Polymath A (mostly) technical weblog for Archivale.com

November 20, 2012

The little electric car that should

Filed under: Personal — piolenc @ 3:00 pm

What the world needs is a low-cost, low-tech, short-range electric commuter car.

What it is offered is a $40,000 car that costs twice that much to build, has a battery pack that costs thousands of dollars to replace and lasts (maybe) three years and is loaded with expensive, high-tech gear that nobody knows how to fix. All this to satisfy a market that, if it exists at all outside the enviro-guilt-ridden, Global Warming true believer precincts of Hollywood, is a tiny fraction of world automotive demand.

This is the result of what the military calls “mission creep.” The taxpayer – not the seller or the prospective buyer – is paying for development, so why not set the bar a little higher – insist on 150 miles range between charges instead of the 50 that most commuters in the world will need. So what if that makes the car unaffordable to most motorists, even with subsidies – it looks great on a corporate brochure and it provides unimpeachable political “green” credentials for the company.

It doesn’t have to be that way. A century ago, clunky batteries and primitive drive trains controlled by simple switches propelled delivery trucks that served major city centers quietly and smokelessly. Electric cars got people safely from home to work, to the store and back. The greatest success story for electrics is one that is almost unknown in America, and that is the classic British milk float, making its early morning milk deliveries without even waking the family dog.

When the US government first got into the business of trying to revive electric vehicles in the 1970s, it sensibly concentrated on fleet vehicles, specifically local delivery vans. Its battery development effort emphasized updates to existing battery technology – longer-lived and lighter lead-acid, cheaper and more power-dense nickel-iron cells. Progress was made, but then the shortages ceased, petroleum prices dropped and program funding evaporated. The accumulated knowledge is still out there, though.

Now entrepreneurs like Jesse Blenn and his associates, operating on a shoestring, are trying to do something really radical, namely give the world a product that real people might actually want to own. Their product is the CambyoCar – small, light, cheap… and supremely adequate. Typically, while billions of dollars of the taxpayer’s money go into the sinkhole of politically favored “green” industry, Jesse is trying to raise a few tens of thousands of dollars – and having a really hard time of it.

The Bay Area Book Exchange – a new kind of bookstore

Filed under: Personal — piolenc @ 2:46 pm

Just visited the Bay Area Book Exchange site. It’s brilliant! My first reaction was to smack my forehead and say “why didn’t I think of that?” It is yet another way to make money while giving things away. The primary attraction of this store is a huge array of books free for the asking – books that were donated for that purpose.

People trying to sell worthless old books are a persistent nuisance of any brick-and-mortar used-book dealer, motivating many to move to an Internet-only business model. Unfortunately, that move also excludes good stock that would otherwise walk in the door under its own power, and forces Net-based dealers to spend time and money hitting the auctions, pawnshops, thrift stores and so forth looking for the one book in twenty or in one hundred worth having in inventory.

This new way of doing things leaves the Exchange with a graceful way to reject the worthless stuff – it goes on the giveaway shelves – while retaining the good stuff for sale. The giveaways attract visitors, some of whom will find what they want in the “sale” section and add to the Exchange’s revenue. It isn’t mentioned in the pages I’ve read, but I’m sure they’re not depending completely on donations, and do shell out cash for the odd really good item. They obviously also offer book searching and brokerage service.

I keep finding more advantages in their way of doing things. The giveaways are not subject to sales tax and all the related BS, like the 25-cent tables used by other book dealers to attract customers. That is tantamount to a profit, because what is saved in processing costs is more than what the 25-cent sellers earn.

Goodbye, OBL

Filed under: Personal — piolenc @ 2:36 pm

Before the late Osama bin Laden fades into obscurity, I think it’s important that we recall his contributions to human knowlege and to human relations.

It was he, future generations will recall, who led us to enlightenment by confirming the truth of anthropogenic global warming and firmly pinning blame on the industrialized world. In one of his signature audio tapes broadcast on 29 January 2010, he declared:

“Discussing climate change is not an intellectual luxury, but a reality. All of the industrialised countries, especially the big ones, bear responsibility for the global warming crisis.”

Oddly, he refrained from pointing out that the most industrialized countries are also those with the smallest Muslim populations, thus missing a chance to demonstrate a clear connection between religious apostasy and environmental iniquity. Now, tragically, that chance is lost forever.

This great moment in world history has never received the attention that it deserved. Imagine the effect on subsequent events if this central figure had not revealed this truth. Millions – perhaps hundreds of millions – of the benighted and bemused might have doubted that Man’s exhalations and flatulence will soon turn the planet into Waterworld. It could only be OBL’s courageous advocacy that has kept Global Warming foremost in our thoughts and helped us banish the doubts that might otherwise have arisen there after the East Anglia scandals.

And while we’re giving credit where it is due, we owe it to History to mention his contributions to structural engineering. About halfway through the construction of the World Trade Towers, the asbestos ban came into effect and the steel columns supporting the building ceased to be sheathed in asbestos. Thermal protection, authority predicted, was not needed, or at least was not worth the danger that asbestos fiber might get into the lungs of the building’s occupants. In one brilliant stroke, OBL proved them wrong by the ingenious expedient of running fuel-laden airliners into the buildings just above the asbestos limit. He also showed that there are worse things to worry about than mineral fibers.

And as long as we’re on the topic of 9/11, it is only fair to point out that this generous act, scurrilously mislabeled “terrorism” by ignorant apostates, actually brought about an intellectual and spiritual renewal in America, land of the Great Satan. Where would we be without the 9/11 Truth Movement which, by seeking to give credit for the destruction of the Towers to the CIA, the Bush family, space aliens or Men In Black – anyone but OBL – has shown (obliquely, to be sure) its deep reverence for the sagacious founder of al-Qaeda.

Some things should never be forgotten.

Why I don’t feel sorry for municipal bond holders

Filed under: Personal — piolenc @ 2:31 pm

Thirty years ago, people were telling me to invest in government bonds for security and guaranteed yield, and wondering why I was too stupid to follow their advice. Now, people are wringing their hands over the plight of bond-holders facing default, and wondering why I don’t sympathize. Let me explain.

It all comes down to the question of what is a government bond. For the sake of contrast, let’s compare a typical “muni” or municipal bond, issued by a city government, with a bond issued by, say, a commercial bakery.

Both are instruments of debt; essentially, the bondholders are lending money at an agreed rate of interest for a specified period to the organization issuing the bonds.

The key difference between them is in who contracts the debt, and who pays.

When the Directors of ABC Bakery, Inc. vote to issue corporate bonds, they are, in effect, promising that at the appointed time ABC will repay the bonds’ principal, plus the agreed interest, out of what they make selling dinner rolls and cinnamon buns. If the appointed time arrives and the bonds are not repaid, ABC will be forced to repay them, and if they can’t, to liquidate and pay off bondholders from the assets of the company. The directors who voted the bond issue, being shareholders in the firm, stand to lose all of their investment in that event. If irregularities are found in the bond issue or in provision for repaying it, they may even be civilly or criminally liable as individuals. All of these circumstances tend to promote prudence and forethought.

When the City Council of Anytown votes to issue municipal bonds, on the other hand, the councilors are not incurring debt on their own behalf. Instead, they are promising the bondholders that, when the time comes, the Council will reach out and rob the taxpayers of the city of the amount needed to repay the debt. The councilors are not themselves liable except as individual taxpayers, and there is no recourse against them in the event of a default. They have much to gain, and nothing to lose by a bond issue. In effect, they can buy their constituents’ votes with the taxpaying citizens’ own money.

In other words, government bonds are a promise to steal and to use the proceeds of the theft for the purpose of bribery. By any consistent standard, such a bargain would constitute criminal conspiracy. Unfortunately for the honest taxpaying citizen, this particular crime is sanctioned by the very government on which he relies to suppress crime. And let’s be very clear on this: trust in government bonds is based on the fact that the taxpayer is not offered the option of not paying; he pays or else. This is no secret, and anybody who buys this debt knows that it is being contracted on behalf of third parties, without their consent. In the case of long-term bonds, those who will be forced to repay them may not even be born yet.

As a practical matter, and especially for the sake of political self-preservation, politicians avoid imposing special assessments or earmarked tax increases to cover bond repayment, and try to stagger bond maturities – all so that bonds can be paid out of current revenues.

But now consider what happens when government revenues decline, as is happening nearly everywhere today. Faced with three alternatives: cut spending, raise tax rates or go deeper into debt, politicians typically dismiss the first alternative out of hand on the grounds that the people have a duty to maintain them in the manner to which they have become accustomed, reject the second because of political risk, and eagerly embrace the third. Now, instead of using bonds for capital improvements and paying for them out of revenues, debt is being incurred to cover shortfalls in the very revenue that is supposed to retire it. It doesn’t take a degree in applied mathematics to know that this is unsustainable, and that a point will soon be reached where the interest on the bond repayments will exceed revenue. At this writing, this is almost the case in at least one American city.

These days, buying government debt isn’t just a criminal act; it is also a stupid one. Yet people continue to “invest” in this cynical bargain, apparently confident that somebody can be made to pay it off.

And they wonder why I don’t feel bad when it doesn’t work out that way.

November 17, 2012

Free Flight: Personal Leased Air Transportation

Filed under: Aeronautics,Personal — piolenc @ 7:33 am

“The question isn’t who is going to let me; it’s who is going to stop me.” – Ayn Rand

In this day of no-fly (and soon presumably no-train, no-ferry, no-bus, etc.) lists the question of unimpeded personal long-distance transportation needs to be taken a bit more seriously than it was in a recent Feedback Friday (http://www.DollarVigilante.com), which read in part: “Sadly, you have to have your own sailboat, motorboat or airplane in order to get outside of all this…” implying that these were hopeless aspirations for most of us. Taking it strictly at face value as meaning owning a seagoing yacht or an aircraft with decent cross-country performance, it’s probably true. But the very technological developments that make possible nude scanners are also making possible the physical liberation of travelers on a very large scale.

Private (i.e. non-commercial) aircraft and boats get you out of the nude scan/grope line, but not completely out of the clutches of government. Where the bureau-rats get you as a boat owner is at the shoreline – forbidding anchorage in the open bay and forcing boat owners to pay outrageous slip fees and submit to ridiculous registration procedures, stopping boats for “safety” checks and to look for drugs, etc. For aircraft it’s the airfield for “papers” and outrageous hangar fees, not to mention the ridiculous cost of the vehicle itself (due to government certification requirements) and of government-approved training to obtain a government-issued license to fly it.

But aircraft have that vital, vertical third degree of freedom that gives them the potential to eliminate contact with officialdom, if only they can get away from using government-designated landing grounds. Very short takeoff makes that possible in many places*; vertical takeoff gives an almost infinite supply of off-airport landing areas. Specific aircraft to fulfill this need may come up in a later post, but I’d like to discuss the problem of Free Flight in general terms first.

Back in the 1930s and 40s it was generally assumed that airplanes (or aircraft of some kind) would eventually become as common as the family car or “flivver,” and serious effort was put into making the perfect “air flivver” for Everyman. Much of that effort, ironically, was sponsored by government, with many NACA reports coming out during that period concentrating on shortening landing and takeoff runs, making airplanes that were stall- and spin-proof, improving pilot vision, improving fuel consumption, reducing noise and so on. Much effort also went into seaplanes, flying boats and amphibians during this period, mostly for military purposes but also to the benefit of “bush” flyers.

The emphasis in all this was on safety, but every bit of progress also held the promise of improved access to air transportation. Low stall speeds and improved low-speed maneuverability make short, unpaved fields usable. Reduced noise makes it possible to operate near homes without causing a riot. Amphibious operation makes every sheltered body of water over a certain size into a potential airstrip, and every boat ramp a terminal.

The one factor that prevented the flowering of Everyman’s Airplane was the need for skill. Make a ‘plane as safe as you like, it still takes much more skill and discipline to operate one than a car, and the penalty for error tends to be greater, too. Forgetting to check your automobile’s fuel level before departure leads to a humiliating roadside call to the Auto Club; forgetting it in your airplane can result in “catastrophic terrain intersection” in the language of modern accident analysts. These constraints cannot be blamed on the State; they are imposed by Nature. Flying an airplane from water requires additional skill and judgment. As for the helicopter, a priori the ideal “air flivver” because of its ability to take off and land vertically, even more skill is needed to fly it, to the point that not every candidate for a helicopter license is able to achieve the necessary level of skill despite the expensive instruction.

The means of overriding that obstacle didn’t exist in the 1940s, but it is ubiquitous and cheap in our day. I’m referring to the digital computer, of course. It is true that much effort has gone into analog autopilots over the years, and many airplanes are equipped with them, but they are essentially cruise controls – they allow the pilot to relax a bit in the middle portion of a long cross-country journey. The most difficult and dangerous phases of flight, namely takeoff and landing, are still the pilot’s full responsibility, and he still has to perform every detail task related to communication, information gathering and navigation, using primitive methods and in the cramped, noisy and often drafty confines of a cockpit.

More recently, digital computation and control have entered the cockpit, with the benefits that we all know so well in other phases of life, namely programmability and expandability. This has led to a vast reduction in pilot workload in commercial airplanes, because the complexity of the tasks assumed by the flight controller has enormously increased, aided by much more extensive sensor input and its ability to directly control the servomechanisms that already exist to amplify the pilot’s control effort. Certain commercial airplanes now can literally land themselves, with the pilot standing by to take over if something goes wrong with the navigation aids or the controls themselves, or some obstacle appears on the runway.

The homebuilt,  experimental or sport aircraft movement has paid attention to these developments and is already making extensive use of digital electronics in navigation and trip planning, but applying electronics to direct control of the airplane is more difficult because most homebuilt and small factory-built airplanes still have direct manual operation of the control surfaces via mechanical linkages, so all the actuators have to be retrofitted and plumbed into the airplane’s existing systems. Besides the expense, there is the question of motivation: the sport airplane crowd probably don’t especially want an airplane that flies itself. Their joy comes in exercising and improving their piloting skills, not merely in getting from point A to point B, and most sport flying originates and ends at the same airfield in any case.

Factory-built airplanes, which in a free market would quickly incorporate every technical advance making flying easier, instead lag decades behind experimental aviation. This is entirely due to the State, which requires every component of a certified aircraft, down to the last rivet, to be government-certified for use in aircraft. This makes the cost of aviation electronics astronomical, further shrinking a market that was small to begin with, and ensuring that mainstream manufacturers will never invest the development costs to bring any truly modern gear to the civil market. Barring a revolution, then, the solution to our problem will not come from the “spamcan” vendors.

Accessible personal air transportation will have to originate outside the usual civil aviation channels, even though it will draw on aviation knowledge and technology. The controls component will originate almost entirely outside of commercial, general and sport aviation because its purpose will be to take the skill out of flying and essentially make the operator of a personal cross-country aircraft a passenger who decides the destination and may make some routing decisions (“stay away from Milwaukee,” “pass Mt. Shasta to the south”), but takes no part in direct control of the aircraft’s flight path. Interestingly, this kind of autonomous control is currently in existence for one application only – unmanned aircraft, mostly used by the military. Digital autonomous and semi-autonomous controls exist for drones, including rotary-wing drones, which can be ordered to follow a certain flight path, the remote operator intervening only to order deviations and to engage targets (and even then he only designates targets, and does not have to directly control weapons). The same controls can work equally well – better, in fact, because they don’t depend on communication links – with the operator on board.

A typical trip scenario, freedom fashion:

When Mr. Freeman wants to make a business trip to a distant city, he orders an aircraft, which lands at the appointed time on his back lawn or any convenient rendezvous such as a nearby park, rooftop or sports field. He hops aboard, identifies himself to the aircraft’s computer and makes himself comfortable. The aircraft weighs itself (by reading sensors in the landing gear struts) and informs him that, although he and his luggage are within the weight limit for the trip, he will need to stow part of his luggage forward as the aircraft is currently tail-heavy. Mr. Freeman moves one of his bags to the forward bin, the computer confirms that the machine is now within its center-of-gravity range, and the aircraft takes off without further delay. During the flight, he enjoys the view, which is panoramic because the flight altitude is low to keep away from government-controlled aviation corridors. If he gets bored with the scenery, he watches a movie on the built-in displays, catches up on his electronic correspondence using the aircraft’s built-in networking facilities or reviews the presentation he will make at his destination. If the flight is too long for the aircraft’s fuel endurance, it will land itself at a suitable refueling stop – probably the back lot of a truck stop or fuel dépôt, as the aircraft will probably be burning diesel fuel or heating oil. Depending on the level of service contracted for, Mr. Freeman will either have to pump the fuel himself or an attendant will do it for him. Likewise, contractual arrangements will determine whether Mr. Freeman uses his own funds directly to buy the fuel or the fuel is paid for by the aircraft leasing service and included in its charge to Mr. Freeman. Upon arrival at destination, he will either be deposited right at his hotel, if the hotel has a pad and is willing to accommodate non-government-sanctioned transports, or at a suburban taxi stand where a taxicab contracted for by the leasing service is waiting to take him and his luggage to his final destination. If the aircraft is amphibious, landing on a convenient river or bay and using a dock or boat ramp might also be an option.

It is clear from the scenario that a very different ownership structure from that currently existing in general aviation is assumed there. One of the many reasons that private aviation is currently so expensive per mile traveled is the very low rate of utilization of the aircraft – a fewscore hours a year, usually. All the aircraft’s overhead (much of it incurred by compliance with government regulations) and debt service gets charged to those very few flying hours, and the arithmetic result does indeed make it look like a rich man’s indulgence. Of course there are rental aircraft fleets in existence today, but they consist of overpriced “certificated” aircraft which are required to be maintained by overpaid licensed Airframe & Powerplant mechanics and are only eligible for rental by government-licensed pilots, so the savings expected from our hypothetical unregulated rental fleet, serving the general public, are not realized. All that changes if the aircraft is part of an autonomous rental fleet that is rented out to multiple non-flight-qualified customers on a per-trip basis over the course of the year, spending only enough time in hangar for regular maintenance.

Where would these unlicensed aircraft come from? Not from Cessna, Piper or Beechcraft  – not for the foreseeable future, anyway – because those General Aviation manufacturers are under the thumb of government and would likely be punished if they built aircraft intended for use off the plantation. At first, production would likely be a cottage industry, with small producers specializing in various hardware components, specialized software firms providing flight control programs and data files such as digital terrain maps. Assembly and checkout would likely be carried out by the customer – the leasing service – or by a specialized assembly firm working directly under their supervision.

The rental scenario and the decentralized production scheme both imply comprehensive standardization. This tends to raise strong doubts in many freedom-lovers, who associate standardization with coercion and centralized authority. It is worth pointing out, however, that voluntary standards existed long before government stuck its oar in the water, and that even today the most effective and widely followed standards are still voluntary ones – think Internet Protocol, WiFi and USB connectors – adopted because it is in the best interests of the adopting parties to use them. Today, there are many de facto standards in effect within the homebuilt aircraft movement, administered (if at all) by membership-based organizations operating without government sanction. This is the closest thing to a free market that exists in aviation – homebuilders who want proven hardware go with the standard offerings, leaving the risk-takers and pioneers free to think and work outside the box.

Initially, somebody is going to have to take a financial risk, but it need not be a very big one – certainly not like the risk exposure of launching a new “spamcan” or factory-built personal airplane. Imprimis, there will be no up-front “compliance” costs which are a major component of cost of any new certified aircraft venture, and a disproportionate component of the cost of bringing a small aircraft to market. Secundus, the infrastructure required for promotion and administration already exists. Tertius, production tooling will likely be paid for by the individual manufacturing subcontractors, each of whom will be taking on only a small part of the project and therefore risking relatively little capital.

The biggest obstacle is, sadly but predictably, the human factor. To be commercially successful, this business must be marketable to people who have not adopted the freedom philosophy. To be a technical and production success, people who don’t necessarily share our principles have to be persuaded to participate, or people who do share our principles have to be persuaded to acquire the necessary skills. This is so because, sad to say, most people in aviation, including the sport aviation/homebuilt phase of it, are conformists when it comes to obeying government regulations, which they associate with safety and order. They are of course far from being the only ones to make this error, but it is inconvenient that a skilled workforce that would be most helpful in this project holds mostly negative views regarding the actual practice of freedom. We may have to build a skilled production and operating force entirely outside current aviation circles. We will have to market the service to travelers purely on the basis of cost and quality, and let the lesson of liberty teach itself by experience.

Fortunately, this should not be too hard. Convenience is hard to beat: pickup à domicile or nearby at a time of the traveler’s choosing, and delivery to destination, or to within a cab ride of the destination. The latter is a pretty strong selling point if the destination is other than a major hub, because the cost and the time in transit both rise drastically in that case, when flying commercial. In fact there are no direct commercial flights to smaller cities and towns; the traveler first flies to the nearest hub, then changes, not just flights but also airlines and terminals usually, retrieving and re-checking his luggage in the process, then flies on to his final destination, often after cooling his heels in a major airport, with his baggage, for hours. If the enhanced convenience, privacy and comfort of a free flight service can be accomplished at or below the price of a commercial ticket, it should be an easy sell. Even if sanity returns to commercial travel and the TSA goons are forced to seek honest work, the service should survive for decentralized travel and for feeder service to major airports. In our hypothetical trip, for example, Mr. Freeman’s autonomous air taxi ride might be the first leg of an international journey.

“Bootstrapping” is an option. The first increment of the business developed could be a franchised chain of local pinpoint aerial parcel-delivery services – the aircraft smaller, short-ranged and not man-rated. Most of the software – particularly close-quarters piloting and dispatch algorithms – developed for this phase will be applicable to later phases. This would be followed by local air taxi and airport feeder service in areas of high traffic congestion, infrastructure bottlenecks or both. Both would form the basis for development of still longer-range vehicles and for building a network of service depots across the country.

* A late and much lamented, very eccentric friend liked to go “camping” with his nimble and somewhat modified Aeronca Champ, a high-wing taildragger in the proud tradition of the Piper Cub, but larger. He would fly out into the countryside at low altitude, pick out a pleasant-looking spot – a country road, a meadow, a ridgetop – and land there. He would then conceal his Champ, which was painted a dull grey (or perhaps that was just the color it had turned over the years), and spend the night sheltering under the wing, eating whatever he’d brought with him and reading by flashlight. (His rudimentary airplane lacked an electrical system, and I had helped him rig up an external battery for his handheld VHF radio, which he would take home for recharging between flights.) If he was feeling too lazy to pack a picnic he would seek out a nice spot close to a truck stop on the Interstate and thus gain access to the truck stop’s café and rest rooms. He loved to point out that he could find clean air, tall grass, peace and quiet a short walk from a freeway that was at times jammed bumper-to-bumper with vehicles. If he felt the need for fuel he would buy some at the truck stop, having brought along a jerrycan for the purpose. When he wanted to move on he would push his machine out of its hiding place, swing the propeller of its sixty-five-horsepower engine to get it started, leap in and take off before anybody was the wiser. His arrival would not have been noticed because he would glide in with the engine barely ticking over, so his departure would come as a complete surprise to whoever was in the neighborhood. This is not a solution for Everyman – Ed could get away with it because of his intimate familiarity with his very low wing loading, balloon-tired airplane and because of the flying skill he had accumulated over decades – but it does point up the possibility of operating a nominally very noisy aircraft in places that are generally thought to be too densely populated for such an activity.

May 12, 2012

Notes to myself about starting a translation agency

Filed under: Personal — piolenc @ 6:25 am

Lately I’ve resurrected a project that I abandoned over ten years ago, namely that of starting a translation agency. My main reason for dropping the idea was that I was doing fine as  an independent translator working for agencies and sometimes for individual clients; that being the case, I saw no reason to incur the additional responsibility of running an agency, with all the attendant hassles. Now that I’ve seen the caliber of agencies in the market and have found all that I have dealt with far from ideal, I’m thinking again about “starting something.” Here are some jottings about what I want to accomplish, and what I want to avoid.

1. Do not start a “Me Too, Ltd.” translation agency; it has to be truly and recognizably different from the norm – and preferably much better – to be worth doing.

2. Don’t encumber the agency with an overburden of corporate trappings; start off as “F. Marc de Piolenc, Translation Broker” or the like, doing business as a sole proprietor, then add on and alter as the actual market demands. Initial capitalization is limited to a cash reserve (operating capital) greater than or equal to all outstanding liabilities to translators.

3. No salaried employees; contract for all services except direct customer relations – order taking, scheduling, QC and invoicing will be performed in-house by the proprietors.

4. No freebies – translators get paid for everything asked of them, including tests. (Whenever possible, ask translators to submit samples of work already performed.)

5. No “we’ll pay you when we get paid.” Maintain a cash reserve sufficient to pay translators on a fixed, agreed schedule, whether or not the client pays on time (or at all). Expand the business only as and when a larger cash reserve is available.

6. Demand quality and be willing to pay for it. Accept – initially at least – a lower profit margin in order to get the best translators and still offer an attractive price to prospective clients.

7. Provide clear and simple pricing to customers, and issue clear and simple instructions to translators.

8. Translators translate – other services are procured at additional cost or provided by the agency. It is not reasonable to expect translators to deal with elaborate formatting and page layout for the same price as straight text unless the agency provides a template that the translator can use directly.

9. Politely refuse work that imposes an unreasonable delivery schedule. Insist on time to do the job right – and then do it right, and on time. Leave the “instant” translations business to the slimeballs, and wait for their disillusioned ex-customers to come to me with more reasonable expectations.

10. Accept down payments on larger jobs, but only from established customers, and only on jobs allocated to proven translators. Down payments cannot count toward the cash reserve, because they are liabilities too – to the client, in this case – until the client accepts the finished job. Segmentation and progress payments can be used to mitigate this problem.

11. Until a pool of proofreaders is established, accept only work with a target language known to me, Sharon or my immediate circle, and in familiar fields of endeavor.

12. Educate the customer – help him to recognize good quality work and give him an honest idea of what he needs to budget to pay for it. Show a clear separation between translation tasks and other work that might be required.

13. Be a translation agency – not a tax collector or social-insurance agent. The translator is responsible for his relations with his local tax authorities. The agency will provide documentation required BY THE TRANSLATOR directly to him and him only; what he does with it is his business.

February 23, 2010

Education for a Strong Republic

Filed under: Personal — piolenc @ 11:00 am

Address at Puga-an High School
Barangay Puga-an, Iligan City
March 31, 2004

Principal Dante Sumagang; staff and faculty of Puga-an High School; students; graduates; ladies and gentlemen:

Today it is my honor and pleasure to address you on the topic of “Education for a Strong Republic.” This is a question of vital importance to the future of this, and many other countries.

The word “republic” has had various meanings through the ages since it was first used in the ancient world. Even today, it has different meanings in different places, so it behooves us to define our terms carefully before proceeding. Very simply, a republic is a country governed by its people.

A republic has a government, but that government has no authority or power of its own. Instead, it is the agent and servant of the people. Like any agent in the business world, it has a contract with the people that defines its duties and powers. We call that contract a constitution.

So one distinguishing feature of a republic is the relationship between the people and their government, which is that of master to servant, or better still of employer to employee. Let’s borrow an example from the world of employment to illustrate that relationship, and to see what can go wrong with it.

I like to use the analogy of a security guard hired by a department store. At first, his job is straightforward: he scans the crowd in the store for pickpockets and purse-snatchers, and makes sure nobody brings weapons or explosives into the store. If a customer in the canteen gets drunk and starts to annoy the other patrons, the guard gently removes the rowdy from the store.

So far, so good. The guard learns his duties and performs them conscientiously and discreetly. The customers and salespeople go about their business, hardly noticing the guard unless there is a disturbance. And there are very few disturbances, because the guard is vigilant. But the owner is not satisfied. After all, that guard is being paid for full-time work, but he spends most of his time just sitting and staring. Pretty soon, the owner starts to assign the guard other tasks “to keep him busy.” Customers often come into the store burdened with purchases that they have made elsewhere. For their comfort and to discourage shoplifting, the owner sets up a check-room where customers can deposit their packages as they come in and retrieve them when leaving. Who will accept the parcels and later give them back? The guard, of course—after all, he isn’t doing anything most of the time. The owner congratulates himself that he is saving money, because he has not been obliged to hire a new employee to provide the new service.

All seems well until one day, while the guard is busy checking in packages, a notorious thief enters the store and steals a lady customer’s handbag while she is paying for her purchases at the checkout counter. The thief escapes before the guard can be alerted. Human nature being what it is, the owner does not blame himself; instead, he scolds the security guard! When another theft occurs, he fires that guard and hires another. The new guard is just as overburdened as the old one, so the same thing continues to happen. At last, the owner admits that the guard has too many tasks to perform…and hires an additional guard. Naturally, this increases the store’s operating costs, so the customers are required to pay higher prices for the goods that they buy there. If he had considered the matter logically, the owner would have set up a self-service checkstand – perhaps with lockers – and given the guard only his original duties to perform, then lowered his prices to their former levels. The result would have been greater security at a lower cost. Not to mention better sales due to more competitive pricing.

We see the same error being committed on a much larger scale in the world at large. In every republic on Earth, citizens are demanding more of their governments than they can reasonably perform. When government fails to meet expectations, as it inevitably must, citizens assign it more resources..and even more responsibility. Few of us pause to consider whether we could perform these functions for ourselves at a lesser cost, and fewer still consider that a government that can give us everything we want must necessarily be able to take from us everything that we have. Like the store owner in my little fable, we are always demanding more, always getting somewhat less than we demand, and always paying more for it than it is worth.

In the process, we are creating a weak republic. Actually, we end up with something that is not a republic at all, but …something else. A strong republic, then, is one whose government has very few tasks to perform. Those tasks are tasks that the citizens either can not or should not perform for themselves. Because the government’s duties are limited and clearly defined, it is able to learn them and do them effectively and consistently. A strong republic, unlike those we know, has very few laws, but those laws apply equally to everybody and are enforced consistently and fairly.

A strong republic must be founded on strong citizens. I don’t mean that we should all be weight-lifters, but we must be good citizens, and that means understanding our responsibilities to our fellow citizens and our prerogatives with respect to our government. And that, ladies and gentlemen, is where education comes in.

What can education do to promote a strong republic? Quite simply – everything. Not all of education occurs in school, of course. The absolute fundamentals of a free man or woman’s preparation for life—respect for the lives and property of others, honesty, maturity, being true to one’s given word—all of these moral and ethical traits are best learned at home, by the example of one’s parents and one’s elders. But the schoolroom is still the training ground of the effective future citizen.

The history of Mankind is, by and large, the history of failure. Failure to act appropriately and to make the right choices, but above all failure to learn from the earlier failures of others. The academic subject that teaches us to avoid repeating these failure is of course History, which would seem to make it the most important subject. But it isn’t. If we are to learn the lessons of history—or of any other subject matter—we must first have the language skills that make us able to understand what we read and hear, analyze it and discuss it intelligently with others. If we can read effectively and with full understanding, then we can teach ourselves the things that our teachers and our parents can not teach us or don’t have the time to teach us. Without language skills, nothing else is learnable. By my reckoning, the preschool teacher who gives a child his earliest acquaintance with letters and numbers is perhaps the most important link in the long chain of instruction that leads him to graduation.

But which language? I believe it must be English. I say that knowing that this claim will not please everybody.

Am I claiming that English is a better language than any other that we might learn? No. The science of linguistics teaches us that all natural human languages are equivalent, and my limited experience seems to confirm that finding.

Do I believe that English speakers are endowed with superhuman intellect, that they are somehow better than those who do not speak English? Again, no. I’m sure that any one of us could think of several stupid people who speak English very well indeed. I see some of them on television almost every night.

What the English speaker does have, and what the non-English-speaker is deprived of, is access to the largest body of literature, on every subject, that has ever existed. English has also become the de facto language of commerce and of diplomacy, where it has displaced French. It is the official language of air commerce; when a Nigerian airline pilot asks for clearance to land at Orly, near Paris, he makes his request and receives his instructions in English. English is also the language of the sea. If you go to sea, and expect to do anything more rewarding than swabbing the deck or cleaning out the head, you must know English.

Those responsible for the education system of this country recognize these facts, though they may not be happy about them. And so, as a result, the language of higher education—university and postgraduate studies—is also English.

I see that English is taught as a separate subject in most schools, and taught as well as the resources of the school permit. But another vital subject—Civics (which now includes what little history is included in most curricula)—is taught in another language that is just as foreign to 30% of the population as English is. Still other subjects are taught in a mixture of English, Tagalog and the local language.

What does this do to our hopes for a strong republic? One of the few lasting benefits of the colonial period was the imposition of a common language for law, instruction and commerce. Better yet, that language as of 1946 was English, then as now a window to the world. Since that time, education has been Balkanized—fragmented. When English was the lingua franca, every student, no matter what dialect he spoke at home, had the same obstacles to overcome to master his formal schooling, and the first was learning English. As he gained confidence in that language, he found greater and greater ease in studying all his other subjects, which gave him a very strong incentive to master English.

Now, however, when a non-Tagalog-speaker enters school he finds that he must learn two foreign languages to finish his studies. What is more, he has little incentive to do well in English, because a good grade in English contributes only one-fifth or so of to his grade average. To pass Civics he must do well in Tagalog, and in most other subjects he can be sure that his teachers will switch to the vernacular to explain any difficult point. Learning a foreign language is difficult; to call forth his best effort, a young student must see an immediate benefit. That benefit is not offered in the non-Tagalog-speaking provinces.

And now consider the burden on schools. In the United States of America, which spends more on public schooling per student than any other nation on earth, most public Elementary, Junior High and High Schools have given up teaching even one foreign language. Yet every provincial school here—even those whose students must build their own desks—is expected to teach two. What amazes me and astounds the Peace Corps workers I have talked to, is how well many schools manage to perform this demanding task.

Even so, provincial High School graduates who do not speak Tagalog at home are at a disadvantage in seeking higher education and better opportunities, unless of course their parents were wealthy enough to send them to private schools or to hire language tutors. But it is the poor who need higher education most; it is they who need to qualify for better jobs. And the poor of this and other remote provinces are held back not so much by their poverty (there are scholarships available, after all!) but by the extra difficulty that they have to qualify for entrance to university, or for a high paying job with promotion potential.

This linguistic ghetto—created, ironically enough, in the name of national unity—contributes to discontent and a feeling of “us versus them” that has no place in a free, strong republic. It limits access to the many sources of instruction that make it possible for any determined citizen to fit himself for his role in the republic. In short, it ensures that much of the country’s population consists of second class citizens, who will have a more difficult time than most in improving their lot. This creates the temptation to employ short cuts, and some of those short cuts involve banditry and other forms of crime and delinquency. These in turn make it necessary to devote more resources to government, deprive the citizens of more of their liberties and thus further weaken the body politic.

What can educators do? Well, resources may be unevenly distributed, but ability and ambition are not. Some of the faculty members and administrators here today will someday gain high positions in the Department of Education—positions from which they can influence policy and determine the nation’s future by forming its future leaders. Some of the students and graduates seated here will gain government office; they, too can have an influence.

I hope that they will remember, then, how vital to our future it is to have a country ruled wisely by its people, and administered by a government that does few things, but does those things very well indeed.

Thank you!

Address to the School of Computer Studies, MSU/IIT

Filed under: Engineering,Personal — piolenc @ 10:16 am

This was written and delivered nearly six years ago. Unfortunately, the part about the consequences of using unverifiable software in life- and mission-critical applications has come tragically true. My other prediction—that this problem would be recognized and that packagers and end-users would insist on open-source software—has still not come true, and there is no sign of it happening.

Think about that the next time you go in for a CT scan.

Address to the graduates of the  School of Computer Studies, Mindanao State University/Iligan Institute of Technology
Recognition Night
29 March 2004

Current trends in information technology, and their implications for young professionals and end-users of all ages.

Good evening.

While preparing these remarks, I realized that it has been almost 35 years since I first sat down in front of a computer. Actually, that is not strictly true; the computer was 100 miles away, at Dartmouth College’s computer center. I was sitting in front of a very tired Teletype terminal at my prep school in Andover, Massachusetts, laboriously typing a 20-line BASIC program on its user-hostile keyboard, anxiously watching the fuzzy letters – all uppercase, of course – appear on the rough roll of yellow paper. At the same time, a pattern of holes was being punched in a narrow strip of paper tape. When I had finished this work, which was done offline, of course, I fed the tape through the reader attached to the terminal, where it was converted into frequency-shift keying and sent to the expensive giant at Dartmouth. The terminal typed “READY.” I gulped and typed “RUN.” A few seconds later, the terminal came to life. I awaited with bated breath the outcome of my first computer job. At last, the print head moved, and printed

JOB TERMINATED
CPU USAGE 3.0 SECONDS

…my entire CPU time allocation for that month.

I had programmed my first endless loop.

In those days, it was clearly understood that “real” computers would always be large and expensive and require a dedicated staff to keep them running and to see that they were put to the most profitable use. For most computers, that meant “time sharing”—allowing remote users like me to submit batch jobs and obtain the results for a set cost per CPU second. The trend in computer design was toward larger, more powerful machines capable of accomodating a larger number of users, and thus of making more money for their owners.

If anybody had made the prediction at that time that within ten years, computers would appear that would give a single user – the owner – more computing power on his desktop than the biggest 1969 mainframes, he would have been laughed out of the room. That was science fiction – and not very good science fiction. It was the same kind of implausible literature that gave Dick Tracy a videophone that he could wear on his wrist. Nonsense!

But that is exactly what happened. In my collection, still running but no longer in use, is a Polymorphic 8813, which came out in 1977. It came with a BASIC interpreter, a FORTRAN compiler, and even some prepackaged business software, including WordMaster, a primitive word processor. Its 8-bit microprocessor was clocked at 256 KILOhertz – about a thousand times slower than today’s machines. There was no hard drive. Later models had a 5-MB capacity eight-inch hard drive available as a special option, in a separate cabinet with its own power supply, but the cost was too high for most users, including me. Even so, it worked, and I continued to use it even after I had bought a more modern machine—a Kaypro 4 running CP/M. The “Poly” continued to serve until I assembled my first IBM AT-compatible machine in 1990. I even bought a 300-baud Hayes modem for it so that I could do my university programming projects at home, upload them to the VAX at UCSD and get back the results. Finding a parking space at UCSD was almost impossible at times and the computer lab was always crowded, so this early foray into telecomputing saved much time and aggravation and encouraged my early experiments with bulletin boards before I defected to the Internet in 1995.

I keep that old Poly to remind myself how quickly and drastically the dominant paradigm of information technology has changed. The Poly came out right on the cusp of the first big change, from multiple remote users and batch processing on a mainframe to a single user and interactive processing on a cheap desktop box. The next big change—packaged programs—was just starting; the Poly’s owners were still expected to churn out most of their own applications.

By the time I bought the Kaypro, that had changed. Most computer owners were users only – they bought packaged software and used it to perform whatever task needed doing. The hobbyists who had sustained the early development of microcomputers were still around, some of them working on commercial software applications, but many still following their original hobbyist inclinations and uploading ingenious, compact utility programs to specialized bulletin boards, from which appreciative users could download them free of charge. In many cases, they published the source code so that users could “port” the software to their particular platform, because this was still the heroic phase of microcomputer development and there were many competing architectures and operating systems.

The so-called IBM PC changed all that. In a bewildering series of switches, IBM had first dismissed microcomputers as mere toys, then endorsed the S-100 passive-backplane hardware standard (the basic architecture of my old Polymorphic, updated to allow for a future 32-bit data buss). Then it had come out with its own brand of microcomputer, with a motherboard-based architecture designed by Intel that didn’t have the slightest connection with the S-100 buss. Never mind. It was IBM, and it could no more be questioned than you could question the wholesomeness of a Disney cartoon. The final nail in the coffin of all the alternatives except Apple was, oddly enough, a setback for IBM. For some reason that I still don’t know, there was no intellectual-property protection on the motherboard architecture or the IBM PC expansion buss. The only thing protected by copyright was the firmware – the BIOS written on a ROM housed on the motherboard. When first one, then another clone company figured out how to make a BIOS ROM that was functionally identical to the original IBM BIOS, but contained no pirated IBM code, the way was open for other manufacturers to make computers that were functionally identical to the IBM PC, but cost a great deal less. The market for software that could run on the IBM PC expanded, and with it the number of programmers willing to enter that market. The commercial software publishers quickly realized that identical hardware platforms meant that they need only license the executable code to the user; the source code could be kept secret. The era of unaccountable, unverifiable software had begun.

Now forgive me, but I have to break the chronological sequence to go back to the early 70’s and another important development that affects us now very much. That, of course, was the development of the Unix operating system. Up until that time, an operating system was always understood to be an executive program that served as an intermediary between applications software and the basic system hardware, besides performing some housekeeping functions. Almost by definition, an operating system was programmed in machine language or in the assembly code of the particular, proprietary hardware architecture that it was meant to serve. Unix, on the other hand, was designed at the outset for portability—it was intended to be used on a variety of platforms. To make Unix portable, it had to be coded in a high-level language, not in proprietary, machine-dependent assembly code. To accomplish this, the developers of Unix had to create another pioneering work whose full significance was not properly appreciated at the time—the C programming language.

Existing high-level languages assumed that any system-level instructions would either be handled by the operating system or by the sloppy expedient of inline assembly code. There was no provision in FORTRAN, ALGOL or any other language for manipulating system-level entities. Clearly, however, a high-level language that would be used for writing an operating system had to have that capability, so C was created to provide it. Professional programmers and enthusiastic amateurs fell in love with it—its immense power, recondite syntax and general illegibility appealed to what I believe is an ingrained masochistic streak in computer geeks. Pretty soon it became the generally accepted programming language, almost eclipsing all predecessors. C and its daughters Java and JavaScript clearly dominate the languages market today.

I’m sure that the original creators of C did not intend to create a monster, but that is what their creation became when it left the mostly professional, competent circle of Unix geeks and got into the hands of amateurs and job-shop code butchers who were paid by the line. You see, to give a programmer access to the system from a high-level language, C had to discard most of the safeguards and error traps that by then were standard in compilers for other languages. It has to be possible, for instance, to assign a logical quantity to a floating-point variable or vice versa, something that is frowned on in any other context! That power appeals to nearly everybody, but giving it to any but the most experienced programmers is like giving a loaded .45-caliber pistol to a three-year-old child.

Unix, too, has its dangers. It is modular, highly versatile and very tolerant of shenanigans that other OS’s won’t put up with. It lends itself readily to networking and can allow a user on one networked workstation or mainframe transparent access to resources on another, possibly very distant platform. This is one of the factors that encouraged its early porting to microcomputers despite its hunger for system resources—one of the things that computer users lost when they migrated from mainframes to micros was the ability to easily share resources. Networking has made up for that. Unfortunately, properly configuring and administering a Unix system is not a simple matter, and moving Unix to the individual, minimally trained end-user’s desktop will require a good deal of adaptation and efficient automatic configuration.

Most of the commercial software—systems and application—in use today is programmed in C, and all of it is defective to some degree. Mind you, any large, complex project inevitably has some flaws. But add the power of C, and you have a recipe, not only for serious functional defects but also for deadly, lurking vulnerabilities that give a malicious programmer opportunities for unprecedented mischief. As this software moves into mission-critical applications—control of power dispatching over continent-wide grids, medical radiation therapy machines, life support, military applications, guidance systems—the potential for harm multiplies.

Believe it or not, this excursion into recent IT history really has a purpose, and that is to provide the factual background for a prediction. So far, I have set the stage, and all the actors in the digital drama that will be played out over the next few years are in costume and in character. We have:

· A systems software market dominated by proprietary operating systems programmed (mostly) in C; their source code is kept secret—only the binary code is distributed. The users find out about flaws in the code when something goes badly wrong with it, or when a malicious programmer detects and exploits a vulnerability.
· The alternatives that are being offered are various flavors of Unix, all programmed in C. Some are proprietary, some are not, some are disputed; some of the proprietary versions are “open source,” others not.
· The operating environment of software, already complicated by the many configuration options available to individual users on isolated workstations, is further complicated by the ubiquity of networking.
· We see growing concern among users about the dangers inherent in defective software, and growing pressure to impose standards and require testing, at least for certain critical applications.

Now for the prediction. Right now there is a growing sense of disgust with the unverifiability of commercial software. That disgust has long pervaded low-level users, but now bigger players are getting irritated, and their worries carry more weight. The US government has, for many years, sought to produce or to buy a “trusted” operating system – one that would allow concurrent processing of non-sensitive and highly classified material on the same machine or even over a network, simultaneous users being allowed access only to the material that they are authorized to work with. I first heard about this project more than ten years ago. There is still no trusted operating system available, and none in prospect. Recently, three cancer patients in South America suffered fatal overdoses of radiation when the software furnished with the radiation therapy machine gave plausible—but totally false —results and those results were uncritically accepted. This will continue to happen, and happen more and more often, until fundamental changes are made in what software comes to market, how it gets there, and what happens afterward.

The fundamental problem is software testing, software verification. Up until now, the model used for testing has been borrowed from other fields of engineering, where it has worked fairly well; this is functional testing. You take a new car model, and you run the prototype over a torture track and see what breaks. You figure out why it broke, redesign and repeat the test until you have a suspension that you can trust. A test pilot puts a new jet through harrowing maneuvers with part of its systems shut down, disabled or impaired to prove that certain essential functions continue to operate.

What is slowly being realized is that this doesn’t work in the digital world. The model simply doesn’t fit. An aircraft or automobile is fully characterized by a seemingly large, but still finite and knowable, set of parameters, all of them under the control of the designer and builder. That is not true of a computer. A general purpose digital computer is whatever its software tells it to be—a powerful calculator, a typesetter, a 6-piece orchestra—and whatever its connected peripherals allow it to be. A computer cannot be exhaustively tested; at best, it can be programmed to perform certain standard tasks—the Sieve of Eratosthenes, for example – and its performance of those tasks measured. This is where the so-called “benchmarks” come from that are published in the computer press. This is fairly obvious and well understood.

What is not as clearly realized is that the same holds true of software. In order to justify the expenditure on programming labor, software must be able to serve the largest possible group of users. Because those users have many configuration options available to them, the software must tolerate those variations and still run. It is fairly easy to prove that a program does what it is intended to do, at least in a certain environment. The tasks it is intended to perform are known, so the software can be given the correct input for each task, and it is easy to check that the proper result is returned. Even so, the operating environment of modern commercial microcomputer software is so complex, and changing so rapidly, that even basic functional testing is rarely complete when a program is released. During the life of a given software version, incompatibilities with configuration options, peripherals, drivers and so on will be discovered. The reputable software publishers maintain compatibility lists and keep their customers apprised of what works and what doesn’t. Even so, these problems are quickly discovered and obvious, and so are relatively benign.

The more serious problems arise when a program receives input that the writers never envisioned, and therefore never tested. In other words, the problem here is to test that a given program not only DOES what it is intended to do, but also does NOT do what it should not do.

That second task requires that the publisher subject the program to all possible incorrect input and to all possible incompatible operating environments, and thereby prove by exhaustion that the program will do no harm—or at least, no intolerable harm. This is impossible, for the simple and obvious reason that the set of incorrect and unexpected inputs is infinite.

Well, then, if functional testing won’t work, what is left? Clearly, we have to be allowed to look under the hood. In other words the source code must be published. This does not mean that the source code must be put into the public domain—after all, the novel that I bought today at the bookstore has its text published (there would be nothing to sell otherwise), but the copyright holder retains his right to his work. Why “open source?” Simply because no software publisher, however great his resources, has the time or the personnel to run every possible “what if..?” scenario on a chunk of code. Publishing the code and inviting critique makes the entire world your testing laboratory, every interested professional a member of your Quality Assurance staff. Of course, it makes every kibitzer a licensed critic too, and that can be irritating, but in my not-so-humble opinion the gains outweigh the inconvenience. Now you may receive a message in halting English from Vladivostok saying something like “I am not finding error trapping for input buffer overflowing…,” which is much better than fielding your product, only to have a criminal hacker in Podunk discover the vulnerability and exploit it. It also encourages the production of compact, well-documented code —nobody likes the entire world seeing his dirty underwear. That in itself is a benefit, because sloppy coding makes a source-code vulnerability analysis that much more difficult.

Will Open Source solve all our problems? No. We will still be human, still capable of making mistakes. Lousy Open Source software is still lousy software. The key advantage of Open Source is that it makes the detection of errors at an early stage more probable, success more easily achievable, and disasters less likely—provided of course that the valid criticisms are recognized, heeded and acted upon.

I venture to predict that, within five years, open source software will be mandatory for mission-critical applications in government and medicine. After that, corporate IT chiefs and server administrators will make the same demand, and the practice will quickly spread to the ordinary user’s desktop.

As for WHAT Open Source solution wins, my crystal ball is a bit cloudy. All the alternatives to the secret-source Microsoft/Apple axis are currently Unix versions, but we’ve seen that Unix is not necessarily the answer to a maiden’s prayer. I sometimes have weird dreams in which Bill Gates has an ecstatic vision, emerges wide-eyed from his inner office, and says “PUBLISH!” There will soon come a time when that will be the only way for him to preserve his market share; the question is, will he recognize the fact and act in time

February 7, 2010

Keeping What is Yours

Filed under: Engineering,Personal — piolenc @ 3:03 pm

Personal Security in the Age of Intrusion

This post is taken from chapter notes for a book project with the same title as the post, and for another called The Tropical House.

The book that got me started in designing fortified hiding places was called How to Hide Almost Anything by David Krotz, published by William Morrow in the Seventies. Another book, with a title like “Secret Hiding Places,” from one of the small alternative publishers, was also helpful. Both emphasize concealment and disguise, but not resistance to forced entry once the hiding place is compromised. I’m fairly sure that Krotz also doesn’t cover the “dual port” problem, though I don’t know because both books are in my container. As mentioned in a chapter draft for my Tropical House book, there are three main considerations in providing secure storage under your own control:

1. concealment and disguise
2. resistance to forced entry
3. ease, rapidity and frequency of access

The nature of the threats that you face will dictate how much weight to give each factor, and that will in turn constrain your choice of solutions.

1. In the extreme case of secure storage for a vacation home that is vacant for most of the year, concealment is very important because thieves have a lot of time in which to work on penetrating your security measures. If they can’t find your strong-room to begin with, they can’t penetrate it, and if you’ve kept a low profile—so that nobody even suspects you have a cache—that’s better still. This point may be much less important in the case of a dwelling that is occupied full-time, unless of course you are forced to locate storage near areas accessible to strangers, like utility meters or service entries.

2. This is self-explanatory. How resistant you make your arrangements depends on what class of thief you are expecting, and how long he might have to work in the worst case. Also: how severe the consequences of a successful penetration would be for you. Here the possibility of duress has to be considered, too. If it is possible for somebody to be holding a gun on you or a loved one and demanding that you open up, special provisions need to be made. Here again, discretion is a great help. If the badguys don’t know you’re hiding something, they won’t demand that you reveal it.

3. This is often neglected in planning, but it’s of paramount importance. Suppose you live in one of the many countries that severely limit personal firearms ownership, and prohibit firearms ownership by foreigners entirely. Now suppose that you are fortunate enough to obtain a firearm with which to defend your household. You have to conceal it, but it does you no good if it’s locked in an impenetrable concealed gun safe that takes a half-hour to open. Here ease and speed of access are going to weigh more heavily than security factors. Likewise if you keep a “bug-out kit” of travel documents and ready cash, which you might need to get your hands on in a great hurry. In an extreme case, like the gun locker that I designed for a motor home belonging to a fellow who made frequent forays into less savory areas of Central America, you might have to settle for concealment as the only security measure, with a magnetic latch the only impediment to opening the locker.

I mentioned the “dual port” problem. This is an outgrowth of the fact that the biggest criminals in the world today are governments, and it is very easy in many parts of the world (including places in which the rule of law prevailed until quite recently) to get unlimited authority to search a home or business on the flimsiest of pretexts. In such an event, the outcome is foregone: your hiding place WILL be found and it WILL ultimately be opened. Here the only effective countermeasure is provision for unloading the cache from another access or “port” so that the cache will be found empty, or better yet loaded with only innocuous items of personal and pecuniary value. This is tied in with the duress problem generally, with the added complication that there is no point in providing a duress code that alerts the authorities, because they’re the ones doing the stealing.

Disguise is a popular way to conceal small objects, with novelty houses selling fake switchboxes and outlets, and booksafes. The problem, aside from limited capacity, is that these subterfuges are well known. I would never use a switchbox, junction box or any other electrical or plumbing item for concealment unless it could still function in its ostensible capacity: you should be able to plug something in or flip the switch or turn the valve and have it work. The item should also match all the others of its kind in your house in brand name, model and degree of wear, which is hard to arrange. I particularly distrust booksafes. The commercial types are completely useless because they stand out on nearly any normal person’s shelves. This is an item you HAVE to make yourself from a real book that fits in with the rest of your collection. Aside from the pain involved in committing vandalism on a useful piece of literature, the risk of a perfectly innocent person opening the “book” and discovering your secret is fairly high in most cases. One might easily be tempted to use modern hi-fi equipment as hidden storage; much of the equipment sold today is packaged in big, expensive cases to look as substantial and imposing as older vacuum-tube equipment, but consists mostly of empty space at the back, with a few tiny printed-circuit boards crowded with solid state components just behind the front panel. There is an obvious security problem here, namely that a thief might make off with your stereo for its own sake, not knowing that he has taken your stash with it. Another problem is cooling the equipment (remember, it has to work as advertised). You may have to install a pancake cooling fan or two to make sure your amplifier gets enough air…and plan to do any servicing and cleaning yourself!

Over the years since I got interested in this, I’ve worked out a generic two-layer strong room or safe design which works well in a wide variety of circumstances. The outer layer is designed for minimal security but rapid access, while the inner layer incorporates more robust security measures, and can also be made fire resistant. The outer layer is secured by an electronic lock, while the inner layer would typically have a mechanical combination lock. The outer layer holds rapid-access items like defensive firearms; the inner layer holds deeds and share certificates, bullion and bullion coins, your draft Memoirs, compromising pictures of the First Lady and the stableboy, etc. As a practical matter it is not usually possible to fireproof the outer layer, so any documents and currency stored there may have to be enclosed in their own fire resistant pouches. There is another obvious weakness to the outer layer, namely the need for electrical power to operate it. A backup battery may have to be provided, and it will have to be located inside the strong room so as not to give away its existence. In the past, finding a way to conceal or disguise the keypad that operates the outer lock, while ensuring easy access, was a problem, but there are ways to deal with that today that also give much faster access than tapping a four-digit code.

January 16, 2010

Virtual Freedom

Filed under: Personal — piolenc @ 11:43 pm

Since the early Seventies I have been interested in New Country projects, which aim to set up communities that are completely independent of any nation-state. Usually, the plan is to build an artificial island or platform on the high seas; the more ambitious plot to launch themselves into space. Here, though, is a pleasant fantasy that occurred to me when I started reducing the New Country challenge to its essential elements.

The Virtual Alternative

Most “new country” projects include some form of isolation, at least in the initial stages of development of the new country. This seems like an obvious requirement, both to place oneself clearly outside the jurisdiction of any nation-state and to avoid notice while developing effective defense systems.

Secession allows one to form his own country in place, but has the obvious disadvantage of arousing an immediate reaction from the nation-state whose territory is thus alienated.

This piece is about an hypothetical alternative that grants the desired benefit of a new country—separation from oppressive nation-states—while preserving access to the many resources of established countries—manpower, information, means of production, transportation networks and so on. This alternative I call virtual separation.

Concealment

What do we really want from a new country project? In a word, freedom. We want to exchange goods and services with one another without restriction or taxation, and with complete privacy. We want to speak our minds without inhibition, no matter whom we may offend. What is really needed in order to accomplish that?

For trade, we need to be able to produce, store and transport our goods without fear of intervention by the State. That can be achieved in three ways:

  • one is to put the goods and production facilities physically out of reach of the State’s goon squads, a very difficult task if a superpower is involved.
  • another is to place them where it is difficult or dangerous to reach them, perhaps in a place where intervention would put the intervening power in conflict with another power in the region.
  • a third – and the one we will focus on in this section – is to conceal them. This can be done by the usual physical methods, of course, but technology now gives us ways to hide things in plain sight.

To illustrate the concept that I call virtual separation or virtual concealment, consider the following scenario:

A police officer sees a vehicle performing an illegal maneuver, and pulls the driver over to the side of the road. At the police officer’s demand, the driver produces his national ID card, driver’s license, vehicle registration, fuel authorization card, safety inspection certificate and maintenance record. The policeman takes all of these to his vehicle, where they are scanned. In a few seconds, the scanner beeps, verifying that the documents are genuine and current. The policeman then issues a citation and releases the vehicle and its driver. Before the policeman even returns the driver’s documents, the citation is transferred electronically to Police Central, where it is printed out and mailed to the driver’s address. And then a funny thing happens…

The citation is not paid. In fact, it is not delivered, for the simple reason that the address on the envelope doesn’t exist. In due course, an arrest warrant is issued and the central ID database is interrogated to get particulars of the individual to be arrested. And then another funny thing happens. It seems there is no record of this individual, anywhere. An alert is put out to stop his vehicle, but it seems that the license number does not correspond to anything in the database.

In fact, the real vehicle that was stopped and the real person who promised to appear in Traffic “Court” still exist, traveling the same roads and living in the same neighborhood, but manipulation of a central database has made them virtually invisible and untraceable.

The right-wing literature in America is full of fearful projections of the day when a biochip will be implanted in every baby at birth, or when government records will be infallibly linked to individuals by their unalterable physical characteristics—retinal pattern, say. These folks forget (or perhaps don’t understand) that all such schemes depend on the integrity and inviolability of a central database where the characteristic—retinal pattern, fingerprint, biochip code or whatever—is connected with other facts about the individual. If that database can be manipulated covertly, the individual can be concealed more effectively than if he were hiding in the bottom of a salt mine. As governments in developed countries become more dependent on electronic data collection and storage, their vulnerability to tunneling from within increases in direct proportion. That vulnerability is of course well known to those concerned with the integrity of databases, and much effort has gone into designing systems that make it as difficult as possible to manipulate databases without authorization, and to make detection and correction of such changes as easy as possible. Such systems must be built and maintained by highly competent people, and as the current crisis of competence hits governments especially hard such people are likely to be found only in the private sector. The day-to-day maintenance of government databases is already falling to less and less competent people, and this trend can be expected to continue as technology races ahead of the bureaucratic mind. Another effect of the trend is that, increasingly, responsibility for maintaining databases falls on junior, low paid employees, because only they have any acquaintance and familiarity with new things. The result is that the higher a supervisor is in authority, the less he knows about what he supervises, so that enormous trust resides IN FACT with a large, diffuse workforce of (relatively) poorly paid people. We have identified three vulnerabilities:

  • to the system programmer who creates the system, and can institute “back doors” or even autonomous routines that accomplish his purposes
  • to the clerk who enters and modifies data in the database, who may simply input false data or delete valid data
  • to the system operator, who may “accidentally” flub a backup, requiring that the system be restarted from a database stored offline.

Make no mistake: whatever the safeguards built into the system, ultimate trust must reside in individual human beings.

I believe that it is only a matter of time before somebody in a key position sees the opportunity to profit massively from the foolishness of centralized bureaucracy. After about five years of careful preparation, he would be ready to exploit his system for profit. Advertisements would spring up on free Web servers all over the world – enroll with us and live free! For a one time setup fee, an annual maintenance fee and a fixed “rebuild” fee charged every time the client’s “legend” has to be rebuilt (to avoid, say, the traffic citation above), the client effectively disappears from government records. Not that there will be an absence of entries concerning him—far from it—but those entries will have no value in oppressing him because they will be secretly under his control. In our hypothetical example above, the driver may choose to pay the fine rather than incur a “rebuild” fee, but if the penalty is serious enough he will report to his “anonymity provider” by secure means, and find himself provided with new plates, new documents and a new database record to go with them. Of course, the anonymity provider’s first client will be himself…

There are some fascinating implications of this. Suppose that one database—manipulated—finds itself in conflict with another that is pristine. WHICH IS “RIGHT?” There is a 50% chance that the valid database will be authoritatively “updated” from the false one. If, as eventually must happen, somebody in power divines that there is massive hanky panky SOMEWHERE, I think there is at least a 50% chance that the bureaucracy itself will try to cover up the breach. Why? Because revealing its existence—when no solution exists—would likely do more damage to the government concerned than would the breach itself. Even more interesting is the fact that, with the right cryptographic and artificial intelligence technique, the fiddle could survive the discovery, exposure and arrest of its progenitor, even if said progenitor decided to cooperate in return for a lighter sentence. The only remedy would be to take down the major systems serving the bureaucracy for an indefinite period, and rebuild them byte for byte from absolutely unquestionable source documents, using absolutely trustworthy people. Who would take the responsibility for making such a decision? Once made, how would it be implemented?

Thus a relatively small proportion of the population of a large developed country could exist among slaves with nearly perfect freedom. To the people with whom they have direct contact, these clandestine freemen would appear perfectly “normal.” John Smith would always be John Smith; his house would look like anybody else’s, and either its taxes would be paid every year right on time (according to the electronic record, at least), or it would be owned by a tax-exempt entity. His midrange car would have current license plates (who notes numbers?). His children would go to school, but nobody would be very clear on just where they went. He would go to work, but nobody would know just where or for whom. Most of his social contacts would be with his own “set” – others like himself – which would give him a reputation for snobbery or standoffishness. In due course it would become painful for John Smith to conceal the enormous untaxed wealth that he had accumulated behind his façade, and he would in due course “move” and appear in a much more prosperous guise somewhere else.

Older Posts »

Powered by WordPress