on the university's Digital Equipment Corp. MicroVAX. But while it was a great teaching tool, Minix really wasn't a fully functioning operating system. It's what Torvalds did in response that was extraordinary. "He was interested in trying to see how an operating system worked by writing one, just as one would learn to ride a bicycle by falling off one. The result was a kernel that contained the basic Unix components - task-switching, a file system and device drivers. In other words, Linux Version 0.02.
Wriston's Law of Capital: Banking legend Walter Wriston was a great thinker who died in the year 2005 and also authored this book, The Twilight of Sovereignty. In the age of electronic money transfers, said Wriston, "Capital will always go where it's wlecome and stay where it's well treated." Capital is mobile, sapital is in play, and if you want some of it, you'd better treat it well. Evidently, the state of Maryland hasn't heard of Wriston's Law. It recently passed initiatives raising the minimum wage and requiring large employers such as Wal-Mart to pay higher employee health care insurance costs. Wriston's Law predicts that these decisions will be good for the economy ... neighboring Virginia's economy.
Sincerely,
Mark Kuharich
Join my free e-mail newsletter called the Software View by clicking here or by sending an e-mail to thesoftwareview-owner@west-point.org
other laws
"Top 10 System Administrator Truths." What are your top ten system administrator truths? We all know them already, but it's still fun re-telling them. Stuff like "90% of all hardware-related problems come from loose connectors", even though you already know it's true, may save you from replacing the "faulty" motherboard if you recall it at the right time.
Click here.