The Coming Fall of the Movie Business

Bill Wyman's recent article on Slate, Groundhog Decade, predicts that the video content industry is going to get smashed up in the coming decade as badly or worse than the music industry was in the last. As he says,

The trouble facing the movie industry right now is the same one the music industry had to confront 10 years ago. This is the summing-up sentence I referred to above:

The easiest and most convenient way to see the movies or TV shows you want is to get them illegally.

This is particularly notable because it's clear from his article that Wyman is not an especially technologically literate guy. (He doesn't seem to entirely understand the interaction of PnP media servers and his PlayStation 3, for example.) When the non-geeks start to find DVDs annoying because they've found something better, you've got a real problem on your hands.

The article's worth a read.


How I Use Anki to Learn Vocabulary though Spaced Repitition

The biggest barrier by far between me and using Japanese effectively is lack of vocabulary. I'm very, very poor at memorizing specific things (as opposed to knowing that something exists and where to go to look up the details), and one of the keys to learning a foreign language, especially one as different from English as Japanese, is extensive memorization.

I decided recently to tackle this problem. To this end I've started using a flashcard program called Anki which uses spaced repetition. While there's a good bit of detail on the Anki site on how it works and how to use it, I found that certain important (to me, anyway) pieces of information were missing, and it took me quite a while to figure out both how Anki really works and how to customize it for my particular learning style.

There were two major issues I had with Anki: it's not configured by default for fairly intensive study (many sessions per day), and it didn't give me, a very poor memorizer, enough repetition of new words early on.

Anki seems mostly oriented around studying once a day, or at least studying any particular card no more than once a day once you've got it right once that day. I prefer to do a lot of short study sessions throughout the day, and have newish cards repeated in several sessions during the day even when I'm getting them correct. (I simply can't see a brand new word a few times and expect to remember it 24 hours later without some intervening repetition.)

I also found that with the default settings, I was running out of material to study after two or three sessions, frequently ending up at the "done for the day" screen, where my only option was to review early if I didn't want to learn some new cards. I've seem to be able to handle only 3-5 truly new items (as opposed to items new in Anki but that I've seen before elsewhere) per day before I get overloaded and can't remember any of the new material, so doing more new cards wasn't really an option for me.

The "only daily" issue was fixed fairly easily in by choosing Settings / Deck Properties... and clicking the "Advanced" tab. Unchecking "Per-day scheduling" made cards come due after the specified number of hours, rather than having them all come due at the beginning of the day even if they'd just been last reviewed a few hours earlier. I then set the initial button intervals down to hours instead of days. Button 2 I set to 0.12-0.16 days (3-4 hours) and button 3 to 0.33-0.5 days (8-12 hours); these mean that once I get a new card correct, I'll get another review (or perhaps several) the same day rather than having to wait until the next day. Button 4 I leave at 3-7 days, however; I use that for cards that are "new" in Anki but actually words I know quite well already.

These settings affect the initial interval, but the changes after that are based on fixed percentages of the old interval, and these can't be changed. The key here is to ignore the "good," "easy," etc. designations on the buttons and instead think of them as indications of how long you'd like to wait to see the card again. Generally, I consistently use button 2 (which gives a 20% increase) for my answer until I'm sure I want the card to go into the "review only every once in a while" category. Even there, if I feel the interval's getting a bit long, I hit 1 the next time the card comes up (even though I got it right) to reset the interval to the initial interval, whence I can extend it as quickly as necessary with the 2, 3 and 4 buttons. (If you do this, make sure that the "Button 1 multiplier" setting is set to 0% and the "Mature bonus" is set to 0 days, or this may not work.)

The next issue for me was how to get much, much more initial review on a new card until I'd really felt that I'd learned it. Far too frequently I was getting a card right, marking it so, but then discovered that a day or even a few hours later it had already slipped out of my mind.

For this I use the 1 button even after I've got the answer right. This keeps the card coming back every study session until I really feel I've gotten it burned into my memory. In other words, I consider the 1 button to mean "keep showing this to me frequently" rather than "I got it wrong." For this to work well, the "Leech threshold" in the deck properties needs to be changed to 99 (the maximum), and turning off "Suspend leeches" is probably a good idea as well. This I found to be no big loss; I can still easily suspend a card I feel is becoming a leech by just choosing "Suspend" from the menu when that card comes up.

The "Button 1 delay" in the deck properties also needs to be tweaked appropriately. This interacts with the session limit in the "Timeboxing" tab of the Study Options. I use three minute sessions and set the button 1 delay to five minutes; this ensures that I won't be shown the card again until a later session. It wouldn't be unreasonable to set this to an hour or more if you're doing short sessions more often, and don't want the card to come back every session.

So, with these settings, my general rule is to use the 1 button until I feel I've got the card known well enough that I feel I'll remember it for at least a day, the 2 button until I can answer the card pretty much instantly, and the default (space) button once I've reached that stage.

This means my reviews would be considered "inefficient" in Anki terms: I'm doing a lot more review on individual cards than I would be otherwise. This isn't really as big a deal as some would make it out to be; once I've got more or less instant recall on a card, a review only takes a couple of seconds if that. Thus, even if I'm doing a couple of hundred more reviews per day than I'd be doing otherwise, this adds only a few minutes a day to my study time. Trading a few minutes of unnecessary study for more quickly building very good recall seems well worthwhile to me.

New cards I tend to add quite slowly by default: 3-5 per day. The issue here is that I get overloaded quite quickly if there's a lot of truly new material. How much is really new can be hard to control, since my deck does include a lot of cards marked "new" that I already know (see below for more on this); I deal with this by finishing my reviews over the course of the day and then, at the "finished for now" screen, choosing "Learn More" and continuing with adding "new" cards until I've seen enough truly new ones that I'm satisfied.

If you want to quickly bulk up your current working set with things you already know from your deck that are still marked new, a handy way of doing this is to run through new cards this way, choosing 2, 3 or 4 for the ones you know, and use the Edit / Bury Fact option on the new cards you don't want to add. That option will temporarily suspend the card; quitting and restarting Anki will remove the suspension.

So this is how I use Anki. It may not be quite how it was designed to be used, and I've been told that I'm doing it "wrong" in terms of spaced repetition theory. But once I'd gotten this setup figured out, and started studying in 3-minute chunks every half hour to two hours throughout the day, my retention improved immensely.


The Sony Reader Finally Reaches Japan

I was in Bic Camera yesterday, and much to my surprise it seems that Sony has finally released its Reader in Japan. What can I say but, "It's about time."

It's unusual for Sony to release a product in North America much before it's released in Japan; particularly unusual is this four-year gap. (The first Sony Reader in the U.S., the PRS-500, was released in September 2006.) But the story is even more odd because Sony had released a similar E Ink reader, the LIBRIé, in Japan years earlier. I first saw this at Bic Camera, too, but it vanished after a few months.

Was it so poorly accepted in Japan, this mecca of electronic devices and novels published on cell phones, that Sony gave up on the idea here for half a decade, even as the concept roared to success in North America?

I've been through several Sony Readers (a PRS-505, a PRS-700 and now a PRS-350) and one of my complaints has always been the lack of Japanese support. These new models in Japan do have proper Japanese support (including reading vertical columns from right to left—the traditional Japanese format) and have a bilingual menu system (Japanese/English). The price, at about ¥18,000 for the PRS-350, is slightly higher than in the U.S.; this isn't suprising as most things are more expensive in Japan, even when made here.

I wonder if it will be possible to upgrade my U.S. reader to the Japanese software? I'm suspecting at the least that it won't be supported; it may be the case as well that the hardware is slightly different, having more memory for dealing with the much larger set of font data required for Japanese.

Regardless, it's still missing the other thing that would make it truly worthwhile for me to use it to read Japanese books: a good dictionary. It does in fact come with one dictionary, but it's English-Japanese, which is not terribly useful for me. I'm surprised that it doesn't even come with a Japanese dictionary, though not so suprised it doesn't come with a Japanese-English dictionary.

Still, they may add other dictionaries one day, or at least add the option to buy them. (Electronic dictionaries are a big business in Japan, mostly in the form of dedicated devices.) If so, it would probably be worthwhile for me even to buy a new Reader for this functionality.


How Cameras Work: Part I

I've been an amateur photographer for a few years now and, being of a technical bent, learned far more now that is ever good for me. So when my brother recently bought a DSLR camera to take pictures of his kid, he asked me for a few pointers. I thought I'd put the write-up up here for the benefit of others as well.

Now there's gobs and gobs of information on the net about cameras and taking photos already, of course, which I encourage everyone to pursue, but there's not so much on the surprisingly simple technical details of taking a photograph. Since the dawn of the digital age especially cameras have been getting more and more complex, but, from the user's point of view, this complexity is all just layered on top of the same basic settings we've been manipulating for close to two hundred years now. So here I'd like to clear through the cobwebs of all of those menus and settings and show you the simple bits underneath. Understanding that should help you understand the rest of what your camera is doing.

A camera consists of two basic parts, a body and a lens. These may come as one indivisible unit, such as in a point-and-shoot camera, or the body and lens might be interchangeable, as with an SLR or a DSLR camera. A camera with interchangeable lenses is almost invariably going to be of reasonably good quality (because there's no market for bad ones); non-interchangeable-lens cameras vary from as good as a good DSLR to complete rubbish.

The camera body's primary purpose is to be a light-tight box with a sensor (either film or a digital device) that can record images and a shutter that can fairly accurately open for a set amount of time and close again. That's it. Bodies often have many, many more features, but all the rest beyond being a light-tight box with a shutter are just gravy, and unnecessary for good photographs. On digital cameras these features often even get in the way of taking good photographs rather than helping, for reasons we'll see later.

The purpose of the lens is to focus light from outside on to the sensor, and on almost all cameras this is the key determinant of image quality. If you've got a poor quality lens, you'll get poor quality pictures, no matter how good the sensor. If you've got a good lens, you'll be able to take pictures as good as the sensor allows.

In the case of film, any modern film will record great images. Some of my best and sharpest pictures have been taken with an old compact film camera (an Olympus XA) simply because the camera had a great lens. For digital, well, of course it depends, but as I mentioned above, pretty much any major manufacturer's DSLR from the past five years will be pretty darn good. (Compact cameras, not so much.) Note that there's a lot more to a digital sensor than megapixels. People love comparing that, because it's a simple number that's easy to compare, but I've got plenty of good photographs that I've blown up to 10x15 (inches) and larger taken on my 6 MP Nikon D40; in terms of quality and the ability to work in low light it will easily outdo all or almost all 10-12 MP point-and-shoot cameras.

As with bodies, modern DSLR lenses tend to be very good as well, even at the low end. The kit lens that came with my D40, a cheap zoom that retails for under $150 separately, is surprisingly good, especially given that it's a zoom lens. Making a zoom lens, as opposed to a prime lens (one without zoom) entails a lot of compromises, which is why a $100 prime lens will often produce better images in lower light than a $500 zoom lens. But if you've got a DSLR, unless you're doing a lot of low-light shooting, the standard kit lens will do just fine for a while, by which time you'll know what else you want and need.

So how does all this work? There are really only four settings you're manipulating, though digital cameras, especially in "automatic" modes, can put a very large amount of mysterious crap between you and those four settings. Here, again, is an advantage of DSLRs; they invariably offer much easier access directly to the settings so that you aren't sitting there futzing with menus or waiting for your camera to figure out what's going on while your photo opportunity goes away.

Let me mention first something about exposure. One of the primary technical things you need to do when taking a picture is to make sure that the right amount of light gets to the sensor. The sensor records only a limited range of light values, and if you put too much light on it, called overexposure, you'll get a very "white" picture where areas that you saw as different brightnesses (details within clouds, for example) all turn into one flat white value as the sensor overloaded. At the other end, if you don't have enough light on the sensor, areas in shadows where you saw details will come out as a flat black area in the photograph. Finding the proper exposure can be tricky because, unfortunately, your eyes see a much, much wider range of light values than a camera does. So you usually need to select a subset of the range that you see to record as an image. If you want to capture detail in the brighter parts of the image, often you'll lose detail in the shadows, and vice versa. So if you see an image where the side of someone's body blends into a white wall, don't necessarily think that the photographer did a bad job; it could just be that there were some darker parts of the image where he wanted to capture the detail rather than having it be a flat black.

The four basic settings that determine a photograph are focus, aperture, shutter speed, and sensor sensitivity. That's it. There's nothing a camera, no matter how sophisticated, is doing that's doing anything but, in the end, manipulating these four settings. (Well, in digital cameras there's actually a bunch of work going on related to how digital sensors work, image compression, and suchlike, but that, except in the most esoteric of circumstances, is something generally best left to the experts who made the camera.)

Focus is probably the most familiar one to most people. What people may not be so familiar with is that focus of an image is not a "yes" or "no" issue; it's often the case that part of the image is in focus and part is not. This is due to the limitations of lenses, and is called "depth of field," usually abbreviated to "DOF." Having parts of the image out of focus is so common in fact that it's now turned into part of the art: photographers will often purposely put the background of an image out of focus to highlight the subject. (This is called "bokeh," from the Japanese for "blur." It's often quite noticeable in motion pictures where you can see the focus change from character to character as each speaks.) This is why photographers generally focus on the eyes in a portrait; faces with the eyes even slightly out of focus tend to look odd because those are what people tend to look at first.

Aperture refers to how large the lens opening is (the "f-stop") and thus how much light the lens lets through to the sensor. With almost all camera lenses this is variable via an iris inside the lens; typical values range from f/2.8 or lower (wide open) to f/16 or more (very closed down). Given that often there's never quite as much light as you'd like, why wouldn't you always just use a large (low f-value) aperture all the time? Well, as with all of these settings, there's a trade-off. A wider aperture does let in more light, but also reduces your depth of field, i.e., the range of distances from the camera over which the image will be in focus. With a very small aperture (e.g., f/16) it might be the case that everything from a meter away from the camera all the way out to infinity is in focus, whereas with a very large aperture (f/1.4, say, if your lens goes that wide open) you might be able to get someone's face in focus but not even her hands, much less the far background. So your aperture setting is both a technical (can I get enough light on the sensor?) and artistic (how much depth of field and how much blur of things other than the subject do I want?) decision.

Note that brightness is perceived and recorded logarithmically, not linearly, so aperture (and shutter speed, as we see below) work in ratios, doubling and halving. The step between f/2 and f/4 is the same size as the step between f/4 and f/8, and the same again as the step between f/8 and f/16. (Actually, a single "stop," as we call the standard step size in photography, is half the step size I described above, or the square root of 2—about 1.414—so a typical set of aperture settings available on a lens would be f/2.8, f/4, f/5.6, f/8, f/11, f/16, and f/22. You'll note that we round off these numbers a bit when we write them down, e.g. f/2.8 instead of f/2.828427....

There is some very interesting physics behind focusing and depth of field of lenses, by the way. I won't go into it here, but it's worth looking up if you're interested in that kind of thing.

Shutter speed refers to how long the shutter stays open and thus, as with aperture, influences how much light gets to the sensor. Low shutter speeds, such as 1/30th of a second (usually marked on the camera as just "30") allow a lot of light; high shutter speeds (say, 1/1000th of a second, or "1000") allow little light before the shutter closes. This is one of the primary means of controlling exposure to make sure that you don't let in too much light, producing an overexposed, very white picture with no detail in the brighter parts, or too little light, producing an underexposed, very dark picture with no detail in the darker areas. But there's also a compromise you need to make here: the longer the shutter is open, the more time things have to move, and so the greater the chance that something will be blurred in your photo. Thus, sports photographers often use quite high shutter speeds to capture an athlete rather than a blur. Even for very still pictures, this becomes critical anyway at speeds of about 1/30 and below, where the movement of your hands and body may blur the entire photo, if you're shooting hand-held rather than using a tripod. In this cases, anything you can do to steady the camera can help: pressing your arms against a table or wall, using the viewfinder to keep the camera pressed against your head, whatever. Also, note that if you're using a telephoto lens (or zoom lens at large zoom), your minimum hand-held shutter speed may be much higher, since the magnification of the image also of course magnifies any movements you make.

Shutter speed interacts with aperture, of course, and is the primary means of getting a choice of apertures. Halving the f-number of the aperture requires a quadrupling of the shutter speed for the same exposure, so if a scene is properly exposed at 1/125 and f/8 I can take it instead at f/4 and 1/500 and get the exact same exposure. I might do this because I'm shooting something moving very quickly and I'd get too much motion blur at 1/125. or I might want to reduce the depth of field to blur the background in a portrait, in which case I might even go further, to f/2 and 1/2000.

Sensor sensitivity also determines how much light you need to make a properly exposed image, and as with the other values, has its trade-off. The more sensitive a sensor, the more noise you'll have in an image. (This is seen as grain in film images, or just "noise" in digital images.) Sensitivity is generally measured in ISO units, and typical values range from 100 or 200 (not very sensitive, but very clean) to 1600 or more. While for a roll film camera you pick your film and are stuck with the same sensitivity for the entire roll, digital cameras have the advantage of being able to change the sensitivity from shoot to shot. (Many cameras even have a convenient automatic mode that will use a low sensitivity when there's plenty of light, but automatically raise it in low light.)

As with aperture and shutter speed, sensitivity as logarithmic; the standard steps are usually 100, 200, 400, 800, 1600 and sometimes 3200 and beyond. But one thing to be careful of is that in the digital world sensitivities and noise for a given ISO vary from camera to camera. For sensitivity this is a fairly small variance; "ISO 200" on one camera may be the same as "ISO 125" on another. But the amount of noise varies a lot: ISO 1600 on a good DSLR may produce an image with less noise than ISO 800 on a point-and-shoot with a smaller sensor. The smaller sensor will, however, produce greater depth of field at the same aperture. And, of course, as yet another part of the trade-off, larger sensors usually cost considerably more than smaller ones.

So these four settings, focus, aperture, shutter speed and sensor sensitivity (ISO) are what you have to work with. Anything else you're fiddling with is just changing one of these. But how do you use them? As this has gotten quite long already, that's going to have to be the topic of a second post next week.


To What Did I Just Agree?

Sony the other day sent me a note (hours before the event) saying that they were changing their PlayStation online agreement and I needed to agree to the new one. No indication of what the changes were, of course, just a lot of text, without even an agreement checkbox, actually. (When and how—and if—I agree to this, I don't even know.)

I hope I didn't give them my immortal soul.

But the problem of these extraordiarially long agreements for simple consumer transactions, and the power imbalances inherent in them, has been well known for a long time and discussed extensively.

Which got me wondering about this power imbalance. It's not just that the other side has a lot more lawyers than I do (an infinite number more, actually, if you consider that to be the result of division by zero), but also that my choices are limited, essentially, to "accept or don't buy that $10 downloadable game." Were I negotiating the kind of contract I negotiate in business, say with a potential employer or employee, both sides are a lot more equal in that I can return a copy of the contract with additions, changes and strikeouts, and suggest that we work from there instead. Negotiations then proceed. But that's not going to happen in the consumer world.

With a consumer contract, since the vast majority of consumers are willing just to click the box regardless of consequences in order to avoid the hassle, the few people who are willing to take the time to read the contract and perhaps disagree with a few terms have no real way to compel the company to change it. Companies with large numbers of consumers can easily afford to ignore those people since it costs them only a few small sales. How can we fix this?

Here's my crazy idea. For these on-line contracts (and perhaps this could be done with paper, too), force everybody to read and truly agree with them by having them, after reading each paragraph, answer a short, simple multiple choice question about that paragraph.

If contracts were less enforcable without this evidence that the consumer had read and understood every provision, a lot more consumers would default to not making the purchase because they aren't willing to deal with the hassle of proving they'd read and understood the long contract. That would cause pain to those who write long and complex agreements, and provide a lot of encouragement to those selling goods of relatively low cost to write short, simple ones instead.

I'm not sure that this is an idea that's ever likely to be put in to practice, but it's an interesting thought.


It Just Got Personal

Through the usual weird six-degrees-of-separation-chain-of-links that happens on the web, I happened to run into the diaries of someone raised by dominionists.

In case you want to get on with life instead of following the links, dominionism is essentially a totalitarian movement to have the world ruled by radical Christians of the U.S. flavour. It's in no way a majority, or even a plurality, but it's far from just a cult.

Reading about this fellow's experiences, I came across a quote from this post that chilled me to the bone:
[O]ne time in 5th grade, I had made a comment that was in fact seen as socially acceptable by the dominionist group that raised me...that I hoped that "godly people took over and did to the atheists what Hitler did to the Jews".
Now think about this. My eyes are blue, my natural hair colour is dark blond, and I'm pretty much your standard person of northern European descent. I'm about as white as they come, and by looks alone I'd probably be quite acceptable to pretty much any western conservative movement, from fundamentalist Christians to neo-Nazis. While I've always appreciated intellectually all of the innumerable forms of hatred for a category, whether it be blacks, gays, Jews, or whatever, all the way up to the plight of the Jews during the Holocaust, I'd never thought of being in one of those groups.

Well, apparently now I am. There really are people out there who would be happy to see me die because I'm an athiest, and they're not the terrorists that Bush & Co. are worried about. Most of them live in "the land of the free."


Citizen Kane, Depth of Field, and 3D TV

Recently I happened to stumble across a fascinating 1941 article called "How I Broke the Rules in Citizen Kane," written by Gregg Toland, the film's cinematographer. One of the "rules" (or Hollywood conventions, as he prefers to think of them) he broke was to use mostly large apertures producing shallow depth of field. Typical apertures at the time were between f2.5 and f3.2; he would regularly use f8. (How he managed to do this is an interesting story in and of it self, and makes the article worth reading.) This let him have much larger depth of field; with wider-angle lenses everything from the foreground to the background was in focus.

To this day this is an uncommon technique in feature films. In fact, we've become so accustomed to a shallow depth of field in film productions that in the last decade or so animated works and video games now emulate this. There's no need for or advantage to shallow depth of field, since in these circumstances there's generally no lens involved; it says a lot that people would go to extra work to emulate what has usually been considered a lens deficiency.

This reminded me of seeing a shallow depth of field in an unusual circumstance the other day that really made it jump out. At consumer electronics stores in Japan lately they now have demonstrations of 3D video systems of the sort that use a pair of LCD glasses. Watching an animated feature playing on one, I noticed that it used the "shallow depth of field" technique. However, this produced a rather odd sensation in a truly 3D image. The image felt more realistic than a standard 2D image in that I had a much stronger sense of being to move my point of focus between foreground and background objects and know just where they were in the 3D space in front of me. But as I did this, I immediately found myself discovering that what was in focus didn't shift with what I was looking at, as it does in the real world. This being both unreal and unconventional gave me quite a jolt.

I wonder where we'll go with 3D, particularly when recording through a lens. Will 3D productions widen the depth of field to avoid the unreal jolt? Or will they stick with the narrow depth of field and let that become an almost unquestioned convention in 3D productions as it is in 2D ones?


Apple & The Dark Cloud of the App Store

Bruce Tognazzini (often known as "Tog"), the well known user interface designer who was once very influential at Apple, recently wrote a column on the issue of editorial control at iPhone App Store entitled Apple & the Dark Cloud of Censorship.

The problem, for those not aware of it, has two parts. First, you cannot download apps for your iPhone (or iPod touch, iPad, or similar device) except through the App Store. (Developers and hackers have ways around this, but the techniques have certain downsides and are more work than most of the general public are willing to do.) Second, Apple's approval process is long, arduous, and somewhat capricious. It can take weeks even to get a bug fix through the process for an application that's already been approved, and many applications that compete with Apple's applications, are of poor quality, or are just not in good taste are not given approval at all. This has been a complaint on developer blogs for some time now, is discussed on Wikipedia, and has even been picked up by mainstream media.

By comparison, handsets and devices running Google's Android operating system don't have these restrictions.

Tog's Take

I certainly agree with Tog that the strict editorial control imposed by Apple is a problem for several different reasons, and I think his solution has some good points, but I think he also goes wrong in many ways in his article.

My less important disagreement is that I don't like his calling this "censorship," and suggesting that this restrains political expression. "Censorship" is a pretty heavy word, applying as it does not only publishers exerting editorial control, but to governments restricting the ability of someone to say something anywhere at all. There are certainly many other avenues in the world for people to publish their political opinions. Should the government, say, force Apple to publish opinions it disagrees with, that would be nearly as bad, from a free speech point of view, as government censorship of opinions. Apple must be free to publish what they like on their platform just as others must be free in the same way. If this hurts Apple's competitiveness (which I think it's clear it does, in certain ways), that's Apple's decision. (Of course Tog, as a shareholder in Apple, has some say in this decision.)

But my greater disagreement is that I think there are other factors in this issue that Tog is downplaying or ignoring. We can't know, of course, exactly what Apple's motives are in exerting fairly draconian editorial control, but I doubt it's strictly to do with avoiding offending users, be they people who dislike porn or people who dislike political cartoons mocking certain religions.

I suspect that the two biggest issues Apple believes it's dealing with is having a good quality user experience for downloaded applications and avoiding legal complaints, as well as informal ones, about content.

The Quality of the "User Experience"

As Tog points out:

Apple is, in my opinion, doing all of us an important service by looking at each and every app to test for conformance to interface and anti-malware standards.

It's true that this is a valuable service, and one that many Windows users wish was available to them, but there's more to it than this.

At the beginning of the 1980s the home video game market had taken off in a big way. One of the biggest changes from previous programmable consumer devices was the ability of third parties to develop and distribute programs. Systems such as the Atari 2600 had a cartridge slot; any company that wished could write a video game, produce it as a cartridge, and sell it. This resulted in an enormous proliferation of software for the platform.

Of course, with such openness came software of varying quality. There were a lot of bad games out there, and a lot of disappointed consumers. Hardware manufacturers in particular appear to believe that this was one of the major factors in the subsequent video game market crash. From that Wikipedia page:

There were several reasons for the crash, but the main cause was supersaturation of the market with hundreds of mostly low-quality games. Hundreds of games were in development for 1983 release alone, and this overproduction resulted in a saturated market.

The market stayed depressed for years, and Atari, the largest and most successful video game manufacturer of the time, never really recovered from this. Dozens of other companies and software development divisions vanished forever.

The recovery eventually came with the Nintendo Entertainment System (note how it was not referred to as a "video game" console) which contained some special locking hardware within the cartridges and the console to ensure that nobody could produce a game for the console without Nintendo's approval. Since then, no major console system has appeared without such hardware or software locks and a strict developer and game approval process (far stricter than that of Apple's App Store) on the part of the console manufacturer.

This persists to this day even with console systems, such as Sony's PSP Go Playstation Portable system, that use only software downloaded from the vendor's on-line store (rather than discs or cartridges purchased through retail outlets) and compete, at least on the games front, with the iPhone. While Sony has introduced a new PSP Mini category of games with a much cheaper and more streamlined approval process, it's still more strict than Apple's.

Now some of you may be asking yourselves at this point why we have such a different situation with personal computers and their software. Computers came out of a different, and much more "high-maintenance" culture than consumer electronics. Computers started out as expensive, complex systems, often from multiple vendors, requiring expert help to install and maintain them, and despite the improvements and expansion of the audience, haven't changed that much. Even with Mac users it's not considered unusual to spend hours debugging and fixing problems with one's computer. When consumers turn on a video game console or a phone, however, even though it's just as complex as a computer, they expect it just to work.

Publication and Liability

While the previous section relates somewhat to how people see an "Apple" product, that side of things goes deeper. As we saw above, Apple exerts a lot of editorial control over what software is available for the iPhone, and even if they were not concerned about the quality of individual applications themselves, it would be very difficult to avoid at least ensuring that applications don't interfere with other software on the phone. Let's say someone downloaded an application that turned her iPhone into a "brick," that is to say, broke the entire phone. Apple no doubt believes, and probably correctly, that fixing this problem is going to end up in their lap, and "buy a new iPhone", or even "we'll wipe the entire phone, including your personal data, and give you a fresh software install" is not going to cut it.

Also, because iPhones (and some iPads) are radio devices, Apple has legal obligations about what what they can permit the device to transmit. If a radio device can be easily tweaked to do something like jam other people's phone calls, the FCC and other national communications regulatory authorities hold the manufacturer responsible. Most radio devices these days allow for a great deal of software control, and Apple simply can't avoid taking all reasonable steps to ensure that software to make its radio do bad things doesn't get on to the phone.

(This problem can be mitigated through "sandboxing" applications—but see below.)

So we can see that, for legal and some eminently reasonable marketing reasons, Apple needs to exert a fair amount of monitoring of and control over what software goes on their phone. Unfortunately, even if you don't care what people think about your company, with control comes potential legal and even criminal liability.

Consider: if you rent someone a car, the brakes fail while he's driving it, and he runs over someone, it's quite possible (and most people would feel reasonable) that you will be considered to have contributed to that injury or death. Even just knowledge of a crime can make you a criminal because you become an accessory:

A person with such knowledge [that a crime is being, or will be committed] may become an accessory by helping or encouraging the criminal in some way, or simply by failing to report the crime to proper authority.

The extreme example of such a situation for Apple would be if they received an application called "Child Porn" which actually did contain child porn, and distributed it through the App Store. I don't think anybody doubts that they'd be held at least an accessory to this criminal act. Lesser situations, and civil rather than criminal suits, may be less of a problem, but it's still a problem, and in many situations you simply can't know to what conclusion a judge or jury will come.

Apple certainly cannot be blamed for taking this into consideration.


I mentioned "sandboxing" above. This is a technique where you run code in a way that limits what it can do and what effect it can have on the hardware and other applications: essentially, letting it play only in the sandbox, and strictly controlling what effects it can have beyond that.

Even under the best of circumstances, this is a difficult technique—no easier, in fact, than keeping every last grain of sand in the sandbox in a playground. (See the Wikipedia entry for more details.) It's no surprise that Apple can't do this for their native applications, which are what we're discussing here.

However, there is another type of application for which they do do this: web applications used through Safari. And you'll note that these are entirely open: iPhone users can access any web content they wish without restriction. I think that this reinforces my point about liability above.

So What to Do?

I think we can see here that Apple's in a bit of a tricky situation here. As Tog said, "Apple has gotten itself trapped in a box, and it's one that will not neatly unlock." But I think it's clear that they're probably not just being unreasonable jerks here: they have some valid business concerns based on what's happened to others in the past, and they've got some powerful external forces reinforcing those.

My original thought on the matter was to make things wide open, but put apps that didn't come from the App Store in a separate area (a "ghetto," essentially) and have users confirm that they really want to use these. However, the lack of sandboxing is a major problem here: until they overcome that technical hurdle, I don't see how they can reasonably allow this.

So I find myself, though through different reasoning, thinking that Tog's proposed solution is the best one:

The simplest way out of this quandry might be for Apple to expand from having an App Store to having an App Mall, anchored by the App Store and flanked by a small number of independent "boutiques." (No, they would't all be for porn. There could be a liberal boutique, a conservative Christian botique, etc., anything not rigidly corporate-mainstream.)

Apple would require these independents to apply the same stringent interface, safety, and legal standards as Apple, monitoring them from time to time only to the extent that, with parental controls turned on, unsuitable material as defined in their "lease" remains inaccessible. However, the independents, within those few, well-defined constraints, could mount whatever products they so chose. Apple could still, as mall landlord, get a cut of every app sold without getting its hands dirty at the same time.

While I wouldn't put it in terms of "parental controls," it makes perfect sense to have a system whereby users (or their parents, if they're young) would have to agree explicitly to certain terms and conditions for access to each of the non-Apple shops, ensuring that they understand what they're getting into when they enable it, and providing legal cover for Apple if they come across something that they don't like. That would also enable Apple to make its shop even more "family friendly," reducing their chances of getting sued, if that's what they're worried about.

Even better for Apple, this would introduce competition into the application approval process. I can't imagine that this is a profit center for Apple, yet it generates an enormous number of complaints and general ill-will from developers. Being able to have a wider range of applications approved, and having more companies working on the approvals process, would almost certainly help the situation. I would not be surprised to see "premium shops" arise that charged developers significant amounts of money for fast, thorough approval: for some developers it would be worth paying more in order to get their applications approved quickly.

Regardless of whether or not Apple picks this particular solution, I think they're going to need to do something. If they don't, now that Android is on the scene, we're likely to see over the next few years whether such strict control really is preferable to a far more open system.


The Smartest Man in Babylon

Abstruse Goose's comic The Smartest Man in Babylon, which appears to be a comment on copyright, is correct up to the last line:
Now the single "smartest" (and most brazen) thing that any person has ever done...was to convince the world that this abstract sequence of numbers was his property.
Except that he didn't. We did that.

"We," in the sense of people, lawmakers, society, or some combination thereof, in our collective wisdom have decided to award the discoverers of particular sequences of numbers a state-enforced monopoly, for a limited period of time, and with certain limitations, on the copying of these numbers. The ostensible reason for this is to encourage people to spend time and effort discovering (or "creating," if you like) such things.

This general idea makes a reasonable amount of sense, actually. And if you want to know what happens when we don't do this, we have plenty of historical examples to chose from. (Post your favourites in the comments!)

That said, there's no reason things have to be this way. If you feel the current state of copyright law has gone far beyond its original intended purpose and is now doing as much or more harm as good to us (except for certain groups), well, you won't be seeing a whole lot of disagreement from me.

I think one of the key things to remember in this debate, though, is that we, as a society, are in control of such things. And it could well be that all of these young "pirates" who are "stealing" MP3s and the like are simply stepping up and registering their opinion that we need a new balance here. (Or it could be that they are simply naive and don't understand economics very well.) There's no question that we're seeing some sort of backlash against powerful lobbies that are clearly attempting to preserve their own profits gained through rent-seeking.

So let's have a discussion about it. A good place to start appears to be with William F. Patry's book, Moral Panics and the Copyright Wars. (I myself need to finish it before I can comment well on it.)

So think about it. What do we, as a society, want?


Peter Watts: Free?

Well, Peter Watts didn't go to jail. (For those not following this, you can do a web search and find the full story. I don't have the heart to look again at those articles.) This is a victory according to some, and I am happy that he can spend Canada Day home this year. Still, besides having to pay a fine and his lawyer, he's now a convicted felon and may never be able to enter the United States again, which is not an insignificant thing for a science fiction author where conventions are an important part of one's career. He certainly didn't get off scott-free.

But a particular line from Madeline Ashby's post on the sentencing rather summed up my view of the whole thing:
What happened to Peter Watts could happen to any of us.
Now I'm a Liberal, which, when I get into conversations with Americans in bars, appears to mean that I agree with right-wingers almost as much as left-wingers and have only a few of my views in common with libertarians. I certainly respect the risks that many law enforcement officers take every day, even those who are border guards on the Canadian border. (I must admit I do not know how dangerous that job is.)

But there's a difference here: LEOs chose that life, and, if the American ideal is to be believed, they did it to protect the rest of us. So why do I feel nervous crossing the U.S. border these days?

Beyond the day-to-day problems of the U.S., the 9/11 moral panic, NSA wiretapping, Bush rather extending (to use a polite term) executive power, Obama continuing that; all of these scare me. When I go there, it can happen to me, or anybody else, from illegal immigrant to full-fledged citizen. (For some it even happens if you don't go near America; just ask all those who've had a free flight to Gitmo.)

This is why, though I've always appreciated intellectually the ideas espoused by people like Scott Greenfield in his blog Simple Justice, I now feel a lot more viscerally what he's talking about.

It's terrible feeling at first, when you live amongst the free, to think, "I could be next." Perhaps you get used to it.


$200 Million per Arrest

Congressman John J. Duncan Jr. has a fantastic article on the uselessness of air marshals. Best quote:

al Qaeda's most important accomplishment was not to hijack our planes but to hijack our political system.
Edit (2014-09-17): Sorry about turning off comments for this post, but somehow it got popular and spammers started deluging it with comments linking to their own stuff. Weird.


Only a loser would delay the news for her blog.

Not Always Right is certainly full of social comment, but it invariably makes fun of people who don't understand what's going on. That of course means that, if the Internet is involved, it makes fun of people who don't quite understand the Internet.

That makes me wonder why this exchange is there, since it so clearly does involve understanding on the part of the customer this time, rather than the staff member:
(It is a few minutes after closing time and we lock the door. Moments later I hear the sound of breaking glass. I rush over and see a woman on the ground surrounded by glass. She had tried to walk through the door and broke it.)
Me: “Miss, are you alright?! Are you hurt anywhere?”
Customer: “No! No…I think I’m fine.”
Me: “Miss, let me show you somewhere to sit while we wait for security and the ambulance.”
Customer: “No! I don’t have time d*** it! I have to go write this in my blog!” *runs off*
Was it because she didn't just pull out her Blackberry and Twitter it?