Wednesday, June 29, 2005

A Great(?) Convenience:

These days your keyring can include a little fob with 256MB of memory, plus a bunch of mini-tags that act as shopping cards and library cards. I added these mini-tags to my keyring for awhile, then took them off again. What happens, I asked myself, when I lose my keys?

My concern is that the data on the memory fob, or even the bar codes on those tags can help someone figure out where I live and what they can open with those keys.

NO THANKS!

Tuesday, June 28, 2005

Professional Snoopers or Lusty Teenagers…

Please read this quote and imagine trying to making sense out of it twenty years ago:
“Whether this worm is the work of professional snoopers or lusty teenagers -- it's hard to say …”
From a story about the Rbot-GR virus that tries to take control of your PC and send movies from your open camera, perhaps in your bedroom, to the hacker. See this ZdNet story, august 23, 2004, and note the fifth paragraph.

Monday, June 27, 2005

The Grokster Case – The Supremes have spoken. Oy.

The Grokster Case is a complex copyright case, and the Supreme Court decision in the case is also complex. But opinion seems to be quickly coalescing, sadly, on one point. The Supreme Court has created a new legal test, and any company that wants to develop a product related to any aspect of copyright will must be prepared for an immense legal cost to avoid the pitfalls of this new test. Any American company, that is. European and Asian companies will be free to innovate without this new cost.
Companies will now be required to avoid any appearance of encouraging customers to infringe on copyright (whatever that means in the future). Innovating companies must be prepared for a high cost of proving their intentions when sued by anyone with deep pockets who wishes to slow them down. Cory Doctorow put it very well in this excerpt at BoingBoing.

Sunday, June 26, 2005

Only two minutes late...

Driving to my haircut appointment, I thought I'd be about two minutes late, so I used my trusty cellphone to call and announce my tardiness. Of course the receptionist said "no problem." That left me casting about for a good reason - why should I call to tell anyone I'm going to be late by such a small amount?

It turns out there's a good reason, at least in my case. As soon as I made the call, I relaxed and started driving more slowly and carefully.

Friday, June 24, 2005

My money, your cause; go ahead and shoot!

Seth Godin proposes a series of ethical conundrums in his blog. After reading them through, I’ve decided that this sort of thing explains why the Almighty gave us the natural ability to think irrationally.

Thursday, June 23, 2005

“Musician Donations Anonymous “, A modest proposal:

I propose a website, “Musician Donations Anonymous”, that works as follows: Musicians who write and perform songs will register their recordings at the site. People who enjoy bootleg copies of this music will use the site to make voluntary micro-donations to the artists. I’m thinking maybe $.05 or $0.10 per song. Or, for those who have a legal copy and just want to encourage the musicians, say, $.02.
My proposal would be a good test of the much proclaimed good intentions of those who listen to bootleg music. It would also be quite a test of micropayments. And, by bypassing the current owners of the music (usually NOT the artists these days), it should drive the RIAA into a frenzy.

This proposal was inspired by Marybeth Peters’ Bombshell proposal before congress to revamp music licensing in order to enable big music companies to sell singles cheaply enough to compete with the music pirates; and also by commentaries on her proposal by Ari F. at Grafodexia and Ernie Miller at Corante.
Marybeth Peters is the Register of Copyrights. She proposes to fix injustice from within the system. I propose to correct it from without.

Wednesday, June 22, 2005

The game of Skat lives!

It always fascinates me to discover that hundreds of millions of people do something differently from what I assumed was the one and only natural way. For example in Asia, the clapper of a large bell is often outside the bell.

Skat is a card game with a number of whimsical differences that set it apart from the games we are used to. I won’t try to teach you the game nor try to persuade you to play, it; I just want to tell you enough so that you can enjoy the interesting differences with me.

My father taught me to play what American card game rulebooks describe as a variant – Räuber Skat. I knew that Skat in all its variants was wildly popular in middle Europe before WW II. I hadn’t thought much about it lately. But I just discovered that there’s a Skat game for the Palm PDA (strictly in German), and it sells for about $40, an incredibly high price for a PDA game (But I found it for 13 Euro in Europe). So it must still be quite popular. The PDA version follows official German rules and seems to be similar to Räuber Skat. That’s not surprising, as the “main” game, with its Guckser and Tournée bids is less skilful. The enjoyable peculiarities of Skat reside in:

  • The rank of the cards.
  • The scoring.
  • The style of bidding.

Skat is played with a 32 card deck, four suits of A K Q J 10 9 8 7. (I’ve just listed the cards, from high to low, in “French” order.) Skat is always played by three people, dealing 10 cards to each and putting the two other cards in a “blind” that may be used or ignored. There are three kinds of contracts: Null, Grand and Suit. In a Null contract, the ranking of cards is French, and the bidder undertakes to lose all the tricks. In a Suit contract, the trump suit is (from high to low): J of clubs, J of spades, J of hearts, J of diamonds, and the seven remaining cards of the suit, in German order (Ace 10 K Q 9 8 7). (The game of Schapfkpof takes this idea to an extreme, with the four queens ranking above the four jacks.) In a Grand (pronounce it: Grahhhnd), the four jacks ARE the trump suit.

The Null contracts have fixed values of 23, 46 and 69 (there are three ways to try to lose all the tricks, and you get more points for doing it the hard way).

The other contracts have fixed values (each suit is different) that are multiplied by another number. This other multiplier partly reflects the difficulty of your bid. For example, undertaking to win 91 of the 120 card points in play (not just the majority) while not looking at the blind gives you 1 for game plus 1 for no blind plus 1 for Schneider (the 91 points) = 3. Now suppose you also have the three highest trumps (J clubs, J spades, J hearts but not J diamonds). Then you are “with three” so your multiplier is 3+3 = 6. But strangely, if your highest trump is the J of diamonds (so it will be a lot harder to win tricks) then you are “without three” and you get the same multiplier, 3+3 = 6. (to complete the example, if you are playing a Suit contract in spades, the value of your contract is 6 * 11 = 66).

During the bidding, each player who thinks they can fulfill a contract tries to bid the lowest number that the other players cannot equal. There is a certain amount of “shtupfing” (sacrifical bidding, higher than you can probably make), but I think that shtupfing only makes sense if you suspect one of your opponents has a better hand than they realize.
The first bidder first has a conversation with the second one, something like this:
First: 18?
Second: Good.
First: 20?
Second: Good.
First: 22?
Second: Good.
The Second player has answered the First each time by saying that he, too, can play a contract worth that amount. In other words, in Skat, you can overcall a bid by bidding the same amount, not (in all the games you may be used to) a higher amount. Also,note that bids are questions, not declarations. (Some other European games like Frog have the same convention.)
At this point the first bidder can go no higher, so the second turns to the third:
Second 22?
… and so on.
Different, huh? And let me tell you, it’s exciting playing a Grand “without two”, or realizing that you can make a null ouverte revolution worth 69 points. (In that contract, you undertake to lose all the tricks even though your opponents exchange two cards before the play begins, and you play with your cards face up on the table.) An interesting polyglot of a card game!

Tuesday, June 21, 2005

If there were bookware patents…

Something really important and worrisome is happening in the EU regarding software patents, but I won’t bother you with that story. The ministers who want software patents seem to believe that they are needed to protect copyrights, which is just totally misguided. Richard Stallman wrote an article in the Manchester Guardian in which he explains how patents would affect books, if there were book-writing patents similar to software patents. His article is almost entirely non-technical, and really funny unless you’re a software developer, in which case the whole business is just too painful for words. I recommend that you enjoy his column here. A quote:
Here's one example of a hypothetical literary patent:

Claim 1: a communication process that represents, in the mind of a reader, the concept of a character who has been in jail for a long time and becomes bitter towards society and humankind.


Personally, I just want the patent for the concept of telling a story in the first person. Oh, I also want the patent for telling a story as a series of letters. And the patent for telling a story in the form of a diary. And the patent for using a whale as a symbol. And the patent for using an ellipsis to suggest there’s some sex happening … No wait, what I REALLY want is the patent for the concept of keeping other people from making unauthorized copies of a book by placing a copyright notice in it. Oh, the royalties ...

Monday, June 20, 2005

Unintended Consequences:

I've argued that congress should make no law to control any aspect of software or the internet, for the simple reason that it's impossible to anticipate the effects such laws will have, especially since the actual use of each piece of software can be so different from the intended use.
However I was myopic. Unintended consequences are simply the order of the day, and one must be humble about claiming to know the implications of any change. I've been reading about how car thieves who used to hotwire cars now steal them from their owners at gunpoint, since it has gotten a lot harder to start a car without its key. And there was the awful case of the gang who chopped off a car owner's finger to steal his Mercedes, which could not be driven his fingerprint.
And here's a remarkable unintended consequence: Mexicans used to sneak across the border to work as illegal aliens for awhile, then sneak home to their families. But it has gotten a lot harder to cross the border, so now entire families sneak over and live here illegally - permanently - in order to minimize their border crossings. The Tucson Citizen has a report on this, Illegal Migrants staying longer. To paraphrase Abe Lincoln, you can fool all of the people some of the time, but once they think of an angle, they'll fool you right back.

Sunday, June 19, 2005

Are you, or have you ever been, sick?

The New York Times has an article, briefly available here, that thoughtfully considers the following issue: Mental health authorities say that 50% of the population will develop mental disorders in their lives; so one must ask where mental health ends and illness begins. The article seems to address a sensible question unless you point out (as the article does not) that this emperor has no clothes.

If you started an article by mentioning that at some time in their lives, almost everyone will have a physical illness, no one would worry about whether that claim impinged on our perception of good physical health. The real issue underlying this article, which I believe it never mentions, is this: why is the diagnosis and treatment of non-physical illness so far behind the physical?

We have nearly scientific methods of diagnosing many of the physical illnesses, and pretty clear ways of deciding when some of them are cured. We can test for bacteria, white blood cells and antibodies. Most physical illnesses are routinely handled by one style of treatment. (In contrast, the Medpundit blog quotes Chekhov’s saying that "When many cures are offered for a disease, it means the disease is not curable." Most mental illnesses still have wide varieties of treatment.)
Diagnosis and treatment of mental illness seems to be a hundred years behind the standard for physical illness. When is it going to catch up? How is it going to catch up? When will we understand mental illness well enough that our perceptions of it will be more like our perceptions of physical illness?

Friday, June 17, 2005

At last, they get around to MY drink!

After years of reading about how red wine protects against cancer, beer protects against cancer, green tea protects against cancer, potato juice protects against cancer, WD40 protects against cancer, we finally get serious:
Single malt whisky 'can protect you from cancer', conference told. "Single malt whisky can beat the threat of cancer, thanks to high levels of a powerful antioxidant that kills cancer cells ... according to research, single malt whisky contains more ellagic acid [whatever that is] than red wine ..."

However, Lesley Walker of Cancer Research UK was dubious. "Ellagic acid ... can also be found in soft fruits."

Peel me a grape, an ear of barley, AND a cask!

Thursday, June 16, 2005

Semi-technical decisions, More Ways:

(You'll find my piece about making decisions just below; now it's anecdote time...)
At one time our manager sent the entire group to learn to be Empowered. The idea is simple: if only the manager can make decisions, then a large group can barely be more productive than one person. Empowered employees solve most problems on their own and pass just a few up the command chain. We came back full of excitement and proceeded to act empowered. This meant that we made decisions, the effects of which came back to our manager without having first been approved by that manager. Each of these produced an explosion of anger and discipline. We were dis-empowered a week later.

Some of my worst decision-making experiences occurred when the manager had risen from techy ranks. One director of this type usually engineered a solution on the spot, when faced with a serious problem. In the quick moment, no one could argue with him; but in practice his fixes usually failed due to lack of familiarity with relevant detail. The crisis brewed for a few wasted weeks and the troops returned to him with the same problem. One might also say that when the manager is technically adept, the communication barrier gets pushed up one level.

Tuesday, June 14, 2005

Semi-technical decisions: Two Ways:

(You'll find my piece about making decisions just below; now it's anecdote time...)

I worked for one manager who was superb at making decisions based on the very technical information his group gave him. It was instinctual with him, and I think he reacted to a sense of how his people felt, since he clearly could not understand the abstruse critical details. Yet I never saw him make a bad decision on a technical issue.

Another successful manager I know listens to his people’s technical discussions and then almost always leaves them to make the decision. The reason this works is that he has always been brilliant at selecting staff that he can trust this way. Many people will tell you that NATURALLY the best course is to leave the decisions to the techies, but that doesn’t work if they are Savant Idiots (a common malady among us techies).

Sunday, June 12, 2005

How do semi-technical decisions get made?

In a May 10, 2005 NYTimes column Kent Sepkowitz, M.D. tackled the difficult issue of whether doctors should withhold information from patients, and who is in charge when it comes to planning treatment. (The column, here, is already in the Times' paid archives.)
This issue is part of a larger topic, one that is quite familiar to many of us. In the business world and elsewhere, a decision is often made based on technical information provided by someone deeply familiar with the techy issues, and a non-technical person with knowledge of the issue, but little understanding of the techy stuff. It is usually close to impossible for these two to communicate, and yet they must, to achieve a good decision. (Of course the decision-making process gets worse when groups of people represent each camp.)

In the business world, the techy is, perhaps, the programmer, engineer or artist, and the other decider is a manager; the manager usually makes the final decision. In the medical world (until recently) the techy was the doctor and the patient had the direct, first-hand knowledge; but the techy used to be the one in charge. (Sepkowitz raises some doubts about how often the patient should be in charge.)
If you will relate this problem to any area in which you have technical expertise, it becomes clear at once that of course you do not tell the other decider "everything." It's counterproductive for either person to drown the other in all the relevant detail one can think of. How do you choose what's relevant without biasing the decision? Or should you bias the decision?
This decision process fascinates me. In every large company, as you go up the hierarchy, you will find barriers where the person above cannot possibly understand most of what the people below have to say. How does communication cross these barriers? And every time I need to make a difficult medical decision, I know I’m facing the same problem. I think that:

  • The techy's challenge is to present the important information without overly biasing it.
  • The other's challenge is to figure out how to react to the techy.

Now perhaps you think I SHOULD have said this:

  • The techy's challenge is to present the important information in a clear non-technical way.
  • The other person's challenge is to know enough about the field to understand.

But I didn't say that, because my experience is: it would be wrong. Illustrative anecdotes will follow!

Friday, June 10, 2005

The Good Software House:

The User Interface Hall of Shame has a lovely web page about how awful it would be to have a house designed by bad software developers. But what if your house were designed by good software developers?

  • Instead of painting or expensive redecoration, you can change the look of the house by adjusting a few attributes or loading a different skin.
  • If your house's location bores you, you can assign new values to its properties.
  • When you need to enlarge the house, just download a few extensions.
  • The Help function is fully implemented, providing housekeeping, garbage collection, pruning, trimming, clipping and even marshalling.
  • Rather than have multiple bathrooms, the room has a Bathroom Class. Anyone in need invokes a bathroom object and simply deletes it afterwards; no need to wait for a free bathroom, and no need for routine bathroom cleanup.
  • The house is persistent; when a hurricane threatens, Save the house, and (if necessary) Restore it afterwards. (You should also Save the house before making any mods.)
  • Any viruses in the house are immediately quarantined. Intruders have little hope of getting past the surrounding firewall.
  • The house remembers actions that you perform and offers to complete them for you when you start to do them again.
  • The house is full of widgits you can tap to bring up relevant tools and implements, for the task at hand.
  • You can move your windows around and resize them to take advantage of the sun, minimize them if the sun's too bright, or pop up an extra window for more light.
  • But when it's too dark, you can click one window to highlight it.
  • You can temporarily enlarge one room to the full size of the house by double-clicking it.
  • If you'd rather your house did not have an address, you can have an indirect address.
  • The guestroom template instantiates automatically for many different types of guests.
  • Your house is fully commented. But if you forget what anything is for, just tap it with your middle finger to see the list of its functions.
  • The closets are hash-coded for rapid storage and retrieval.
  • The refrigerator is implemented with a linked list and compression, giving it nearly infinite storage capacity. A Least Recently Used algorithm controls its automatic garbage collection.
  • You can use the "Find" function to locate anything in the house. (Be sure to save the search criteria you'll use often, like "find reading glasses.")
  • The bookshelves, closets, storage boxes, etc. can each be told to sort their contents by size, age, weight, name or value.
  • There are many interesting trees in he house, you'll enjoy their pretty nodes and leaves.
  • There's an exercise room fully stocked with incomplete lists, sorts, bags, collections, virtual classes, prototypes and pseudo-code to provide regular mental exercise.
  • When you're tired, you can arrange to sleep for any specific number of milliseconds.
  • There are several generic vehicles in the garage as well as their subclasses. You wouldn't drive a generic vehicle of course, but you might, say, own just one really good radio, put it in The Generic Vehicle, and then all your cars, trucks and planes will have it too.
  • You'll find that the edges, floors and ceilings are in particularly good shape, for these cases were the focus of most of the software testing.

If anything goes wrong with the house, no expensive repairs are necessary, you just restart it. (You can restart it in Safe Mode if a klutz is visiting.) I must admit though, that the upgrade to House 3.0 is expected to cause pain. Your neighbors will admire your house and want to see all of it, but unfortunately it's too late to give them a walk-through; the walk-throughs were completed during the early stages of construction.

I'm just tipping the iceberg here. Please comment to add improvements to the Good Software House!

Thursday, June 09, 2005

Motive is Everything?

We all take risks from time to time, but it's worth it to us when we have a good reason. Take Gary McKinnon, who stands accused of causing the US government $1 billion of damage by breaking into its most secure computers at the Pentagon and Nasa. He is likely to be extradited to America to face eight counts of computer crime in 14 states and could be jailed for 70 years. Friends said that he broke into the networks from his home computer to try to prove his theory that the US was covering up the existence of UFOs. I quoted most of this paragraph from the story at the 'This is London' website. I hope he found the evidence he was looking for. It's sooo hard to find a Lack Of Evidence...

Wednesday, June 08, 2005

Genuine, umm ...

As computers get better at copying things, especially in 3D, our sense of what is genuine and original is going to be under attack. But I think that in this great battle of artistic standards, we can leave the in-store ad I saw today on the sidelines. The new idea here is that you bring your photograph to the store, and it gets converted into a painting that features genuine artistic digital brushstrokes. I'd like to set myself up as an expert able to distinguish the genuine artistic digital brushstrokes from the fake ones, but I doubt there'd be much call for my services.

Tuesday, June 07, 2005

Wastebaskets for Packrats:

I had trouble finding wastebaskets in my local Target store because they were not where I expected to find them. They weren't even where several Target employees expected to find them. But if you're a packrat, you'll agree that Target is definitely on to something: wastebaskets are in the storage department.

Sunday, June 05, 2005

The Placebo Affect*

Seth Godin has created the term "Placebo Affect" to describe something that Marketers do - they sometimes manage to take any old product and give it caché‚ by interweaving it with a story, an idea, that catches on and inspires. He says:

"We don’t like to admit that we tell stories, that we’re in the placebo business. Instead, we tell ourselves about features and benefits as a way to rationalize our desire to to help our customers by allowing them to lie to themselves."


As a firm believer in the Placebo Effect, my feeling is that Seth Godin has perverted a fine concept. Placebo pills help people to harness their bodies' natural healing abilities. Godin's "Placebo Affect" helps people to believe that yet another same-tasting vodka is really special. There's more difference than similarity here. Much more.

(Here's a web site discussing the placebo effect in some detail. But it approvingly mentions the Danish study that basically added 114 placebo studies together, and concluded that there is no such effect. Other placebo studies have shown that some ills are affected by placebos and other kind sof ills are not. By adding placebo studies of both kinds together, the Danish study produced a statistically meaningless total that fails to debunk the placebo effect.)

Friday, June 03, 2005

Silly Loooonghooorn (1):

We’re getting demos and hype of Microsoft’s upcoming LongHorn operating system. Some of the buzz would put me to sleep if it weren’t so pitiful. E.g.:
“For example, saving a file in Longhorn brings up fields for user-defined file metadata such as the author, the title of a document, the date of creation, the document's category, and keywords; if completed, these metadata fields can later be used to speed up and improve searches. By displaying the fields instead of hiding them in a Properties dialog box as Windows currently does, Longhorn presumably will encourage users to supply the metadata.”
Your word processor already has these fields. Do you fill them out? Would you fill them out if no one were pointing a gun at you? What a pointless idea.

On the other hand, Longhorn will allow you to make screen icons so large you can actually see them on a high resolution screen. That would have been a great idea in 1995, and it’s still a great idea.

Thursday, June 02, 2005

The clear symbiosis between illegal music downloads and the music market:

Cory Doctorow has stated it (at BoingBoing) with such remarkable clarity that you can hardly think otherwise: the trade in illegal audio downloads is exactly what is fueling sales of songs. First the numbers, from Forbes:
"According to the NPD Group, consumers purchased 25.9 million songs in March 2005, 52% more than they bought a year ago. Consumers also downloaded more than 242 million songs illegally this March, up 25% from March 2004." Then Doctorow's observation: "If the only way to load 10,000 tracks on an iPod was to buy them at $0.99 each from the iTunes Music Store, would there be much market for iPods? And without the market for iPods, what happens to the Music Store?"

So the availability of no-cost music causes people to buy players with large amounts of storage, which they then fill, at least partly, with purchased music.

Bear in mind also, the "worst" cost (to the music industry) of those illegal music downloads. This calculation changed just recently when Yahoo opened their new music service. Yahoo charges about $6/month to download all the music you want without DRM restrictions. So in general 100 people who download 100 songs a month for a year could have paid yahoo $7,200 to do the same thing legally, or about $0.72 per illegal copy.

In about 1984, the company I was working at decided that they were buying far more floppy disks than they needed. Every programmer needed a bunch of floppies, but the purchases were high enough to suggest strongly that many people were just taking floppies home for their personal use. Management set up a "checkout" system, where you had to sign out every single floppy and indicate what you would use it for. At that time, floppies cost us about $1 apiece. We complained that the people time necessary to check out the floppies would cost the company more; and of course developer time is precious. After one or two cases where a developer who had been kind enough to work till midnight could not check out a floppy from the locked cabinet - and delayed a deadline - the checkout system was dropped.

The draconian DRM systems the music companies are developing to protect their precious audio from copying seem to fall into the same category as the checkout system. The current lost sales are a small cost of doing audio business. Enforcing DRM will cost us and inconvenience us and damage the audio market. We've got to protect this industry from itself until they come to their senses.

I'm not condoning illegal copying. Here's a suggestion: the big audio companies should do what one classical pianist does, and release free tracks of music in unpolished, noise-afflicted form. There would be lots of live performances and recording session tracks. Then people would buy the finished, high quality tracks of the music they like.

Wednesday, June 01, 2005

Why is there such a difference? (Hardware & Software)

I previously blogged about a useful, but rather delicate and gimmicky eyeglass cleaning machine. Now let’s think about the very different requirements for user interfaces of software and hardware. In software, the most common goal is that no matter how the user manipulates the thing, it will act reasonably and not break.
Can you imagine someone complaining that they pressed all the buttons on a telephone (or a car, or an electrical service panel) and it did something unacceptably weird?

Some hardware products are required not to pose a death or disfigurement hazard (a rare requirement in software), but breakability is mostly the user’s responsibility. That’s amazing, considering that software can usually be restarted with a reboot, but a broken razor is useless or will cost money to fix.
The difference in our expectations for these two kinds of products is intuitively obvious, but I think it would be difficult to put in words. (And we have a yet another set of user interface requirements for books.)