Brownian Musings

Copyright and the Consitituion

2011-09-15T12:44:00.000-07:00

Let me start by making it very clear that I am no constitutional scholar. I am the furthermost thing from a lawyer, as I am an engineer. So please forgive me if I go too far in applying plain common sense and a literal Interpretation of words and grammar to this topic.

First of all, let's start with the United States Constitution. Article I, Section 8, Clause 8 of the United States Constitution (often referred to as the "Copyright Clause") forms the basis on which Congress established the U.S. Patent and Copyright law. That clause is as follows:

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.

I have two questions concerning the constitution and copyright.

Question 1

How does this clause establish that Authors (and Inventors for that matter) can sell their rights?

And even if we grant that Authors and Inventors can sell their rights to their writings and discoveries, how does that grant congress the power to allow companies and individuals who are not authors nor inventors to continue to sell these exclusive rights?

This is important when you consider how Apple, Microsoft and others just purchased Nortel's patents. Nortel was assigned these rights by actual inventors (mostly their employees, one presumes). But now Nortel is selling them to Apple and Microsoft. In the end, these "Discoveries" are controlled by parties that had nothing to do with the development of these Discoveries. These parties can now deny other companies access to these Discoveries, or extract payments for access to these Discoveries, and all without having done anything to "promote progress".

Ignoring the fact that there is little way to justify this state of affairs in the context of promoting progress, it remains that this clause seems to have been expanded to mean:

To promote the Progress of Science and useful Arts, Exclusive Rights to Writings and Discoveries may be created by Authors and Inventors. These rights can then be freely assigned and traded to other individuals and corporations without limitation.

Now, I likely didn't do the best job with capturing what patent and copyright law does. But without question the reality of copyright and patent law requires a huge stretch of the Copyright Clause.

Question 2

How can a copyright term of an Author's life + 70 years be constitutional?

Here the law is quite clearly creating something beyond the Copyright clause. By definition an Author is *not* granted exclusive rights to their writings by current copyright law, because by law the exclusive rights must extend past the death of the Author!

It would seem reasonable to assume that the "limited Times" referred to in the constitution should at least be possible for an Author to possess. The assumption that these rights might pass from the Author to other parties isn't the same as a law that requires the rights to pass to other parties. In other words, it seems to me that the Copyright Clause implies that it should be possible for an Author to possess the rights granted. If there is no way that "limited Times" of the law can be granted "to Authors" (because, after all, they must be dead for 70 years before they expire) then surely there is a conflict between the constitution and the law here.

Right?

The Death of the Business of Journalism

2011-08-18T11:51:00.000-07:00

(I would like to post a link to the following comment, but can't figure out how:)

Jeff Jarvis - 9/18/2011 7:04 AM -
Preparing for my first class of the season. This is the session in which I brainwash all the incoming students. I used to go through a litany of numbers about the fall of the business of journalism. I figure I no longer need to do that. Stipulated, your honor. Today I want to talk about how they must reconsider -- though not necessarily reject -- all assumptions about news: business, form, relationship....

Just curious if we are seeing the fall of the business of journalism, or just change back into a model of journalism that once existed in the past? Consider that Benjamin Franklin could make a huge stir in his day by inserting his work into a paper published by his brother.... With the advent of technology over the last 75 years, no such small scale operation could make such an impact. But the technology didn't stop there, and now again a blogger can again make a huge impact by posting their work on their own site.

I used to make the same kinds of observations about privacy. Nobody in a small town is surprised when people they don't really know still know about their business. With the move to cities (mostly due to the automation of food production, i.e. technology) people became accustom to the fact that there are too many people in a city for anyone to track what they personally are doing. Then technology kept rolling, and suddenly we can automate the collection of information about what people buy, where people go, how much they spend, and what catches their interest. Once upon a time everyone knew all about what you did because you lived in a small town. Now everyone can know what you do even if you live in a city with millions of other people.

Technology isn't so much changing the rules as it is rolling back the clock.

I know this is simplistic, but there is a truth here. The production and sharing of information was once a personal thing, and within small groups of people and without expensive technology, it has a certain empowering and at the same time confining nature. When we farmed out the production and sharing of information to corporations (that could afford the technology that individuals could not), the production and sharing of information took on a particular nature that was both impersonal, and maybe more manageable as a business. Then as the technology makes the process once again a personal thing, we find that the production and sharing of information becomes increasingly unmanageable as a business.

In other words, as individuals become more empowered to produce and share information, the competition to do so increases, and the ability of business to manage the production and sharing of information declines. But it isn't because people are doing something fundamentally evil by "breaking the rules" when they share information. It is because the process by its very nature is becoming more personal.

The business of journalism is going to be more diverse, with more competition, and with more attention to the fact that information is now both produced at a personal level, and shared at a personal level.

Intellectual Property

2011-08-15T15:15:00.000-07:00

You gotta love http://mimiandeunice.com. Here is one I saw today:

Open Source is a funny thing. Clearly it is a total ripoff for someone to gain ownership of the contributions of other individuals and other companies who are doing the real work that makes a Open Source project successful!! And everyday, another Open Source project suckers more individuals and companies into donating their work to make yet another Open Source owner rich!

But is this so?

As a project/product/application is developed, hard problems are encountered. And sometimes a really nifty solution is recognized and implemented. Then what does the funding group, the company/organization/agency do?

Typically the company/organization/agency will grab said "nifty solution" (which isn't part of their core goals/aims/product), and stamp it as "Our Intellectual Property."

This does not increase the value of the company/organization/agency, but rather serves to lock up the solution.

99.75 percent of the time (according to an Internet source), "Nifty Solutions" built to solve particular problems within a project get frozen in place. They don't get reused by other parts of the organization, because the other parts of the organization don't want to have to come to understand that code. They cannot be modified because there isn't any budget to support any problems that might arise from modifying the code. And these "Nifty Solutions" can't be used outside the organization because they are stamped "Our Intellectual Property".

So other "Nifty Solutions" are built. Again. And Again frozen in place. Over and over. Sometimes new "Nifty Solutions" are built just to replace past "Nifty Solutions" nobody understands anymore.

So how to save the "Nifty Solutions" from their frozen fate?

Open Source.

The power of open source isn't so much that someone can own it and reap the financial benefits of owning/controlling all of the contributions made, though to some extent they can and do. The real power is that as a contributor, an individual or company can bend an Open Source project to better address the problems they face in their own applications/products/programs. And then later they can use the better versions of the project in other parts of the organization. They can hire people that are already trained to understand the Open Source project.

As users of Open Source, we the developers get to benefit from updates to solutions, which we may or may not integrate into our projects as we see fit. We can be more productive. When we can deliver to our "Over Lords" (whoever actually pay us) "Free" performance improvements and "Free" new features, it makes us developers look good. We get paid better because our knowledge makes us more valuable.

Those of us that use Open Source as we move through their career can KEEP our knowledge, and can continue to apply our knowledge TO OTHER projects. Even if we change the company/organization/agency that we work for.

Contributors and users of Open Source gain by using and contributing to Open Source. They have better choices of jobs (i.e. they know Linux, Hibernate, Tomcat, Eclipse, Maven, etc. etc.) Th work of contributors does not get frozen as "Our Intellectual Property" by an organization that is never going to leverage it further. Contributors thus don't have to build new versions of the same solutions over and over to the same problems and watch their solutions get frozen in place, yet again. Once contributed into Open Source, a solution can be leveraged over and over, can be extended, and can even be rewritten when the problem becomes more fully understood.

And the funny thing is that most Open Source projects do not result in huge financial windfalls for the owners of the projects. They may hope for those windfalls, but until they come they are the ones scheduling work, testing, the releases, feature development, etc. The winners day in and day out are the users and contributors to these projects who get to see their work used and applied and built upon.

Nobody wants to see their code frozen in place. Write a good solution within a corporate or government context, on a project, and watch it Freeze, even as you type. This is because the suits at the executive levels can't understand why they should not claim any and every "Nifty Solution" as their own 'Intellectual Property".

And at the same executive management types will not understand why their "Intellectual Property" does not scale, does not grow, cannot be used by other parts of their organization, and does not generate those golden "license fees". But wait! They can avoid costs by using Open Source project X! They can agree to that.

And that is why so many developers contribute to Open Source.

Stimulating the Economy

2011-07-07T06:44:00.000-07:00

There are not many things we can do that reduce the government's size, reduce liability of businesses, lower the costs of goods to consumers, increase competition, stimulate research and development, increase investment in businesses, and create jobs.

All without the government spending a single extra dime. In fact, we are talking about cutting government here.

Eliminate patents. Altogether, and across the board. If it is good for banks, why isn't it good for high tech? For entertainment? For drug companies?

Here are my observations:

1) Without patents, we would see the creation of more new companies. New companies could immediately go into businesses that today which they currently avoid because of the risk of being sued for patent infringement by hosts of Large, established businesses. AMD, ADM, Exon, Ford, Sony, Microsoft, Apple, IBM, CISCO, and thousands of other companies in every industry use patents to limit competition and delay new technologies. Their advantage as established businesses is in their product line and their execution of their business. Their disadvantage is that they don't want to disrupt the same, and patents allow them to protect the status quo.

2) Healthcare costs would be reduced. Drug costs would plummet. Drug companies would have to compete with generics from day one. Research would open up where key patents prevent innovation. Recent audits of Drug companies how shown that they have vastly inflated the costs of developing new drugs. Drug companies are raking in the dough, while we foot the bill. Mostly the tax payers pay for research and drug development (by way of government grants and subsidies) and Big Pharma profits. No patents, no more free ride.

3) Smart phones would become legal. You didn't know your smart phone was illegal? Well given that to be legal you would have to have a license to a host of patents held by a huge range of companies, such as Apple, IBM, HP, Google, Microsoft, Motorola, HTC, Nokia, Sony, Samsung, CISCO, AT&T, Verizon, Sprint, AMD, Intel, and literally dozens if not hundreds of other companies. Nobody has such rights, and in fact many of the above companies (and others in the field) are suing each other over their various smart phone patents. Obviously nobody has the rights to produce a smart phone. We have smart phone technology despite patents, not because of them.

4) The elimination of patents would lower product costs. License fees for intellectual property are a drain on our economy and a drain on innovation, and a drain on companies trying to produce products, and thus a drain out of YOUR pocket. Prices for many products would immediately drop without the risk of patent lawsuits and license overheads. Money paid on licenses fees is money that is diverted from hiring people, from investing in one's business, and increases the cost of doing business.

5) Innovations would come to market faster, not slower. Why? Because a patent never gives you the right to make a product. A patent can ONLY give you the right to keep someone else from making a product. That is because every modern product is covered by hundreds if not thousands of patents, mostly held by big corporations. A single patent can force a product out of the market, but just because you have a patent does not prevent hundreds of other companies from blocking your product with their patents.

6) Companies would create more jobs. Want companies to INVEST into their business? Then take away the government imposed barriers to competition! No company can afford to wait until next year to advance their business if another business can spring up and take away their customers! Patents allow companies to keep their prices high and their bank accounts full as they meter out the investments into their operations. Competition in the actual execution of business is what we need to get companies off the dime.

Does it make any difference to vote Democrat over Republican?

2010-09-17T15:43:00.000-07:00

"IF you cannot see any differences between the GOP and Democratic party, you have absolutely no power of enlightenment, critical thinking or observation. "

The difference between the parties is like the difference between cutting your wrist with a dull and rusty knife, or a dull and rusty spoon. Absolutely there are differences by definition, but there are no differences in their actual functionality when applied to the task at hand.

Pick some issues, like abortion, education, stem cells, the war, health care. The parties have some defined differences in their positions. But these are issues raised to gain office without any significant difference to the people. Status quo reigns for abortion and education. Corporate interests win for everything else.

As a result, I am sorry to say, I cannot see any real difference between the Democrats and the Republicans. Both are in the pockets of special interests, even if those special interests do vary a little bit. So we can define a few differences. Can we define the similarities?

Do both parties submit legislation mostly written by special Interests? YES
Do both parties pander to superficial political issues so as to avoid discussing Debt, Corporate Welfare, our continued loss of rights and privacy, the need for corporate trust busting, etc? YES
Do both parties lie and slander their opponents in their political ads? YES
Do both parties misrepresent the issues facing the government to gain political advantage? YES
Do both parties vote to promote their party's power over voting in the interest of the people? YES
Do both parties persecute party members that vote for their constituents interests over their party's interests? YES
Do both parties slant laws and districts to promote their party's interests over the interests of fair representation of the views of the people in their elections? YES
Do both parties stack the laws in favor of the parties over Independents and third parties? YES
Do both parties block campaign reform that takes away the advantage of money in elections? YES

Is any one candidate guilty of every one of the above offenses to the people? No. But this is EXACTLY why voting for parties instead of voting for candidates is so horribly wrong. Because both of the major parties ARE guilty in EVERY election of EVERYONE of these offenses against the people. And just because parties cannot completely corrupt every candidate for office does not mean they do not succeed in corrupting nearly all elected officials of their party over time.

In fact, nearly everything the parties agree on from session to congressional session is designed to maintain the status quo and favor corporate influence in government (on which the status quo is built).

I am no different than any true blue party believer, at least in a way. I want to think that voting one party or the other is the way to advance my own welfare too. However the evidence that I have no such choice is way too overwhelming to deny.

Senator Kay Bailey Hutchison Believes Monopolies are the Answer

2010-09-02T11:54:00.000-07:00

I wrote to my Senator, Kay Bailey Hutchison, asking that she withdraw her support from Bill S 3828 IS, the so called ‘Innovative Design Protection and Piracy Prevention Act’. Here is what she said:

Dear Friend:

Thank you for contacting me regarding copyright protection. I welcome your thoughts and comments on this issue.

Copyright protection has been central to America's prosperity and job creation. Movies, books, computer software, television, photography and music are among our unique American products and some of our most successful exports. United States industries depending on copyright protection employ nearly 4 million workers and produce over $65 billion of our exports - more than agriculture and automobile manufacturing.

Protecting content in a high-technology age is a new and daunting problem, and copyright protection is an important challenge as the broadband revolution offers even more far-reaching possibilities and opportunities. With new speed and interactivity, the entire store of movies, music, books, television and raw knowledge can be made widely available. I believe copyright protection is a foundation of innovation, and copyright law should work to ultimately protect the best interests of consumers. Intellectual property is the creative core of the information age. I will keep your views in mind should the Senate consider legislation addressing this issue.

I appreciate hearing from you and hope you will not hesitate to keep in touch on any issue of concern to you.

Sincerely,

Kay Bailey Hutchison

United States Senator

http://hutchison.senate.gov

Now the CBO looked into the kind of statistics she is quoting here, and couldn't find any credible source for these numbers. Basically, she is quoting the content industry here. What, is she really just a stooge? It doesn't matter what a special interest says, a Senator SHOULD look at the numbers and demand that they be justified by facts.

Here is the link to the CBO's take on copyright. THEY couldn't find a source for these numbers. THEY found that these numbers do not take into account harm to the market from strict copyright enforcement. THEY found that these numbers do not take into account benefits to the market (More hardware, more software, more Internet access).

On one point she is right. As computers become more powerful, the old way of doing things is going to become increasingly impossible to enforce. If movies can be shared on a whim, how are you going to stop sharing? If information can be shared on a whim, how are you going to stop learning? And exactly what is to be gained anyway from protecting huge, expensive, massive corporations whose only value add to a digital product is that they can distribute it.... Wait the Internet can do that! Maybe they can provide studios... Wait! The hardware to do that is dirt cheap and can be done by the artists.... That they can control the markets? Wait! Why should we let them?

The content industry could make content easily available TODAY for a REASONABLE price. The reason they do not is that they want HIGH prices, not FAIR prices. Piracy isn't driven by people's desire to steal content, but by a lack of reasonable ways to access content legally, easily, and safely. Hollywood's income set records last year. More is being spent on music then ever before (even if less is being spent on recordings). The solution to the problem of high technology is competition and free markets. It is not fighting useless battles to regulate technology because we do not want to let people to have access to the products and services it allows.
All of this is fine, but had NOTHING to do with the email I sent Senator Kay Bailey Hutchison.

The Fashion industry in the US is the largest and most powerful in the world WITHOUT this bill. The only thing that copyright can do is allow the Fashion Industry to be dominated by large corporations. I hate to be cynical here, but the only benefit I can see to Senator Kay Bailey Hutchison is that large corporations give more money to campaigns.

Obviously emailing her directly didn't do any good, because she didn't even try to address the need for this bill. This is just a form letter telling me why she votes for the content industry, i.e. she takes them on their word at best, or supports them for their campaign contributions at worst.

I can understand why I just get a form letter when I write her. She hasn't time to actually read what her constituents have to say. But she ought to at LEAST generate a form letter that is on topic for a bill she is sponsoring.

Maybe public pressure can do some good?

Ha ha! just kidding!

Credit Card Reform....

2010-01-15T12:10:00.000-08:00

A consumer's payment history and average account balance is a typical justification for raising interest rates on credit cards and debt.

But at the same time, these companies make their most profits from customers who pay more fees, maintain higher balances, and pay higher interest rates.

The justification generally given for raising rates on such customers is that they represent greater risk than more disciplined customers. I would like to challenge that assumption.

The fact is, over time, the sum of interest and fees paid to the creditor reduce that creditor's risks. For example, a history of paying a bill 5 days late each month (along with the associated late fee) adds to the creditor's return on the said account.

Clearly over time, the creditor can document the return on a given account. And clearly as the return increases on an account, the risk of that account decreases.

So take as an example an account with an average balance of $1000 over five years. If this account has an interest rate of 20 percent, then after five years the return on that account has been $1000. Let's throw in an average of 3 fees of 30 dollars per year (overlimit fees, late payment fees, etc.). Assuming that the account is in good standing at the end of the five years, the return to the creditor is $1450 ($1000 in interest + $90 per year in fees).

The observation is that credit card companies will almost universally increase the fees on the customer with the fees while at the same time make nearly 50 percent more revenue on the same customer!

The simple reform I would suggest is that credit card companies be required to calculate the Return Total (RT) on every account, just as they are required to calculate the APR on every account today.

RT = total of all Interest + all fees paid by the consumer.

Then, quite simply, the interest rates of any balance < RT be limited.

What limit would be fair? 5 percent over prime for an unsecured debt? 10 percent? But surely interest rates like 29 percent or more are absolutely unjustified. And once a consumer has paid such rates over time, what does such a consumer gain? Shouldn't there be some return to a customer who pays such rates over time?

The obvious justification of limiting interest rates in this way is the observation what matters to the profit of the credit card company is the ratio of the money they are paid to the money they have loaned. While credit history, credit scores, driving records, and payment histories might be reasonable tools by which to set an initial interest rate on a loan, what really matters over the life of an account is the money it makes for the company.

A customer with a poor credit rating and poor payment record may be suffering from a host of influences (illness, tragedy, depression, volatile income, etc.) that do not necessarily reflect on their honesty or ability/willingness to pay.

This is a simple and objective means of insuring a customer gets credit for their actual worth to a company (i.e. their contribution to the company's profit) rather than allowing credit card companies to gouge people making an honest effort to pay their debts.

Copyright and the MPAA

2009-11-20T18:34:00.000-08:00

Let’s suppose I write a really good joke. It is an absolute knee slapper. Now technically it is copyrighted. But people who hear it (since it is short and sweet) simply steal it, telling it to other people. Who tell other people. Who steal it and tell other people.
I want the phone lines keyed to my joke. I want computers to detect the joke being told, and emit annoying fog horn sounds to prevent the theft of my property. I want iPhones and other smart phones to monitor all conversations (not just phone calls!) and emit ring tones to prevent the spread of my joke should their microphones detect someone telling my joke within the range of their microphones.
Unreasonable?
The fact of the matter is that there isn’t any way to stop the spread of copyright material through a population when it is as trivial as a good joke. And as technology advances, it is going to be easier to transfer a movie or a song than to tell a joke. So we can either make all communications illegal. Or give this up.
Hollywood is going to be just fine. Their revenue is growing, and true counterfeiters of media are dying away. There just isn’t any money in counterfeiting something you can simply download for free. The fact is fans of a good movie or actor or director are going to go to the Theaters ($), buy the official DVDs or Bluerays ($$), buy shirts and posters ($$$), go to concerts ($$$$), and who knows what else. People would rather pay a REASONABLE price for an easy view of a movie that fuss with a Torrent that might not work.
Quit whining that the world isn’t constructed to your (MPAA’s) preferred specifications! Quit trying to get the government to pass laws to bend physics and the advance of technology into something that exactly models the past! Instead, make a good product and accept the pay for a day’s work and give up the idea that your members should be paid forever for anything they ever did.

Disruptive Change

2009-11-11T17:03:00.000-08:00

Matthew Komorowski produced this wonderful graph of the rise of hard drive capacity per dollar cost over the last few decades. Matthew derived the following fit to his data:

Now I just had to grab a spreadsheet and plug in some numbers. Here is some interesting predictions that result. Now nobody is saying that this equation will certainly fit future data. But the argument that it will is pretty hard to ignore. So for the moment, let's not argue that point, and assume that the equation will hold for another 20 years, more or less.
First of all, how much does hard disk storage cost? Well, unlike bread or gold or lumber, disk storage prices are on a continual slide downwards. That is what Matthew's equation given above tells us. So how drastic is this fall in hard disk storage costs? Consider:

In 1989 the cost of 1KB fell below a penny
In 2001 the cost of 1MB fell below a penny
In 2013 we can expect the cost of a GB to fall below a penny
In 2025 we can expect the cost of a TB to fall below a penny

I bought a 1 Gigabyte drive somewhere around 1992 for my company for $5000. Today I can buy 1000 times that same storage for about $70. (EDIT: was $250, but that was a very dated price!) Amazing!

Still, KB this and GB that doesn't necessarily mean anything to our intuition. So the question becomes, how might we relate these drives to something more every day, something tangible. Talking about the number of phone books that can be stored on a drive just doesn't cut it anymore, so how about considering HD Video?

Edit: The use of HD Video as an example in this exercise should not be taken as prediction on video formats to be used in the future! Likewise, our projection of the use of HD Video in the past is an intentional anachronism... We are just using HD Video as a yardstick here, nothing more.

Well, if we assume a hour of HD Video takes about 4 GB, we can divide the "Price Time Line" by how many seconds, minutes, etc. of HD Video a $100 hard drive can hold. So what does this "Time Line" look like?

In 1981, your $100 got you about a second of HD Video
By 1993, your $100 could cover about a minute of HD Video
By 2000, your $100 got you about an hour and 15 minutes of HD Video
By 2006, your $100 could buy you 1 and 2/3 days of HD Video

And here is where things get really interesting...

By 2012, your $100 will be able to buy you nearly 2 months of HD Video
By 2016, the same $100 will be able to buy storage for almost 1 1/2 years of HD Video
By 2020, your $100 will be able to buy you 1 1/2 decades of HD Video
By 2024, your $100 will buy you 1.45 Centuries of HD Video

And if this trend holds for less than 20 years ....
By 2028 you will be able to buy space to store 1450 years of HD Video for $100...
That has GOTTA blow your mind.
Here is my spreadsheet. The shaded years are years for which we have data and fit the curve described above:

Self Organization

2006-12-12T09:52:00.000-08:00

Biological Systems are different in a fundamental way from non-living systems. This shouldn't be too much of a surprise, as one system is a living system, and non-living systems are, well, non-living.

But from a Self Organization point of view, living things are very different. Their behaviors may follow simple rules that give rise to structures that look like structures built by non-living systems. For example, one may find parallels between geological structures on Earth with those on the Moon. However, geological structures on Earth are often found to be hugely influenced by living things, while the moon is a decidedly un-living structure.

Living things have behavior which can evolve. That is to say that the ability of an organism to pass behavior on to future generations is based on an organism's ability to live and reproduce. If an organism does so "on the cheap" by aggressiveness, then such behavior will increase. Perhaps it will increase until it does not provide a significant advantage. Then organisms that invest in cooperation may displace aggressive organisms by virtue of numbers.

Such dynamics do not exist in geological formations. So while self organization may occur, it is not influenced by the progression of "parent" geological formations. It simply is what it is.

To take this a step further, if a parent passes on a behavior to its children, it must do so via some transfer of information. Living systems can transfer information by 1) DNA, 2) changes to the environment, and 3) modeling. The DNA of an organism represents the genetic transfer of information. Environmental changes would include the writing of books, building of cities and roads, etc. Teaching would involve the direct modeling of behavior to the young.

All the way up to this point, the discussion has been about how organisms transfer behavior to other organisms. But we are entering a new phase. We now have the ability to transfer behavior not just to other organisms, but to our tools. And better yet, our tools are then able to transfer that behavior to other tools and to mix and match behaviors from many sources.

Of course, this magical tool is the computer.

Computer systems allow information to be defined and transferred to other systems without involving people. If a sales tax changes, then without computers we would have to communicate that change to every person who computes sales tax. With computers, we merely have to communicate that change to every computer system. In both cases, the behavior (i.e. what sales tax to charge) is changed, but in the latter the change does not require a transfer of knowledge to the person, but to the tool.

Computer systems can be considered organisms from the point of view of Self Organization. As such, they are configured mostly through direct informational transfer. Their DNA becomes the set of installation files used. Then like Bacteria, computers can pick up new functionality and drop functionality as they encounter applications, and/or programs are un-installed.

It is difficult claim that computers can learn through the same mechanisms that we use. Digital transfer, while very like the exchange of DNA, isn’t really the same thing. Computers are increasingly exchanging digital information, very little of which constitute exchanges of functionality. Exchanges of information always do one of three things: 1) Post events, 2) Provide Data, 3) Provide Functionality, or 4) Provide nothing. While computers do clearly exchange information, it is difficult to map such exchanges to biological processes in a meaningful way.

This topic will surely be worth musing on in the future.

The Halting Problem

2006-12-06T17:33:00.000-08:00

Alan Turing gave a wonderful proof in his 1936 paper proving that no general algorithm exists which can answer halting question for a given TM in a finite number of steps. The Halting Problem is merely one where we ask if a particular TM ever halts.

Turing first demonstrates that you can simulate any TM with a UTM. Then he describes the behavior of such a general algorithm given itself as an input. This construction will clearly halt, since we are assuming it can answer the halting question in a finite number of steps. Then we modify it so it behaves like a 2 year old child, i.e. after it has decided that it will halt, it then doesn’t. Or if it decides it doesn’t halt, it then it does.

This proof as a programmer has always bothered me. Code does what it does. If you have an algorithm, it executes to the end. Something about this proof was fishy, because even if you limit the behavior to looking for simple deterministic failures, it still doesn’t work.

The problem is that Turing’s Proof doesn’t work because it cannot be constructed.

Observation 1: No modification to a program results in the same program.
Observation 2: Even very safe, fool proof, easy, not a problem changes can break a program.
Observation 3: If you can’t really build what Turing describes, then you can toss the proof as well.

Approach A:

So with great care, let us see if we can really build Turing’s example assuming we have some way of examining a program and determining if it will halt.

Assume A is a machine that can take an input B and answer if B halts. To build a TM to run A, you would build a tape: AB. A is the algorithm used to determine if B halts, and B is a UTM that executes B in simulation. Some how or other we are going to assume that A can look at B and determine if B halts.

By Turing, the proof against the existence of A is that you could then modify A building an A’ so that if B halts, A’ does not, and if B does not halt, A’ does. Then, given itself as the input, the result is by definition wrong. A’ cannot possibly answer if it halts given itself as an input, for it will always do the opposite.

The problem is in the phrase “giving itself as the input”. This in and of itself isn’t possible. In other words, the machine you wish to test is A’B for some arbitrary machine B. Thus you would write A’A’B. However, the behavior of A’A’B doesn’t have to match the behavior of A’B since A’ and B are different machines.

But, you may insist, you only wish to test if A’ itself will halt when evaluating itself. Now what I want to test is A’A’. However the second A’ must have an input to be meaningful, and its input is itself, so that generates A’A’A’. The third A’ needs an input, generating A’A’A’A’. This continues infinitely. In other words, you can’t construct a Turing Machine which takes itself testing itself as an input because it would require an infinite number of symbols on the initial tape. This isn’t allowed by Turing’s definition of a Turing Machine.

If you look at the halting problem proof as described in the link (trouble) the problem with asking what trouble(t) does is that t must have an input to be meaningful. If t is itself, where does that version of t get its input from?

Approach B

Okay, but maybe there is a way round this. Suppose we write an A’’ which first copies itself into the input position on the tape (building around itself whatever UTM stuff necessary), then executes as normal. Now all we need on the tape is A’’ and once it was through with its copy operation, we would have A’’A’’ on the tape. Which would be just fine, as A’’ no longer needs an input tape (as it is going to just use itself as an input). Furthermore, once A’’ gets an answer (if it stops or not), then it will do the opposite. Run forever if it tells itself that it will halt, or halt if it tells itself it is going to run forever.

The interesting thing about this construction is that the actual recursive behavior used in the code is itself open to analysis. If the algorithm A’’ can come up with an answer, then that answer will be wrong by design. This is because after getting the answer, it does the wrong thing to make the answer invalid. As a programmer, I find this very harsh. Code doesn’t care about what I think it will do. It just does what it does.

So let’s take this a step further. Suppose I have another gear in my Turing Machine which lets me execute an infinite number of instructions, in order, in the time it takes to execute one instruction. After running in this gear, the TM sets the state of the TM to either ‘a’ if a finite number of instructions executed, or ‘b’ if an infinite number of instructions executed.

Now let’s assume that we are going to answer the halting problem by simulation. In other words, we are going to make a copy of ourselves, then simulate our own execution. If we are still running at the end of an infinite number of instructions, then we will declare that our algorithm does not halt. If after giving time for an infinite number of instructions to execute, we only execute a finite number of instructions, we will declare that our algorithm does halt.

What will happen? Well, quite clearly we are going to go into an infinite loop. Not because our approach to answering the halting problem was invalid, but because this approach runs into the same construction problem described in Approach A. Now suppose I build a tape with the original A and have it examine A’’. In other words, we build a tape AA’’ and ask it what happens.

The answer is that A can quite correctly report that A’’ does not halt. Why can A report that A’’ doesn’t halt, but A’’ cannot? The answer is that A’’ gets caught in a loop copying itself, trying to construct the tape we proved could not be built in Approach A.

Thus while in fact the halting problem may not be answerable, I don’t think Turing’s proof in the original paper proves this to be the case. It does prove that you can't construct an algorithm to prove it does halt by first copying itself.

This isn't mysterious. The simple little copy at the beginning puts everything into an infintie loop. You have proved you can't build a valid A'', but you haven't necessarily proven that no A exists.

Microsoft Researh

2006-12-06T15:04:00.000-08:00

Microsoft research? Why does that seem to be such an oxymoron?

Here is my comment on the matter:

If I have to call Microsoft over another activation problem...

Why should I as a business owner or shareholder spend my money to do a task whose result isn't a benefit to the business, but to some other company from whom the business bought a product? In other words, when a business pays someone to solve an "activation" problem, they have paid someone to insure that Microsoft was paid. The business receives no benefit, but they are out the money anyway.

When Microsoft pours money into research on how to develop technologies that seek to avoid theft of their product, that is fine until part of their solution increases the cost of ownership. When Microsoft pours money into "securing digital rights", that's fine until part of their solution increases the cost of access to content.

Microsoft and others are struggling to survive in a future where computers have nearly unlimited disk space, increasing numbers of processors, vast memory spaces, and high bandwidth to other computers. Very soon we should be able to run multiple operating systems on a single computer at the same time. Running on virtual machines will be the norm, if for no other reason than to allow applications the freedom they need to run and not step on each other or get killed by viruses and compromised by spyware.

Everyone would be impressed if Microsoft was embracing this future and working to leverage all this power for the sake of the user. Instead, Microsoft appears to be working late into the night doing everything they can to insure each day dawns according to the same old paradigms that made them billions in the past.

Social Justice and Technology

2006-12-05T13:34:00.000-08:00

This is a complex issue which effects the quality of our lives, the environment, our access to opportunities, etc. I’m not really going to go there. Nothing I write here should be interpreted as a claim that all is well in the distribution of wealth in the U.S.A.

Instead, I would like to point out that Technology is gradually wiping out the differences in the quality of the products and staples of life enjoyed by those at the high end of the income scale and by those at the low end.

Take tennis shoes, for example. The super rich can spend more for a pair of tennis shoes, but the quality and utility of those high end shoes will not differ in any interesting way from those a middle class person might wear, and not greatly from someone at the low end of our society. We rather take this for granted, but 100 years ago, this was not true. Technology is approaching parity across classes for tennis shoes.

Take computers. 100 years ago, a computer was someone who did calculations for a living. If you really needed a large number of calculations done, you needed to pay some one to do them for you. Not many people could afford to balance their checkbook that way in the early 1900’s. By the 1960’s computers were available to the most wealthy companies. Today, calculators are easily available to nearly anyone, and are embedded in so many items it becomes difficult to estimate how many calculations are performed on a daily basis for each person of any social level. Most people with checkbooks today have access to free online services that do most of the calculations required to balance a checkbook.

Take music playing devices. 100 years ago, a big divide. Today, not much of one.

Microwave Ovens. 100 years ago, couldn’t get one at any price. Today, most people with a kitchen have one.

The observation to make is that for so many “things” that we have, the super rich cannot buy at any price a version of distinctively greater quality than the middle class can buy on their own budget. And for many “things” the quality is pretty much the same for everyone.

It is so easy to be blinded by the fact that the rich can buy more and have to work less for it. However, for the first time in History, most of the basic staples consumed by the classes, and most of the most common luxuries enjoyed by the classes are of roughly equivalent quality and manufacture.

And as Technology progresses, the differences will continue to shrink.

We will solve the actual distribution of wealth problem in a future post….

Are Computers Alive?

2006-11-29T08:23:00.000-08:00

The question isn't as simple as it seems. We do manufacture the hardware, but much of this process is automated. So when we say "we manufacture", we mean we do some of the processes and leg work needed to manufacture computers, but most of the complexity is handled by automated systems (i.e. computers). The software manages the masks laid out in the processor, that builds the memory devices, guides wafers through the system.

The software actually sets up the OS, the applications, the preferences, the screen savers. In fact, while it seems that “we install” software, all we really do is just provide a few prompts. Even in very complicated installs, the number of values and files installed by the software vastly dwarfs the parameters supplied by the person doing the installation.

So if a computer can be said to “build itself” then it does so using a huge number of processes and components already built for it.

But do we build ourselves? Or do we use a huge number of processes and components supplied to us? In fact, we do do rely on a vast set of supplied parts! We eat other orgamisms and use the building blocks they provide (proteins, fats, carbohydrates) and build our selves from that.

If we rely on "manufactured parts" from sources outside ourselves, and computers rely on "manufactured parts" outside themselves, then the difference isn't absolute, but just a matter of degree. If we are alive and require all this outside help to construct and maintain ourselves, can we say computers are not alive because they too need outside help to construct and maintain themselves?

Do computers replicate? Of course they do. The "genes" of a computer system is its software. And software exists as installation files, CDs, disks... We select and install the various software components we want. We don't build applications from the ground up for each computer. Instead, we build one application, build an install file or disk, and then we replicate that.

We take the mix of software we want in a system and we install all of those pieces. And a computer is “born!”

Like "genes" in life, we see different software installed on different systems. As a result, you get different behaviors in different systems. If a "gene" (i.e. an application) is really, really advantagous, it gets replicated more often. If a "gene" is really, really useless, it does not get replicated, and it "dies out".

Do all computers replicate? No. Most just do their job. Does that make them not alive? Well most ants don’t replicate either. Only the queen and a few drones replicate. If ants are alive, than all we need to find is a few computers that replicate to consider them alive. And in fact only a few specialized computers manufacture computers, and only a few developer machines generate new software.

Living things evolve. Do computers evolve? Well of course they do! Evolution proceeds by mutation. We don’t invalidate any mutation because of the cause. In other words, just because people manipulate the genes of some organisms doesn’t make the resulting organism un-alive. The fact that people manipulate the “genes” of computers by programming, doesn’t mean they don’t change. Evolve means change, after all. It doesn’t mean “only random change”. Mutation is a difference from one version of a gene in the next generation version of the gene. We don’t write software fresh with every build of an application. We simply introduce change after change, testing at each step, and selecting the changes to keep and which to toss.

If we assume that evolution can be a guided process, then the fact that we are active in the evolution of computers doesn’t make them un-alive.

If we ever come around to acknowledging software systems as being a form of life, it will be an amazing thing. To have something take over our culture, our economy, our lives, and not even know it is really alive....

So, if not Windows or Linux, then what?

2006-11-28T14:49:00.000-08:00

Sound Byte: Operating Systems are going to give way to Configuration Systems.

What is it that we will be doing in these systems? Imagine 1000 top of the line computer systems (by today's measure) sitting in your laptop. It would be like having 1000 computer systems (nodes) all sitting in one box! Each node will have its own processor(s) and its own memory. Each may even have its own persistence. Each will likely have its own unique configuration. They would have to have their own unique state, or there would be no point in multiple nodes!

Today we have issues configuring and maintaining one such system. This really looks like a mess. But can we really avoid it? A focus on very big, very fancy, and very fast processors puts off the problem. But clearly we are going to have to shift to multiple processors for future increases in processing speed. Then eventually we will need ever wider paths to memory. Multiple paths would be best. Then we will want multiple caches (for process independence and speed). Then we will want multiple persistence paths for all the same reasons.

And towards the end of the next century, systems may be more powerful then even this.

The trick is to ask the right question. And that question isn’t a matter of replacing these Operating Systems. Such a laptop may in fact run Windows, Linux, DOS, OS X, etc.

So it may not be a question of "then what?" It is instead it is a question of "How?" as in, "How do you configure such a beast?"

The trick is to look at biological systems and examine how they pull off the exact same configuration issue at the cellular level. Regardless of how many cells are involved (like one for bacteria or trillions for a human), the organism gets a library of functions, i.e. its DNA. Each cell gets the same library, i.e. the same DNA. As each cell divides, it collects state information from its parent and from its environment. That state information allows the cell to choose a subset of the library to deploy. What subset it deploys determines the configuration of that cell.

To build such a system for configuring the systems of the future, we need to refractor the code that we already have into modules that can be selectively configured and deployed. This includes all the functionality that today we consider to be part of an "OS". We need a tagged representation (like XML) that allows the available modules to be defined. And we need a system to collect triggers from both code running in a node as well as code in "adjacent" nodes to allow for the selection of modules to be deployed.

Once configured, each node would behave (for the most part) just like a computer system behaves today, with some exceptions. In a box running 1000 nodes, we don't need 1000 copies of Windows. Maybe we have one copy that provides the UI. We don't need 1000 file systems. Already today’s Operating Systems have orders of magnitude more files exposed to the user and to the OS itself than are in any possible way useful. The only Instead, each node can manage its persistence in the way it makes the most since. Only those nodes that need to "publish" files to the UI would have to do so, and they would do so in just that way. They would publish them. If the Windows of the future cared to provide an "Explorer", then the user would be faced with only files of interest.

Developers may need to see everything in a node. Okay. Done. But really, what user really needs to see all the 1000 files an application needs for itself? If the app is broken, they can only delete it and reinstall it anyway.

There are more things to be said here. How do you test such systems? How do you reconfigure such systems? How do you extract the useful information (state information to be preserved in this system or future systems) from implementation information (the applications, the temporary state information, etc.)?

Let’s stop with some questions.

Why might we consider computers to be a different kind of life?
Why might we reject the idea that computers are a different kind of life?
If we considered computers to be a different kind of life, where does it keep its DNA?
If several computers together deploy an application, and as a living thing thus share a common set of DNA, where is it?

Operating Systems never did exist, and the ones we have are doomed...

2006-11-16T10:30:00.000-08:00

This is one of my very favorite rants. And it is one that generally has traction with people that just want to use computers, and has no traction with people that have been sucked into the way we build and define software.

I am not saying we don’t bundle up software and call that software an Operating System.

I am saying what we call an Operating System is nothing more than discription given to a bundle of software.

The main feature of an OS is that it provides a way to distribute applications and run them without requiring each and every application to provide software for running a file system, driving printers, driving displays, etc.

However, other applications provide the same abstractions. The Java Virtual Machine (JVM) provides the same kind of abstractions for Java programs even though the JVM isn’t usually considered an Operating System.

VMWare provides software to virtualized a computer, allowing one to install an operating system and run applications even though the VMWare Virtual Machine is really just an application running on another operating system.

This all makes sense if you go back to the only formal mathematical model of computation, the Turing Machine. The only thing you will find in the formalization of a Turing Machine is computation. A Turing Machine has no place to put an Operating System.

There is no spoon.

Software cannot tell if it is running on a real machine or a virtual machine. And thus Operating Systems are simply software.

But that isn’t what will really kill Operating Systems as we know them today. What is going to kill them is the death of the Von Neumann Architecture. Operating Systems make sense when the goal is to provide applications access to memory, the cpu, storage, and I/O. In a system where you only have one CPU or you can pretend you only have one CPU, managing the resources in such a system is very important.

Consider the computer of tomorrow. It is going to be a system where the number of CPUs is vast, i.e. greater than 100, and it will grow from there over time. Each of these CPUs (or Cores ) will have access to memory spaces of several Gigabytes and grow from there. Bandwidth to non-volatile storage is going to be very, very high, or the number of paths to storage is very, very high (equivalent to having more than 100 hard drives of a terabyte or more each) and they will grow from there. The number of displays will be dynamic and may even be accessed from many other systems. The bandwidth between CPUS will be very, very high, and will get faster over time.

In a system like this, it is not hard to imagine that one will be able to run many of what we would call Operating Systems of various times all at the same time. At that point of time, why run an OS at all? Instead, the problem will shift from asking “How do we acquire the resources needed by an Application” to becoming “How do we configure a portion of the hardware to run an Application”.

The Von Neumann Architecture is not so much going to die as it is going to replicate itself so many times it is going to force us to consider other, more simple and basic ways to configure and run these things we call computers.

Richard Dawkins Rants...

2006-11-15T13:51:00.000-08:00

Consider Dawkins Post "Why There Almost Certainly Is No God"

America, founded in secularism as a beacon of eighteenth century enlightenment, is becoming the victim of religious politics, a circumstance that would have horrified the Founding Fathers. The political ascendancy today values embryonic cells over adult people. It obsesses about gay marriage, ahead of genuinely important issues that actually make...

As a Child, I could see that a "Belief in God" was inseparable from a belief in "Morality". In other words, if I believe in the existence of “Right” and “Wrong”, then I must then by definition believe in something beyond what I can learn via the Scientific Method.

"Right," "Wrong," and "God" equally elude any scientific justification, for they are all equally religious concepts.

Then, knowing almost surely that I could not possibly be right on every point, I chose to believe in God as I do. I continue to believe in God because I continue to view a framework of Morality as necessary in my life. Could I be wrong? Most likely I am. However, maintaining a Moral framework for my life is important to me, and the choice I made and continue to make to believe in God is as reasoned a choice as any other.

My choice to believe in God is as scientificly reasoned as Dawkin's choice not to believe. The difference isn't the facts, but how we weigh the factors in the decisions we make.

Richard Dawkin’s campaign against religion is at best naïve. At worst, who knows? He gives us his reasons why he chooses not to believe in God, and implies that the rest of us *shouldn’t* either. Yet this is the flaw in his arguement, i.e. How do we test for the level of "should-ness" or "shouldn’t-ness" in an idea like God?

This test for "should-ness" is going to be complicated. Dawkins wrote a whole book to lay out the "should-ness" issues for God, from his point of view. What ideas are worthy of such tests, and which can we accept at face value? Or even on a whim?

"Should" I be concerned about the progress of Technology? Of Education? Of Stem Cell Research? Of Gay Marriage? Of the study of Evolution? Of the writing of the next Harry Potter Novel? Of the current Spiderman comic? What scientific test can tell us what public issues are "important" and what are "not important"?

I would point out that the overwhelming majority of organisms can't comprehend any of these issues. How can "importance" be proven to exist for humans in respect for these issues? "Importance" clearly doesn't exist for any non-living things, nor for most organisms.

It would seem "Importance" is just as much of a religious concept as "God". How is it that Richard Dawkins can accept the existance of one as a basic assumption in his life, and use it to motivate himself to attack those that accept the existance of the other as a basic assumption in their lives....

Personally, I find it odd when someone gets bent out of shape over religion because they feel deeply that religion serves no purpose and hinders our advancement as human beings. Almost without exception, all groups of humans develop religious concepts. And humans are the only organisms that develop technology. While one might not belive in God, one would have to wonder (from a scientific perspective) if religion isn't simply part of us as organisms.