If you can't make the little mistakes because of language design, you'll be hopeless at fixing bugs in your individual architectural design.
That's ridiculous.
The issue is that programming is done on many levels. The high level is the program's architecture, and then there are lower levels than that (the number depending on the level of abstraction, how many libraries are used, etc.) until you get down to the base level of the machine code itself. Why does this matter? Simple: the lower levels can only ever add bugs; they can't compensate for problems at the higher levels. If your program architecture is a buggy POS (and high-level deadlocks would be a good indicator that this is indeed the case; large-scale resource management is always a key aspect of architectures) then whatever you implement that architecture with is still going to result in a buggy POS.
Of course, what's really hard is that it is difficult to work at very high levels of abstraction. Easy to make a subtle mistake, hard to detect the error, hard to correct it consistently through the other levels. Thus bugs are very easily introduced...
Maybe its a good language once you learned the hard way why its a good language by working with others.
The hair shirt approach to programming? Or perhaps the "get off my lawn" school?
Yes, loads. But with the ignorance you're showing, you must work for some part of the rest of Microsoft; after all, they never seem to listen at all to what MSR's up to and instead focus on turning out the same not-very-innovative crap over and over.
They could loose a little bit. And make it back up in a couple of book/item sales. Just depends what they want to consider ROI.
What's more, Amazon have a long history of using ultra-thin margins and making it back on volume. They own several markets completely through that strategy, yet without locking competitors out so that they don't have excessive regulatory trouble. Very very clever. Very dangerous to anyone competing with them. They also have a lot of eCommerce patents, which would make it ultra-hazardous for the likes of Apple to go after closing down distribution of the new Kindles (since the retaliation would likely involve effectively closing down iTunes, which is currently critical to Apple's profitability).
For example the ps3 sold until recently at a loss (sony even states this in their quarterly earnings reports).
But if they're taking the profits on the games through a different route (i.e., doing accounting tricks) then the loss on the device is not important. What's more, they now have a period when they are going to own the console market; the development costs on the PS3 are largely recouped, the competitors are needing to do major product refreshes to catch up, the manufacturing efficiency is high (and costs are low), and the platform is picking up nice licensing fees from all the third-party developers producing software for it. It won't last forever, but it's going to be pretty sweet for a while even allowing for the widespread depression. Which part of the business officially clocks up the profit... that's an accounting matter (and will probably be tuned to minimize tax liabilities) but it's the whole that's winning.
Nintendo and Apple on the other hand use their name to sell at a premium (and not at a loss).
Apple are also good at the business of getting money from a significant fraction of the product use cycle, ensuring that they get their cut from a lot of the things that are done with their devices. However, they have the problem that they've annoyed content providers a lot and have a lot of competition coming at them very fast and hard from many directions. They might also be about to attract significant regulatory attention; being wildly profitable by providing what people want is fine, but doing it by over-sharp business practices is not...
Nintendo are just in the wrong part of the product cycle right now (the Wii had a naturally shorter lifespan than the PS3 because so much of it was less technically advanced) and may have had a few problems with the situation in Japan this year. If they survive (and there's no particular reason to think they won't) expect them to be back with future products. It's just unfortunate for them that the thing they were banking on a bit (3D handhelds) hasn't worked out so well as they'd hoped; human visual systems are tricky. Though to be fair, the big threat to Nintendo is probably Apple (and the various manufacturers grouped around Android) as those are devices that are likely to threaten Nintendo's handheld stranglehold utterly. It'll be interesting to watch what happens from the sidelines.
Why doesn't anyone else take their laptops and add an aluminum case and 50% markup?
Because it's not one thing, it's lots of small things. For me, the big benefits are a full Unix environment, really good suspend support (really important for a notebook) and convenient backup software. Oh, and I haven't broken the case yet, unlike with the PCs I had before of the same age.:-) YMMV.
Why have one-size-fits-all healthcare? That'd be like having one OS, or one brand of clothing, or one chain of restaurants, or one make of car. We don't ask government to manage our food purchases, or where we choose to live, or what jobs we work at. Why this fascination with government-run healthcare?
Because there's ample evidence from elsewhere in the world that having a single mandatory system reduces costs in healthcare, which are massively out of control in the USA. It also helps in that with a single system bearing all the costs, those running that system are strongly motivated to find cheap ways to alleviate situations (which in healthcare means that there's much more of a focus on prevention rather than treatment; it's better to not get sick at all than to get better.) That's not to say that it's necessary to have no private sector involvement at all, but with a very large government buyer about that everyone is eligible to use, the private sector is motivated to provide better service (less waiting, better care) to justify their extra charges and to show people what they are getting for the money. (This is the system that the UK has; it's definitely not 100% government run.)
So no, the Constitution does not grant the federal government any say in the welfare of the citizens of the States.
It depends on the extent to which you believe the welfare of the states is determined by the welfare of the citizens (and other inhabitants) of those states.
Mojang shouldn't be able to trademark Scrolls, however they should definitely be able to use Scrolls without risk of exactly this lawsuit.
They should be able to, but only in a particular artistic font or in conjunction with something else that makes it very clear that it is talking about Mojang's game. The whole point of trademarks is that they are non-functional decoration to allow customers to know unambiguously whose product they are purchasing (and using, but that's actually secondary in legal terms). Making trademarks be "strictly enforced but very specific" is how it's supposed to be. Trademarking an undecorated dictionary word, even if just within the scope of a particular industry (e.g., computer software), is an attempted abuse.
You can't form a contract by simply sending someone a unilateral set of terms and relying on their silence to confirm their assent to it.
You can in some jurisdictions, but it does depend on the exact set of terms proposed. In particular, the person receiving the document can't be forced to take any action by such one-way contracts because they have not agreed to anything (so the contract terms would have to say "If you - the receiver - do A, then I will do B." so that the acceptance is still clearly signaled.) Remember, contracts are fundamentally agreements and you can't have an agreement without all parties actually agreeing.
Not that I'm a lawyer, but I did work on a project that looked at the legality of SLAs so I have actually looked into this.
If you are really serious about reform, and you applaud USPTO's rejection of this term, you must also support revoking the Microsoft trademark on "Windows".
The trademark term is "Microsoft Windows", and that's obviously permitted. Had Apple tried to get "Apple Multi-Touch" then they'd have had no problem (but it wouldn't have had the squatting-on-a-term effect that they wanted).
Why the hell would you want a statically typed language to replace Javascript? Multiple people have said they want this but I can't imagine why. Of all the problems with JS that's the last thing I'd imagine people would have a problem with.
Because they want to personally have to write lots of type thunks for inter-converting between interfaces to existing components. Once you add in all that stuff, you slow your statically typed program back down to around the sorts of speed that you get in scripting languages. This does tend to be the stuff that isn't picked up by published benchmarks though, since there programmers of languages with static types go to the effort of making everything optimally typed throughout; that's great, but too often doesn't bear out in the real world (where expedient hacks are far more prevalent).
One curiosity: many people who write statically-typed programs insert their own scripting language library to handle things like configuration. Most such little languages are horrible, slow and vastly underpowered...
But you do raise another interesting concern, how do you get the containers back to the USA? If a full container ship travels to the USA and takes back to China 4 or 5 times as many empties as it took over eventually the empties will start collecting at terminals of countries with huge export and little import trade.
This happens. Then someone buys up the (by now, going cheap) empty containers for other uses. For example, if the price of the metal the containers are made from is high enough, it becomes worthwhile to recycle as scrap. On the other hand, the filling up of storage space for empty containers also encourages port (or storage) owners to charge higher amounts for their part in all this; after all, the port owners want trade to continue and storage for empty boxes is just an irritating land-hungry sideline.
I wonder if there is a way to actually provide physical keys to computer systems.
Yes. That's smartcard-based login systems, and they've been around for decades. The main downside is that they're relatively expensive due to the need to have all that extra hardware and someone on-site to issue new cards — that can't be outsourced to another location, well not outside the city where this is happening, because cards will get broken from time to time and need replacing by someone who's trained to check that the card is going to the right person — so they tend to only be used in situations that can justify it (e.g., government offices handling large amounts of personal data).
My inbox will contest that. I get spam from some pretty reputable UK companies, despite the fact that it is illegal.
There's two different types of spam. One is commercial email that is sent legitimately but which you don't want, and the other is the stuff that is being sent by the true mass spammers which uses false identities. The former, you can block with your email client just fine because it's not pretending to be anything or by anyone other than the truth. The latter, that merits the use of real anti-spam services (block lists, etc.) While yes, you don't really want either, it's the latter which is a deep problem (the former is just advertising, and has more in common with irritating ads on websites than criminality).
So if we get rid of the TSA and we hire Joe the plumber to do airport security or just leave it up to the airlines them maybe if we get another terrorist attack the people will shut up and not bitch about lax security.
You keep a rump of the TSA, but instead of running security directly they should be in charge of keeping the private security people from getting too lax.
Self-signed certs are effectively without identity assertions, since the creator can say anything they want in them. The only way you can trust them is if you either learn the true content of the certificate by some other method (copying the file via a known-good mechanism is the simplest way) or get someone else to assert that the certificate is indeed valid (i.e., they're a bodged-together CA). Without those, you simply don't have any way to tell if the server certificate is valid or not, as one certificate looks just like another to a computer; there's no Evil Bit (though there might be an OID for Evil[*]) and I've never met a (non-security-wonk) user who could be bothered to validate a key signature by hand.
To work out the identity of the server — in fact, not just its hostname but the identity of the organization which owns and operates it — we need some kind of mechanism for making trustable assertions. Either that's peer-to-peer or its centralized. P2P (i.e., web of trust) has the problem that it uses transitive trust relationships, and if someone says something wrong (as will happen as things scale up, whether through incompetence or malice) then it's a total mess to clear things up. A centralized system has many problems too, but at least has the ability to enforce rules. That's what's happened to Diginotar: they didn't obey the rules, and so got punished. We just need to remember that this is what should happen to a CA who isn't diligent, and let it stand as an object lesson.
And a CA absolutely should know who they've issued (currently-valid) certificates to. That they didn't... sheesh!
[*] Who am I kidding? All OIDs are evil, whatever they're saying.
I remember hearing about a new capacitor technology that used nanoparticles to allow far more power storage and almost instant recharge rates. Better/worse?
Different purpose. Capacitors (energy stored in electrical potential) are typically used in combination with batteries (energy stored in chemicals); the caps can deal with rapid changes in demand much better, but batteries tend to have much higher overall energy densities and lower leakage currents.
Isn't that true for every single company on Earth?
No. The successful ones tend to be run by businessmen (and businesswomen) since either they get taken over by people who specialize in the running of businesses, or the founder/owner becomes more adept at running the business and becomes a business(wo)man. Businesses that aren't run as businesses at least somewhat tend to run out of money and fold, since controlling the assets — and especially the cash supply — is a core requirement of being a solvent business. There are a lot of "businesses" in that category, though most are small (and likely to vanish virtually without trace).
Parameterized queries require pre-registration with the database engine, and have a host of other gotchas. EG: No dynamically created SQL queries.
No. That's just your sucky database client library. OTOH, if you've got a parameterized query then you absolutely should be able to cache it so that you can reuse it, avoiding a recompile of the query where not necessary. Some client libs can handle that for you automatically, others are more intrusive. (Some languages even have wrapping layers available that make nasty low level libs much more palatable.)
That the language is SQL or something else, that's wholly independent really.
Remembering to escape every variable also has it's own gotchas, including the inevitable security hole when you mistook public data as trusted source data.
Now that is very true. If you're thinking about doing that, STOP! and consider whether it is at all possible for you to do it a different way. If you're thinking about allowing arbitrary users to type in full SQL queries, take great care to document that you're doing it (and that they need to be executed in a very limited database account) so as to avoid amazing amounts of trouble. (Also consider psychiatric assistance.)
The best question on the visa waiver form is whether you are seeking entry to "engage in immoral activities." Who can go through a single day without doing something that someone considers immoral?
I believe there's a more closely given definition in the actual legislation, but it's probably rather long for putting on a normal-length form (I found it once online, but can't remember where; IIRC, it mentioned many specific activities such as drug running and prostitution). If you have to know, you're probably running too close to the edge of legality...
However, Apple registered their design in the EU, so they might theoretically sue in other countries as well, and maybe even be able to use the court verdict in Germany as a precedent.
Except that a majority of EU countries don't formally admit precedence as part of court cases, and those that do aren't particularly inclined to be bound by the decision of a German lower court. Moreover, there are some subtle differences between EU countries in how the laws in this area are specified and tried.
The equivalent in the US would be a decision by a lower state court (though in a very big state; Germany's got more population than California, Texas and New York state combined).
Slashdot Mirror
If you can't make the little mistakes because of language design, you'll be hopeless at fixing bugs in your individual architectural design.
That's ridiculous.
The issue is that programming is done on many levels. The high level is the program's architecture, and then there are lower levels than that (the number depending on the level of abstraction, how many libraries are used, etc.) until you get down to the base level of the machine code itself. Why does this matter? Simple: the lower levels can only ever add bugs; they can't compensate for problems at the higher levels. If your program architecture is a buggy POS (and high-level deadlocks would be a good indicator that this is indeed the case; large-scale resource management is always a key aspect of architectures) then whatever you implement that architecture with is still going to result in a buggy POS.
Of course, what's really hard is that it is difficult to work at very high levels of abstraction. Easy to make a subtle mistake, hard to detect the error, hard to correct it consistently through the other levels. Thus bugs are very easily introduced...
Maybe its a good language once you learned the hard way why its a good language by working with others.
The hair shirt approach to programming? Or perhaps the "get off my lawn" school?
MS does research? For real?
Yes, loads. But with the ignorance you're showing, you must work for some part of the rest of Microsoft; after all, they never seem to listen at all to what MSR's up to and instead focus on turning out the same not-very-innovative crap over and over.
How much do you trust your DNS registrar?
What's more, if your DNS registrar is crooked (or broken into), you're stuffed because you can't go to someone else nearly as easily as with a CA.
They could loose a little bit. And make it back up in a couple of book/item sales. Just depends what they want to consider ROI.
What's more, Amazon have a long history of using ultra-thin margins and making it back on volume. They own several markets completely through that strategy, yet without locking competitors out so that they don't have excessive regulatory trouble. Very very clever. Very dangerous to anyone competing with them. They also have a lot of eCommerce patents, which would make it ultra-hazardous for the likes of Apple to go after closing down distribution of the new Kindles (since the retaliation would likely involve effectively closing down iTunes, which is currently critical to Apple's profitability).
For example the ps3 sold until recently at a loss (sony even states this in their quarterly earnings reports).
But if they're taking the profits on the games through a different route (i.e., doing accounting tricks) then the loss on the device is not important. What's more, they now have a period when they are going to own the console market; the development costs on the PS3 are largely recouped, the competitors are needing to do major product refreshes to catch up, the manufacturing efficiency is high (and costs are low), and the platform is picking up nice licensing fees from all the third-party developers producing software for it. It won't last forever, but it's going to be pretty sweet for a while even allowing for the widespread depression. Which part of the business officially clocks up the profit... that's an accounting matter (and will probably be tuned to minimize tax liabilities) but it's the whole that's winning.
Nintendo and Apple on the other hand use their name to sell at a premium (and not at a loss).
Apple are also good at the business of getting money from a significant fraction of the product use cycle, ensuring that they get their cut from a lot of the things that are done with their devices. However, they have the problem that they've annoyed content providers a lot and have a lot of competition coming at them very fast and hard from many directions. They might also be about to attract significant regulatory attention; being wildly profitable by providing what people want is fine, but doing it by over-sharp business practices is not...
Nintendo are just in the wrong part of the product cycle right now (the Wii had a naturally shorter lifespan than the PS3 because so much of it was less technically advanced) and may have had a few problems with the situation in Japan this year. If they survive (and there's no particular reason to think they won't) expect them to be back with future products. It's just unfortunate for them that the thing they were banking on a bit (3D handhelds) hasn't worked out so well as they'd hoped; human visual systems are tricky. Though to be fair, the big threat to Nintendo is probably Apple (and the various manufacturers grouped around Android) as those are devices that are likely to threaten Nintendo's handheld stranglehold utterly. It'll be interesting to watch what happens from the sidelines.
Why doesn't anyone else take their laptops and add an aluminum case and 50% markup?
Because it's not one thing, it's lots of small things. For me, the big benefits are a full Unix environment, really good suspend support (really important for a notebook) and convenient backup software. Oh, and I haven't broken the case yet, unlike with the PCs I had before of the same age. :-) YMMV.
But it doesn't necessarily benefit both parties.
You're forgetting the 3rd party, the carrier. In their eyes, the other two parties do not count except as a revenue source.
As long as they get paid and in full, they don't care who pays. (Getting paid twice is even nicer, but not really a sustainable business practice.)
Why have one-size-fits-all healthcare? That'd be like having one OS, or one brand of clothing, or one chain of restaurants, or one make of car. We don't ask government to manage our food purchases, or where we choose to live, or what jobs we work at. Why this fascination with government-run healthcare?
Because there's ample evidence from elsewhere in the world that having a single mandatory system reduces costs in healthcare, which are massively out of control in the USA. It also helps in that with a single system bearing all the costs, those running that system are strongly motivated to find cheap ways to alleviate situations (which in healthcare means that there's much more of a focus on prevention rather than treatment; it's better to not get sick at all than to get better.) That's not to say that it's necessary to have no private sector involvement at all, but with a very large government buyer about that everyone is eligible to use, the private sector is motivated to provide better service (less waiting, better care) to justify their extra charges and to show people what they are getting for the money. (This is the system that the UK has; it's definitely not 100% government run.)
So no, the Constitution does not grant the federal government any say in the welfare of the citizens of the States.
It depends on the extent to which you believe the welfare of the states is determined by the welfare of the citizens (and other inhabitants) of those states.
X windows as mainly developed at MIT
but he said successful as opposed to "kludge used in the absence of anything else".
Then it's a very successful kludge. I wish all mine were even 1% as successful!
Mojang shouldn't be able to trademark Scrolls, however they should definitely be able to use Scrolls without risk of exactly this lawsuit.
They should be able to, but only in a particular artistic font or in conjunction with something else that makes it very clear that it is talking about Mojang's game. The whole point of trademarks is that they are non-functional decoration to allow customers to know unambiguously whose product they are purchasing (and using, but that's actually secondary in legal terms). Making trademarks be "strictly enforced but very specific" is how it's supposed to be. Trademarking an undecorated dictionary word, even if just within the scope of a particular industry (e.g., computer software), is an attempted abuse.
You can't form a contract by simply sending someone a unilateral set of terms and relying on their silence to confirm their assent to it.
You can in some jurisdictions, but it does depend on the exact set of terms proposed. In particular, the person receiving the document can't be forced to take any action by such one-way contracts because they have not agreed to anything (so the contract terms would have to say "If you - the receiver - do A, then I will do B." so that the acceptance is still clearly signaled.) Remember, contracts are fundamentally agreements and you can't have an agreement without all parties actually agreeing.
Not that I'm a lawyer, but I did work on a project that looked at the legality of SLAs so I have actually looked into this.
If you are really serious about reform, and you applaud USPTO's rejection of this term, you must also support revoking the Microsoft trademark on "Windows".
The trademark term is "Microsoft Windows", and that's obviously permitted. Had Apple tried to get "Apple Multi-Touch" then they'd have had no problem (but it wouldn't have had the squatting-on-a-term effect that they wanted).
Why the hell would you want a statically typed language to replace Javascript? Multiple people have said they want this but I can't imagine why. Of all the problems with JS that's the last thing I'd imagine people would have a problem with.
Because they want to personally have to write lots of type thunks for inter-converting between interfaces to existing components. Once you add in all that stuff, you slow your statically typed program back down to around the sorts of speed that you get in scripting languages. This does tend to be the stuff that isn't picked up by published benchmarks though, since there programmers of languages with static types go to the effort of making everything optimally typed throughout; that's great, but too often doesn't bear out in the real world (where expedient hacks are far more prevalent).
One curiosity: many people who write statically-typed programs insert their own scripting language library to handle things like configuration. Most such little languages are horrible, slow and vastly underpowered...
But you do raise another interesting concern, how do you get the containers back to the USA? If a full container ship travels to the USA and takes back to China 4 or 5 times as many empties as it took over eventually the empties will start collecting at terminals of countries with huge export and little import trade.
This happens. Then someone buys up the (by now, going cheap) empty containers for other uses. For example, if the price of the metal the containers are made from is high enough, it becomes worthwhile to recycle as scrap. On the other hand, the filling up of storage space for empty containers also encourages port (or storage) owners to charge higher amounts for their part in all this; after all, the port owners want trade to continue and storage for empty boxes is just an irritating land-hungry sideline.
I wonder if there is a way to actually provide physical keys to computer systems.
Yes. That's smartcard-based login systems, and they've been around for decades. The main downside is that they're relatively expensive due to the need to have all that extra hardware and someone on-site to issue new cards — that can't be outsourced to another location, well not outside the city where this is happening, because cards will get broken from time to time and need replacing by someone who's trained to check that the card is going to the right person — so they tend to only be used in situations that can justify it (e.g., government offices handling large amounts of personal data).
My inbox will contest that. I get spam from some pretty reputable UK companies, despite the fact that it is illegal.
There's two different types of spam. One is commercial email that is sent legitimately but which you don't want, and the other is the stuff that is being sent by the true mass spammers which uses false identities. The former, you can block with your email client just fine because it's not pretending to be anything or by anyone other than the truth. The latter, that merits the use of real anti-spam services (block lists, etc.) While yes, you don't really want either, it's the latter which is a deep problem (the former is just advertising, and has more in common with irritating ads on websites than criminality).
So if we get rid of the TSA and we hire Joe the plumber to do airport security or just leave it up to the airlines them maybe if we get another terrorist attack the people will shut up and not bitch about lax security.
You keep a rump of the TSA, but instead of running security directly they should be in charge of keeping the private security people from getting too lax.
Self signed certs work fine
Self-signed certs are effectively without identity assertions, since the creator can say anything they want in them. The only way you can trust them is if you either learn the true content of the certificate by some other method (copying the file via a known-good mechanism is the simplest way) or get someone else to assert that the certificate is indeed valid (i.e., they're a bodged-together CA). Without those, you simply don't have any way to tell if the server certificate is valid or not, as one certificate looks just like another to a computer; there's no Evil Bit (though there might be an OID for Evil[*]) and I've never met a (non-security-wonk) user who could be bothered to validate a key signature by hand.
To work out the identity of the server — in fact, not just its hostname but the identity of the organization which owns and operates it — we need some kind of mechanism for making trustable assertions. Either that's peer-to-peer or its centralized. P2P (i.e., web of trust) has the problem that it uses transitive trust relationships, and if someone says something wrong (as will happen as things scale up, whether through incompetence or malice) then it's a total mess to clear things up. A centralized system has many problems too, but at least has the ability to enforce rules. That's what's happened to Diginotar: they didn't obey the rules, and so got punished. We just need to remember that this is what should happen to a CA who isn't diligent, and let it stand as an object lesson.
And a CA absolutely should know who they've issued (currently-valid) certificates to. That they didn't... sheesh!
[*] Who am I kidding? All OIDs are evil, whatever they're saying.
Especially considering software patents are not legal in the UK
Not true! They're legal, but significantly harder to get than in the US.
I remember hearing about a new capacitor technology that used nanoparticles to allow far more power storage and almost instant recharge rates. Better/worse?
Different purpose. Capacitors (energy stored in electrical potential) are typically used in combination with batteries (energy stored in chemicals); the caps can deal with rapid changes in demand much better, but batteries tend to have much higher overall energy densities and lower leakage currents.
EA is run and owned by businessmen.
Isn't that true for every single company on Earth?
No. The successful ones tend to be run by businessmen (and businesswomen) since either they get taken over by people who specialize in the running of businesses, or the founder/owner becomes more adept at running the business and becomes a business(wo)man. Businesses that aren't run as businesses at least somewhat tend to run out of money and fold, since controlling the assets — and especially the cash supply — is a core requirement of being a solvent business. There are a lot of "businesses" in that category, though most are small (and likely to vanish virtually without trace).
Parameterized queries require pre-registration with the database engine, and have a host of other gotchas. EG: No dynamically created SQL queries.
No. That's just your sucky database client library. OTOH, if you've got a parameterized query then you absolutely should be able to cache it so that you can reuse it, avoiding a recompile of the query where not necessary. Some client libs can handle that for you automatically, others are more intrusive. (Some languages even have wrapping layers available that make nasty low level libs much more palatable.)
That the language is SQL or something else, that's wholly independent really.
Remembering to escape every variable also has it's own gotchas, including the inevitable security hole when you mistook public data as trusted source data.
Now that is very true. If you're thinking about doing that, STOP! and consider whether it is at all possible for you to do it a different way. If you're thinking about allowing arbitrary users to type in full SQL queries, take great care to document that you're doing it (and that they need to be executed in a very limited database account) so as to avoid amazing amounts of trouble. (Also consider psychiatric assistance.)
The best question on the visa waiver form is whether you are seeking entry to "engage in immoral activities." Who can go through a single day without doing something that someone considers immoral?
I believe there's a more closely given definition in the actual legislation, but it's probably rather long for putting on a normal-length form (I found it once online, but can't remember where; IIRC, it mentioned many specific activities such as drug running and prostitution). If you have to know, you're probably running too close to the edge of legality...
However, Apple registered their design in the EU, so they might theoretically sue in other countries as well, and maybe even be able to use the court verdict in Germany as a precedent.
Except that a majority of EU countries don't formally admit precedence as part of court cases, and those that do aren't particularly inclined to be bound by the decision of a German lower court. Moreover, there are some subtle differences between EU countries in how the laws in this area are specified and tried.
The equivalent in the US would be a decision by a lower state court (though in a very big state; Germany's got more population than California, Texas and New York state combined).
I think you get the picture - the picture of a magnificent IBM Model M keyboard, that is.
It's the only keyboard that's certified for dealing with the Zombie Apocalypse, and is ideal for use by lumberjacks too.