Dr. Jerry Pournelle

Email Me

Why not subscribe now?

Chaos Manor Subscribe Now

Useful Link(s)...


Hosting by

Powered by Apache

Computing At Chaos Manor:
October 30, 2007

The User's Column, October 2007
Column 327, part 3
Jerry Pournelle jerryp@jerrypournelle.com
Copyright 2007 Jerry E. Pournelle, Ph.D.

The October 29 Chaos Manor Reviews Mailbag has a letter and discussion of software patents. The Monday, October 29 editorial page of the Wall Street Journal has an article by AEI’s Claude Barfield and John E. Calfee entitled Congress’s Patent Mistakes that is worth your attention. [No link yet, but it will probably be on-line by the time you read this.]

Barfield and Calfee are anxious to defend the notion of software patents, and to discourage Congress from radical restructuring of patent law. Both points are worth thinking about. The patent system is important because without some protection of intellectual property, it is difficult to raise private capital for research and development. The alternative to private R&D is to set up a government bureaucracy for the purpose.

Government uses two methods. One is the National Science Foundation model: NSF doesn’t itself conduct research but it does fund grants. This works far better than one might expect a priori, but over time, NSF simply funds academic bureaucracies. Eventually the goal of NSF funded research laboratories and think tanks is to get more NSF grants, and grantsmanship becomes far more important than physics or biology. It does a fairly good job of allocating the resources that it gets, and indeed I have here and elsewhere argued that NSF is underfunded: that your NSF investment dollar is one of the best spent dollars in your enormous tax bill, and we’d be better off if we doubled the NSF budget. I don’t retract that statement.

On the other hand, I don’t want NSF budgets increased by a factor of a hundred, and I don’t want to see NSF funds as the only source of risk venture research capital. Indeed, the one thing NSF isn’t very good at is funding high-risk research, for the obvious reason that it’s too easy for a Congressional Committee to make fun of such grants. Senator Proxmire with his “Golden Fleece Award” for a one hundred thousand dollar grant to SETI (Search for Extra-Terrestrial Intelligence) comes to mind. If NSF dispensed all the research venture funds the result would not be good. For confirmation, study the USSR and its Academy of Sciences programs.

Another NSF-like government model for funding research is the X programs, and those were wildly successful. The X-1 gave us supersonic flight. The X-3 Stiletto gave us a ship that could take off from a runway, go supersonic, and return to where it started. The Stiletto was a beautiful ship but not very useful. It wasn’t maneuverable at all: but we learned from it. The result was the F-104 Starfighter, which dominated military airspace for twenty years. X projects worked so well that McNamara shut them down in the name of arms control: they generated far too much new military technology.

The problem is that neither NSF nor X Projects get much of the government R&D budget. Most of it goes to direct government bureaucracy.

NASA is an obvious example of the kind of return on investment such bureaucracies typically return: some great successes, but at enormous expenses. For what NASA has spent since Apollo we ought to be halfway to Alpha Centauri, but instead we are likely to lose the second Moon Race to the Chinese.

Incidentally, the late Robert Heinlein was fond of saying that it was inevitable that the human race would go to the planets (and indeed to the stars) but that nothing guaranteed that the language spoken in space would be English. It might well be Chinese.

If the history of the Cold War taught us anything, it was that the Soviet Central Control model doesn’t work very well. We used to say there were three enormous bureaucracies in this world: The Soviet Agriculture System; the US Education Establishment; and NASA. They all work about as you would expect them to.

We have learned, at great cost, that the engines of capitalism are the most efficient mechanisms we have for generating economic growth including the invention of new technology. The Framers may or may not have understood this at the Philadelphia Convention of 1787; they were mostly concerned with limiting the size of government, and they saw economic freedom as one way to accomplish this. They were mostly concerned that the power to grant monopolies was one of the central powers of the English Crown, and they wanted none of that.

Thus we got patent and copyright laws. The First Congress gave us some very simple ones. Successive Congresses have worked to make them longer and better.

The last time Congress tried to fix Patent Law, it created a Court of patent attorneys, which was naturally friendly to Patent Lawyers. The result was both predicted and predictable: more litigation, employment of more patent lawyers, and the requirement for startups to raise capital for legal rather than development expenses. Congress also tried to make the Patent Office self-supporting by charging fees for services, but later Congresses couldn’t resist the temptation to make it a cash cow. Making the Patent Office a profit center and the Patent Courts a fountain of incentives for litigation may or may not have been unintended consequences, but surely no one outside Congress (and the Trial Lawyers’ Association) wanted those results.

The intent of the Constitution in giving Congress power to enable the granting of patents and copyrights is pretty clear:

Congress shall have power . . . To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries.

The purpose is to enrich inventors, not lawyers. Astonishing.

Given that purpose, it is inconceivable that anyone in his right mind would devise the Patent and Copyright Laws we now have in place. The question is, would having this Congress – or any Congress that might actually be elected – start over from scratch effect any improvement? It’s a question worthy of several treatises. Indeed, there’s a good science fiction novel in there, and Niven and I may just deal with the issue.

It does remain an interesting question. If we really were to start over, what would be a good set of patent and copyright laws? Now, given that we don’t live in one world, fold in the International Conventions: do we remain part of them, or opt out? What happens when we do such things?

Do we count on the Members of Congress to know about such things and act wisely?

More on Comcast

I don’t favor a new bureaucracy to enforce Net Neutrality, but I confess the net service providers are working to give their enemies ammunition. The Comcast discussion continues in my advisory discussion group. Rich Heimlich came up with this:

Comcast fesses up to traffic delays

Oh boy.... First they confess to traffic delays.

Engadget link

Then it just keeps getting more interesting.

Verizon just showed up in our neighborhood to lay fiber to all the homes in my specific neighborhood so I'll be going 20/20 service shortly and kissing these guys good-bye. I cut them out of my TV money in 1994 and hope to cut them out of my ISP money for the first time. I've had them for 10 years now.

Bonus Ars Technica link on Comcast employee woes

Rich Heimlich

I’ve been collecting mail on this subject, and I’ll keep you informed as I know more. Meanwhile, one savvy computer user is overjoyed at the prospect of escaping from the grip of Comcast. Ain’t competition wonderful?

Winding Down

The movie of the month is The Kingdom. It’s the story of the FBI investigation of the terrorist murders of American workers in Saudi Arabia. It also gives a partial picture of life in The Kingdom of Saudi Arabia, and that’s why you go see the picture. It’s also a good action adventure film with very fast pacing. The actors won’t win Oscars because this isn’t the kind of movie that generates Oscar performances, but they are quite believable. This is well worth seeing. Recommended.

I can also recommend Dan in Real Life. It’s a chic flick worth a number of husband points. It’s also enjoyable.

The book of the month is Understanding Human History, by Michael H. Hart. This is a very Politically Incorrect book. Hart’s central thesis is that one must take evolution seriously, and evolution happens fairly quickly. His corollary is that history is best understood as a contest won by the most intelligent peoples; and that those who evolved in cold climates evolved the most intelligence. Living in cold climates requires skinning animals, making clothes, building fires; developing sewing instruments; cooperation and language. Those who develop such techniques have offspring. Those who don’t die off without children. In tropical climates other factors (including resistance to diseases) are more important for having children. Since different races evolved in different places, Hart explicitly ranks the races of man by intelligence. Needless to say this is not a popular view.

Hart explicitly discusses Jarred Diamond’s Guns, Germs, and Steel as an alternate theory of human history, and confronts Diamond directly. He also addresses the Flynn Effect.

Understanding Human History covers the entire sweep of human history in 480 pages including footnotes, bibliography, and index. The entire history of Rome from founding to Fall takes only fifteen pages. Needless to say this is not a book rich in details. It’s not an intellectual or military or even economic history of the world. It does attempt to assess the contributions of various civilizations and peoples to modern civilization. You may or may not agree with the assessments. It is certain that academia will disagree violently with just about every aspect of Hart’s thesis.

If you’re wondering what all the shouting about IQ and History is over, this is a readable way to find out what a strict evolutionist makes of the development of human civilization.

The other books of the month for me were technical works on naval warfare in medieval times; clearly not anything of great interest to most. The plot developments in my next novel, Mamelukes (part of the Janissaries series) involves Renaissance naval warfare, so I have been learning about the ships, the crews, and how it was done.

The computer book of the month is James Kalbach, Designing Web Navigation, O’Reilly. This has the subtitle “Optimizing the User Experience” and it pretty well lives up to its name. You wouldn’t suppose that there could be an entire book dedicated to designing web sites for easier navigation, but you’d be wrong. I wish I had this book in front of me when I was designing the Chaos Manor web site. It’s easy enough to read, and goes into considerable detail. I intend to use it for some emergency redesign of my web site. That may take a while because I have a number of books to get out the door, but this book will make it much easier. Recommended.