Dealers of Lightning 104
| Dealers of Lightning: Xerox PARC and the Dawn of the Computer Age | |
| author | Michael Hiltzik |
| pages | 448 |
| publisher | Harper Business |
| rating | 7.5 |
| reviewer | jnazario |
| ISBN | 0887309895 |
| summary | A worthwhile read for hackers and their managers, alike. |
PARC, Xerox's Palo Alto Research Center, was created after Xerox bought the research heavy SDS, (Scientific Data Systems), in the late 1960s. Almost immediately the seeds are being planted for a research arm of Xerox. Great minds are obtained in the process and in the same year the ARPANET becomes functional. The timing couldn't have been better.
What quickly emerges is the story of a large group of people, led by great minds and personalities like Bob Taylor and Charles Thacker. Strong of mind and personality, these are bright, visionary people who know what they want to do and how they will have to go about it. No hesitation, the bigger problems are things like How do you bring the right people together? And once there, what do they need?
Taylor brought together the best and brightest he could find, which is to say he got some of the best minds on the planet.
At every stage of the story, Hiltzik captures the mood, the emotion and the environment. In the early stages, he describes how this wondrous world was hatched out of determination and willpower. Xerox looked on during this early stage, perhaps a bit apprehensively, but also expectantly.
With a lot of freedom to tinker, a strong group of physicists and computer scientists were assembled and began building some of the greatest stuff in the world. By the time the 70s are over, Hiltzik's story is thick with the tension of researchers who design without products in mind and with management which attempts to see the value proposition in everything coming out of PARC.
Hiltzik's tour includes stories of how Ethernet was built, how the first personal computers were created and networked, how WYSIWYG applications emerged, and how so much else was created. He spends a lot of time discussing the invention of the laser printer, originally a dream of an idea by outcast physicist Gary Starkweather. Fighting sneers and doubt all along the way, he persisted and created the laser printer. But management only saw a threat to their core business of toner transfer copiers and the outrageous price of the device. However, they did patent the technology and that one invention alone paid for the entire PARC venture.
Several inventions seem so basic that you have to wonder how a company as apparently adept and bright as Xerox failed to capitalize on. Desktop publishing, which seems like a natural outgrowth of a document-processing company like Xerox, was born at PARC but discarded. Color printing as well was dismantled by Xerox. Other ventures, such as the personal computer and the Smalltalk language, seem obvious as unnatural fits for Xerox.
This is the crux of the book, and why it is such a valuable read for both engineers and management alike. For engineers, it is important to get a feel for how management operates, how they best appreciate ideas as marketable products. The same goes for managers, who often don't appreciate the value of research ideas; in this history, Hiltzik shows how that even when things were on the brink of falling apart for Xerox, management was able to continue its course, hoping the rest of the world would be content to buy only a handful of large-scale copiers.
Ultimately the book's epilogue gets it right, more or less. Xerox didn't fumble their future, though they did fail to understand the value of several of PARC's achievements. This is a hotly debated topic for many who feel that Xerox could have easily demanded hefty sums from Apple, IBM, and Microsoft or simply gone to market first with a mass-market personal computer.
The geek in me loves this book for so many reasons. Hiltzik's book is in the same spirit as The Soul of a New Machine and Fire in the Valley -- it's presented in a really thrilling way. The historian in me loves the modern history of the computer science community, and loves to see how the spirit of PARC has migrated to Apple, SGI, Microsoft, and beyond.
All in all I am very glad I read this book. It's inspirational, interesting, and of course relevant to what I do. A highly recommended book.
You can purchase Dealers of Lightning from bn.com. Slashdot welcomes readers' book reviews -- to see your own review here, read the book review guidelines, then visit the submission page.
Reviews oughta review... (Score:5, Insightful)
So why only a 7.5? What's missing?
Managers were already dumb (Score:5, Insightful)
His interest for networks started from there.
But he also told us how dumb the managers were already. Basically, he told us, researchers had white cards for a whole lot of things, and really invented beautiful things.
For example, the principle of a UI, where you could type and store a whole document and then print it later on was realized there, but a dumb manager refused the idea, claiming it was too complex: all the users want, he said, is a typewriter where you can validate your text one line, print it, and then validate the next one.
No doubt that if the Xerox manegers had been smarter, Xerox would be a far bigger company than it is today.
JB.
JB.
Xerox and a commercial PARC (Score:5, Insightful)
What is most impressive about this book is the way it doesn't condemn the Xerox execs out of hand for not taking up ideas, it slates them for destroying the atmosphere that created those ideas. The execs made a bundle of cash out of Xerox Parc, sure they could have made more but it more than paid for itself as it was.
Where the execs went wrong was because they _tried_ to make Parc more commercial, and more commercially driven. The power of PARC was that it started as basically a University within a corporation, and the corporation gained many valuable elements from it. As soon as they moved towards a more commercial model (Star et al) then the suits began to exert more control and the brains began to leave or get pissed off.
Don't slate Xerox for not capitalising on all of the ideas, slate Xerox for trying to capitalise on PARC and destroying it in the process.
Xerox PARC invented the majority of the important technology today, in the sense that they made it a reality even if others had thought of it first. Your PC has windowing because Apple saw PARC, your PC has ethernet because they needed to network computers, your printer works because PARC made it so.
PARC founded the modern computing world, but commercialism and the attempt to exploit the ideas are what destroyed it. PARC made Xerox HUGE amounts of cash, it was a desire (greed?) to get even more than led to the bright lights leaving.
These bright lights have gone on to bigger and better things, how Xerox must now think "if only".
Re:First Toddler Vomit Post! (Score:5, Insightful)
Cringely's documentary has some interesting interviews about how Xerox gave away a lot of their technologies. Many of the workers thought the management was insane.
While the failures of recognizing the loss of Xerox ideas seem obvious now, one must realize that nothing like that existed at the time. Developing any of those technologies would have involved huge risk and cost. The important thing is that certain people did realize it, and the technologies were marketed. Just not by Xerox.
Ideas are easy. Developing and marketing them takes real work.
Re:Sounds like... (Score:3, Insightful)
"Total ojectivity" is only going to be found in yourself, hopefully after reading enough conflicting viewpoints. Anyone with the fire to write about something is probably a deep fan or critic.
MS visionaries? (Score:5, Insightful)
MS has been a genious at settling the frontiers of computing with a sustainable and growing bussiness model but not in pioneering. In fact I cant think of any technology that ever came from MS that was not derivative. Nor can I even think of a slick integration of technologies (e.g. apple's forte), nor even a novel presentation of a new technology.
Maybe some MS folks can contadict me with a couple trivial examples. But look for a billion dollar company with 90% of the market their creative output is pathetic. Maybe some MS worker bees reading slashdot can say why. Does MS have a creative research dept? if so where's the products?
My dealings with PARC (Score:5, Insightful)
From my dealings with the people there, it was clear that they had the whole research and development thing down. They inspired their people to build things that were unbelievable. But the marketing and sales folks all came from the copier side of the business whenever they wanted to roll things out. Although Xerox folks were great people, they could not bridge the gap between their experience and the future. (See Clayton Christensen's Innovator's Dilemma.)
As time went on Xerox say more and more that they were not capturing the benefits of the PARC developed technology and got desperate. So all good things come to an end.
Re:Xerox and a commercial PARC (Score:4, Insightful)
Re:MS visionaries? (Score:3, Insightful)
Just because Microsoft's visionaries are in management and not technology doesn't make them any less visionary. Now, whether it makes them law-abiding, that's a different question.
That's not what went wrong. (Score:5, Insightful)
First, PARC wasn't that secretive. I saw the first batch of Altos in 1975, long before Jobs. Alan Kay described the first Ethernet as "an Alohanet with a captive ether", which we (being computer design students) all got. We were given an early Smalltalk demo. In the 1980s, I programmed an Alto in Mesa. I've been there many times, and met many of the PARC people over the years. Almost went to work there once. So I know something about this.
The blind spot at PARC was that they, and Xerox management in Rochester, thought that stuff should just work. They visualized boxes that you plugged together in an office environment and that didn't need any on-site expertise to operate. This made sense, because that's what other office products looked like back then. Xerox copiers of the 1970s, while incredibly complicated internally, hid all that complexity; only the Xerox service people had to understand what went on inside.
Early word processors were as simple as possible from the user point of view. Wang was the leader in "shared-logic word processors", which were dedicated time-sharing systems for word processing. A Wang-equippped office had a computer in a box the size of a filing cabinet, running nothing but Wang software and maintained by service people who came when called. The users didn't think of it as a computer.
PARC tried to replicate this with the Xerox Star, a networkable box which contained an suite of office programs. It was expensive, but good. By design, it was not user-programmable.
What PARC didn't see was that the future of computing involved cheap machines running crappy software. The future was CP/M on green screens tied to daisy-wheel printers interconnected with 300 baud modems. The future was DOS, WordStar, and VisiCalc. The future crashed a lot. People at PARC regarded this with horror.
Remember, the original IBM PC was considered a joke by everybody in computer science. It was clear what you wanted - a real CPU like a Motorola 68000, with an MMU and some kind of real operating system, with at least "a MIP, a megapixel, and a megabyte". The Apple Lisa (not the Mac) reflected those goals.
But it just couldn't be done cheaply enough. The hardware wasn't really there to do it right until the late 1980s, when Motorola released the 68030 and Intel released the 386. By then, mainstream computing was locked into the model we all love to hate.
It was all a cost problem. The original Altos cost about $50K each. Xerox Star machines were in the $20K range. UNIX workstations used to be in the $10-20K range (some still are). But PCs launched at $2-3K, and went down from there. And that's why things went the way they did. Not because Xerox blew it. But because it was just too early to do it right.
Re:Hal: What will happen to us, Dave? (Score:4, Insightful)
Let's look at it:
Xerox PARC
Everybody Else Combined
But I guess we "real world" types didn't find any use for the Xerox PARC stuff...
Re:MS visionaries? (Score:1, Insightful)
in the computer industry who can claim
to be innovative and certainly Microsoft
haven't produced anything revolutionary -
rather their skill is in taking existing
ideas and integrating them, making incremental
improvements and, most importantly,
making them available cheaply and on
commodity hardware. Many comments on
slashdot complain about an 'MS tax'
but they forget that when UNIX ruled
the roost the OS might have been
cheap but the choice in the mini or
workstation market was very limited.