Well, the dicing up will limit the power of individual banks.
However, you are correct that the problem was that EVERYBODY was acting stupid. The problem is that when all your competitors are making risky loans and selling them to investors as safe loans you have two choices:
1. Do the same thing. 2. Watch all the investors pull out all of their money and give it to those who do #1.
After all, if you're an investor, which fund would you invest in? The AAA bond fund yielding 15%, or the AAA bond fund yielding 5%?
A big problem is that individual investors are far too insulated from the investment decisions. Most people have a choice of maybe 20 mutual funds in their 401k - usually from one or two investment houses. At best they get a quarterly statement and some audits. Nobody really works for the individual investor in such a setup - the funds make their pitches to their employers. People are told to just set aside 10% of their income and watch it grow, and that investments are too complicated for them to deal with. Nobody is accountable when things go wrong.
The free market only works when consumers can make educated choices. That is why there was a ton of regulation after the 1929 crash.
Yup, that would be the other thing that I would legislate (besides separating the billing items for phone subsidies and plan fees) - the advertised cost of the plan is what consumers have to pay, and not a penny more. For that matter, unless a consumer agrees there can be no incremental costs beyond the advertised plan cost. Carriers may not offer a discount on the plan if you agree to pay extra for texts/etc, either. The whole surprise-$500-SMS-bill thing has to go away.
Add-ons should be add-ons, and if a consumer says that they're not interested in the add-ons then the carrier cannot charge for them, period. By all means they can block SMS or MMS or whatever if the customer hasn't paid for them, but they can't charge for them if they get used if the consumer didn't request this.
In the US they're also like an installment plan, but you pay the installments whether or not you get a free phone out of the deal. It is a BIG ripoff, and it is the reason why I have a collection of perfectly fine old phones lying around - if I pay the same to keep the old phone or get a new one, why wouldn't I want the new one?
Agreed. There should also not be any early-termination fee for any aspect of the service other than the phone cost, and you shouldn't have that at all if you bring your own phone. Also, you should be able to pay out the remainder of your phone rental term without having phone service. So, if my phone fee is $10/month and my plan is $100 per month and I want to switch two months early at most it should cost me $20 (paid over the next two months).
And all phones should be unlocked - at most the company that you're renting them from should have a lein against the phone. I'm fine with them demanding the return of the phone if you fail to make payments for it, but if I cancel my deal with AT&T I should be able to use my existing phone with T-mobile or whatever as long as I pay for phone-only cost for the remainder of my AT&T plan.
Once this happens you could mostly deregulate phones, as they are a commodity. The plans are the part that are a natural monopoly.
It is only a free market if anybody is free to enter the market. That is certainly not the case with a phone company. How much capital do you think it would take to actually create a credible cell-phone provider?
Phone PROVIDERS, on the other hand are NOT. In the US there are really only four with any significant coverage, and only two with fairly wide coverage. That is not really a "free market."
And only T-mobile at this point has plans that don't involve phone subsidies, which makes the commodity status of phones an issue as well. What good does it do me to buy my own phone when the carrier is going to just charge me as if they had given me a subsidized one?
They should be required to separate the phone subsidy in the plan so that you can choose to get the plan with or without the subsidy.
If you cancel service early you could opt to pay the termination fee, or pay ONLY for the subsidy portion of the plan for the remainder of your contract (not the FULL plan). If you don't have this clause then the providers will play games with the subsidy value so that people still pay more.
Will it actually be reasonably supported in such a configuration? Google doesn't even support an open-source version of android that can actually make phone calls. They have lots of source code, but no releases or anything with QA behind it. Integrating proprietary drivers needed to actually make it work is a fairly manual process, and there are no guides/etc.
It seems like they're only willing to make it open-source to the extent that it doesn't cost them anything and it doesn't create viable competition.
From what I've seen Google open source projects don't really operate like normal open source projects. It is a proprietary cathedral model that publishes most of their source code, without much regard to whether the open source product is in itself something anybody would actually want to use. At least with mysql or whatever you can download the source for an actual release, and do a make install and get more-or-less exactly what you'd get if you just used their product.
From what I've seen on the web wave is open source in the same sense that android is open source. There are a bunch of files on a CVS server, with some build instructions that creates a program that does something, but it isn't something they actually intend anybody to actually use. In the case of android their open source code results in an OS that doesn't have access to the cell network or wifi, and which can't even blink the LED on the front from what I understand. Their build instructions are a real mess too as it requires a ton of manual effort to get the proprietary blobs added to it so that it works. It isn't like you can just put some proprietary bits in a directory, hit make, and end up with an image.
Plus, their only implementation runs on Java, and I don't really have half a gig of RAM to waste having that in memory all the time. Most servers that don't get much traffic can run in a lot less!
It isn't a bad concept, but I still don't see it taking off. Maybe if I ran a forum and the open-source version of wave were robust enough I might replace phpbb or whatever with it. However, most open-source forum systems are actually designed to be used by people, and they're not just intended for developers to contribute to one particular implementation of it.
Of course, who is to say what she actually said, but I agree - the word "love" is REALLY strong. Unless you're a teenager you shouldn't be using it to describe every little infatuation.
Relationships at work can be very messy - particularly when they involve somebody you work closely with. You need to take things carefully one step at a time.
Yup. I know somebody who was involved in some software that dealt with clinical trial data and they were ultra-paranoid about this sort of stuff (to the point where access to many fields in the database was need-to-know). A full date of birth was considered as sensitive as a name/address or ID code (and they didn't even track SSNs). Even the age or year-of-birth was considered sensitive if the age was above a threshold (how many 97-year-olds are there?).
They clearly didn't talk to a lawyer versed in this stuff before releasing this data.
You don't want to light up a radar or anything like that because it can be detected FAR beyond its own detection threshold unless you're very sure you need a positive firing solution and that's the way to get it.
Yup. Plus at longer ranges you get the radar return at about the same time as you get the incoming volley from your target who is firing down the bearing of your radar. By the time you fire back the entire area around you is being blasted at.
Now, if you can sneak in to what you have reason to believe is close range (negligible light travel time) then radar can be used pretty safely, since you'll be more prepared to make use of the data than they will.
I agree that drones are the way to go. In fact, missiles might actually become rockets with sensors and a laser on them. You fire them at the approximate area the enemy is in, and then the missile detects the target and fires at it when it is within range. Being a small object it could get in close, and it need not have energy for sustained maneuver or lots of shots. An x-ray laser on a rocket could be a really devastating weapon. However, the missile will need to be able to boost without giving away its trajectory, or it will be an easy target when it is ballistic.
I think the only reason the tank/dps/healer thing works is because of dumb AI. So, you're taking out a whole army of goblins or whatever in some cave. You come up to groups of 2-3 bad guys at a time, in constrained corridors so that your tank can actually block them from getting past.
There are a couple of obvious counters to this. The first group of 2-3 puts their biggest tanks in your way to slow you down, while the last guy does a Paul Revere and every goblin in the lair is headed your way. While they hold you up a bunch of them leave via their back door and circle around to come at you from behind. The goblins of course put their own tanks in front, and defend themselves from prepared positions. If you come across a long corridor you can bet that there will be archers chopping you to pieces from the far end.
However, the game AI just sends people at you in mobs of 2-3 to be chewed up one at a time.
In open terrain any larger group of enemies would certainly flank your tanks and take out your rear forces. They would never bunch up to allow an area-effect spell to do anything.
I do agree with some of the posters that a fantasy game shouldn't be completely realistic. In any REAL war with reasonably matched forces almost nobody makes it through multiple battles alive. Most people kill at most a few enemy soldiers and then are killed themselves. That doesn't make for good gameplay.
However, I am not big on MMORPGs because they are way too formulaic - the grind, roles, etc. I can't just go and do what I want to do unless I only want to experience a small part of the game or tick off my party/guildmates/whatever.
Singularity Sky (Charles Stross) has a similar depiction of space battle.
The bigger your weapons the longer your ranges get - since you can cover a bigger volume of space and the errors in targeting solutions matter less. At light-minute distances it would create an entirely new set of tactics.
Agreed - a big factor in space battles is going to be getting a firing solution on the enemy, and then blasting as much space as possible in the area where he is going to be. That might be a bullet every 10m, or laser fire directed across a cone, or whatever.
To detect the guy you need light to travel from him to you. To kill him you need your weapons to travel from you to him. That is at least twice the light-travel time for your weapon to get there, so as soon as the other guy's acceleration limits him to a volume of space you can target in twice the light travel time then he's dead. Stealth could be critical, as it could let a smaller vessel get closer to a big one and negate its ability to direct energy into a larger volume of space (plus the smaller vessel can probably accelerate more, increasing the volume of space that must be targeted).
I wouldn't be surprised if 99% of the actual shooting is fully automated. The only thing humans will be doing is telling the computer where to go and what to shoot at. The ship would be under constant maneuver with the computer doing its best to get to the general area it is asked to go to, while killing anything it can on the way.
Coordination of multiple ships could be really tricky - your communications also suffer travel delay due to the speed of light. Unless you direct your communications carefully they will also broadcast your position.
I do agree with your point that electronic warfare will be important. If you can trick out the enemy's radar then you greatly reduce his effective firing range, as the volume of space to be targeted goes way up.
In space combat that divergence angle will be essential - or you won't hit your target unless you're at very close range.
Suppose you open fire at a range of 10 light seconds? Do you think that a microscopic beam will have any chance of hitting the target at all?
I think that is the one thing that all this discussion is missing - the concept of VERY long range warfare in space.
When those guys from earth launch to head out to mars to conqueror it, they're going to be under observation and potential attack as soon as the light from the booster rocket travels to mars. Chances are the mars base will be under direct missile attack from earth as well. Now, at those long ranges no weapons will be effective, but I think that the ranges where you need to start worrying about incoming fire are going to be a LOT longer than people appreciate.
Ships will be destroyed when they reach a range where they are fired upon by a ship that can direct a lethal amount of energy into a volume of space large enough to contain most of the possible trajectories the target can take based upon the round-trip time of the sensors plus the weapons fire. Ships will constantly generate maximum thrust in semi-random directions to maximize the volume of space their opponent has to cover. They also have to guess when they are detected, since rocket thrust will undoubtedly broadcast their position at a longer range than a radar return will. Ships might even avoid using radar, as a radar gives your target a bearing to you that is twice as recent as the bearing you have on (since he sees you when the ping gets to him, and you see him when the ping returns).
Just think of what combat looks like when the travel time of light is noticeably wrong - that is what interplanetary space combat will look like. Now, in the near future there could certainly be some closer-range stuff before all the technology matures, but I think that ground-based platforms will have the advantage for a while since they can expend a LOT more energy.
On the forums for "search engine optimization", one discovers that ad clicks from Google search results tend to result in sales, while ad clicks from Google ads on non-Google sites (what Google euphemistically calls the "Google Content Network") don't. 50% of ad clicks come from 10% of the user base, and that 10% doesn't buy anything.
Actually, all that you know is that the people who view your ads on non-Google sites don't buy the items at that moment by clicking on the link and immediately buying it.
You have no idea how many people actually bought your product after VIEWING (not clicking) the ad.
By your argument a superbowl ad has zero impact, because not a single person clicked on their TV screen during the game and bought a Gilette razor or whatever.
I never really understood the obsession with click-through rates in the online advertising world. Sure, it is nice to have it, but I'd think that on TV or newspaper the rate would be even lower than online if such a concept even existed. The reason you advertise isn't so that some consumer drops what they're doing and decides that instead of reading the news they should be buying motor oil for their car. Instead, when they do go out to buy motor oil for their car you want them to be thinking, "I wonder if that Penzoil stuff is any better than what I'm using?"
However, I do agree that an advertiser would probably want to pay more for search ads than they pay for display ads. A search ad is much more likely to have an impact since you're targeting people who you KNOW are looking for your product. My objection is to the suggestion that non-search ads don't really have much value - they're at least as valuable as any other kind of advertising.
And just like in most air combat on earth - the guy with the longest effective weapon and detection ranges will kill the other guy every time.
I don't really see speed being an advantage in space combat. Acceleration certainly could be, as an incoming barrage of weapons needs to be directed into some area of space and acceleration gives you the ability to not be in that space. On the other hand, if accurate and powerful lasers are involved then detection range becomes most critical (unless combat is happening at ranges where ships can accelerate sufficiently to mess up a speed-of-light firing solution).
I think that Singularity Sky had a really good depiction of space combat. Fights happened at very long ranges, so detection was difficult, and speed-of-light delay factored into all the tactics. If it takes 10 minutes for a laser to reach a target, and you see where the target was 10 minutes ago, that has a big impact on your fighting.
I don't see fancy dogfights going on in spacecraft. Honestly, I don't see them going on too much longer in aircraft either (once we get automated weapons that can fire in any direction you don't need to maneuver). If the guy is "on your tail" the computer blows him up with a laser pointed backwards at him, or the ship just flips around to fire at him. There is no lift vector in space. The only reason that there is a lot of maneuver in aircraft fights is that weapons are directed forwards only (although modern missiles are getting close to shooting in any direction), and a flying aircraft is subject to thrust, lift, and weight and they're all huge forces relative to the aircraft's mass all the time. In space you only have thrust, and compared to the ship's orbital velocity it probably won't be all that big.
I think the main advantages of spacecraft in combat will be:
1. Ability to avoid detection. 2. High weapon speed - ie lasers/etc. At long ranges weapon velocity has a huge impact on effective range. If your weapon reaches the target faster, the space that it can be in from time of aiming is smaller. 3. Ability to direct as much energy as possible into a given area - whether laser or kinetic. If you can fire a killing shot into a volume of 1km^3 you can open fire at a much longer range than if you can only kill something in a volume of 1m^3. The target is maneuvering, so you need to cover as much of its possible trajectories as possible. Needless to say, aiming will be automated - the battle could easily be determined by who shoots first. A human will at most indicate intent to destroy a target. 4. Ability to accelerate. If you can increase your distance from the center of the aim point by x meters, the volume of space the laser or flak spread has to cover increases by x^3. 5. Fuel. If you can maintain constant and apparently random acceleration then the guy shooting at you needs to cover a large area with his weapons. On the other hand, maneuvering will most likely give off EM energy aiding detection.
In a nutshell, space combat really bears little resemblance to combat on earth.
I think the design of these sorts of things is basically like steam.
The theater downloads the movie several days or maybe even a week before release. It is probably an enormous file, so it greatly eases distribution if people can do it well in advance. However, the file is encrypted and the theaters don't get the keys.
Then at midnight on opening day the keys are made available to people who are allowed to get them. Now you have 10,000 theaters all hitting the server for keys, but they're only a kilobyte or whatever so no big deal.
I suspect that there might have been a glitch in the servers that provide the keys. Without them, the theaters have a whole bunch of random-looking data.
Now, I have no doubts that in addition to this minimal amount of security they have all kinds of DRM software out the wazoo and of course that means that anything could go wrong.
I don't think there are centuries of relevant court rulings.
Uh, the whole idea of a contract is that you agree on something and then it is a done deal. Further changes require the consent of both parties. Contracts require a "meeting of the minds" and you can't have a meeting when one party decides something and just tells the other to live with it.
If there were then these terms wouldn't be in the contract.
I don't see how the one follows the other. There is a clear understanding in law that slavery is illegal, but I can write up a contract for you to sell yourself into slavery, and you can sign it if you'd like to. It isn't enforceable at all, but anybody can write whatever they want on paper.
Lots of contract terms are ruled illegal by courts. That rarely keeps them out of future contracts. Look at how many contracts disavow liability for even gross negligence. At best those kinds of terms are designed to discourage people from seeking redress in the courts (and it probably works).
If you want to know what your legal rights are with regard to a contract you've signed, call a lawyer that YOU pay. Don't call the guy who wrote the contract. They're under NO obligation to tell you that their own lawyers laughed while they were writing it up since it was a joke.
Now, the problem is that while you have the theoretical freedom to get out of your deal (assuming that AT&T does make a unilateral change), the fact is that AT&T could make life very difficult on you. First, they could just refuse to cancel your contract and demand an ETF. You could of course refuse to pay it (if they have a credit card number for you be sure to cancel the account or be sure you can reverse the charge later, and heaven help you if you have EFT set up). They could then turn you over to collections and write a nasty gram to the credit agencies.
You could of course write a letter explaining the situation. The fact is that this letter won't do you much good - some guy lending your money is going to be just as upset about you actually holding a company to the letter of their agreement as they would be about you not paying your bills. After all, maybe they'll want to screw you over in the future and they can't very well do that if their customers don't bend over.
The only way you're going to get that record expunged is to wait 7 years, or to sue in court, which is very expensive.
On the other hand, AT&T will never be able to actually collect that ETF. To do that they would need to get to court and get a declaration of judgment, and no court would side with them. However, they already have lots of leverage by messing with your credit history. They know that most consumers will just capitulate.
And this is why credit agencies need more regulation. They're basically people paid to gossip about you, and you have VERY little recourse if those records contain incorrect information. In fact, companies can discriminate against you even if you take actions that are perfectly within your legal rights.
Lovely. Too bad that centuries of court rulings say that they can't do that - regardless of what is written in the contract.
Of course, the way credit reporting companies work, it is effectively up to the consumer to try to fix their own credit damage when the phone company claims an ETF and the consumer refuses to pay it. It isn't like the credit bureaus care who is in the right.
Really? At the moment people are having a hell of a time rooting the Motorola Droid, even though many working on it know their way around Linux as well as anybody.
Last I heard it has been rooted already - it only needs to happen once per phone/firmware/etc. Gee, it took all of a week or two. How many people proficient at such things even had their hands on it in the first week on the market?
Besides, there are tools for mounting ext2/3 file systems under windows, or browsing them like a zip file. All somebody needs to do is write a guide about using such a tool, and then everybody knows.
Yup, kind of like rooting your phone. Do you think that the 10,000 people running Cyanogenmod on their G1s all figured out a local root exploit on their own?
I'll admit that the copy protection system Android uses is hopelessly lame, and is basically already broken beyond repair.
Uh, the whole concept of copy protection is broken beyond repair.
To run a program, the CPU needs to be able to read it in the clear. To copy a program, the CPU needs to be able to read it in the clear. How exactly are you going to do one without the other?
The only way you can prevent copy-protection is to retain physical control over the hardware that it runs on. Sometimes you can do that by obfuscating the hardware (smartcard concept), but that is never 100% effective. The only certain way to do it is to run your app over the web or something so that the code never leaves your hands, and then secure the servers REALLY well.
In my thinking the token security on android is enough to be able to say that they gave it a shot at least. Why add stuff that will slow down the phone to that?
Simple create a file system as a file on the SD card. The filesystem will be encrypted and loopback mounted using the standard Linux facilities for this.
Seems kind of silly.
If the apps-on-SD solution runs on ext2/3/4 or whatever, then 99.999% of phone users won't be able to figure out how to mount it anyway. 99.999% of those who could mount an ext2/3/4 will figure out how to root the phone and bypass absolutely any DRM scheme they come up with.
Your threat model is somebody who knows how to mount a unix filesystem, but who does not know how to follow a local root exploit script posted on the internet. Certainly more could be done to obfuscate things, but that's about the limit of it.
Yeah, and still only 9 judges for 330,000,000 people and they never have time to hear many important cases and decide those cases as narrowly as possible? FAIL.
Actually, that bit actually makes sense. The courts of appeals (and of course the district courts) are supposed to get it right the first time.
What else can you do, have 14 "supreme courts"? And what happens when various ones of those issue conflicting rulings? Now do we have a "super-supreme" court? At best we could have an extra layer, but that makes things even more astronomically expensive for anybody who actually has a case that needs to go through all the appeals.
One thing that would help is if the court could issue general guidances to the lower courts before a case reaches it. Those should be kept simple, and wouldn't have the full weight of actual decisions. However, if they can see that half the appeals courts are doing it one way and half are doing it another, they could just say - do it like these guys are doing it.
The problem with the current system is that in order to even get an argument before the court you need to be involved in a suit or charged with a crime. That is a huge risk to get yourself into.
Suppose congress passes some unconstitutional law that says I can't do X, but I'd rather do X anyway. Suppose I even think that the supreme court will rule that X is unconstitutional. If I do X and get caught I'll space time in prison, I'll spend money on bail, I'll spend a fortune on lawyers, and I'll spend half of my life for a few years in courtrooms. As soon as the first court rules against me I might be filing appeals from prison, or I'll have paid fines that I don't get interest back on if the ruling is reversed. And, there is a risk that the court will uphold my sentence and I'll get nothing back. So, everybody just keeps their heads down and follows the unconstitutional law.
Then think about how much work goes into creating the perfect "test case" that some special interest wants to go to the court, so that a ruling comes out that is most skewed in their direction. Why not just have the court issue guidances in the absence of a specific case, and then they can still hear appeals if they need to refine things?
The problem with the supreme court isn't size/scale - it is the over-reliance on case law.
Saving one's bacon is a very strong motivator to getting something done. We need to do more of it. Or do you think that we won't have any equipment problems as we scale up our space activities?
By that logic we should tell somebody to design a better repair robot, or we'll kill them after executing their coworkers one at a time as deadlines are missed. Saving one's bacon is a strong motivator and all that...
Sure, being in a dangerous situation in space does motivate one to get home safely. However, that isn't a reason to put people in dangerous situations in space.
And, many of those situations could have been worked out on earth much more safely. You could people in a bubble and pretend that they are enroute to mars for two years. If something breaks they have to fix it with the stuff on-hand. It would be a serious simulation and maybe we'd even let the situation get almost dangerous. However, once some threshold is met the doors could be popped open and if necessary people could be whisked away to a hospital. It would be a whole lot cheaper than an actual mission, and if the crew came up with a brilliant idea but it didn't completely work out, then we'd actually know about it rather than it being lost when the crew perished without a working radio.
Well, I tend to doubt that any court would uphold an EULA term that prevented you from fixing a defect in software that you paid for, provided that you didn't redistribute it. What would the damages be? Even if you distributed a patch, if that patch were just a binary diff then you're not redistributing anything.
Big companies usually get code escrow agreements in place when they spend lots of money on software, but that usually doesn't help ordinary consumers or small businesses. Even the big companies rarely take advantage of the escrow.
Slashdot Mirror
Well, the dicing up will limit the power of individual banks.
However, you are correct that the problem was that EVERYBODY was acting stupid. The problem is that when all your competitors are making risky loans and selling them to investors as safe loans you have two choices:
1. Do the same thing.
2. Watch all the investors pull out all of their money and give it to those who do #1.
After all, if you're an investor, which fund would you invest in? The AAA bond fund yielding 15%, or the AAA bond fund yielding 5%?
A big problem is that individual investors are far too insulated from the investment decisions. Most people have a choice of maybe 20 mutual funds in their 401k - usually from one or two investment houses. At best they get a quarterly statement and some audits. Nobody really works for the individual investor in such a setup - the funds make their pitches to their employers. People are told to just set aside 10% of their income and watch it grow, and that investments are too complicated for them to deal with. Nobody is accountable when things go wrong.
The free market only works when consumers can make educated choices. That is why there was a ton of regulation after the 1929 crash.
Yup, that would be the other thing that I would legislate (besides separating the billing items for phone subsidies and plan fees) - the advertised cost of the plan is what consumers have to pay, and not a penny more. For that matter, unless a consumer agrees there can be no incremental costs beyond the advertised plan cost. Carriers may not offer a discount on the plan if you agree to pay extra for texts/etc, either. The whole surprise-$500-SMS-bill thing has to go away.
Add-ons should be add-ons, and if a consumer says that they're not interested in the add-ons then the carrier cannot charge for them, period. By all means they can block SMS or MMS or whatever if the customer hasn't paid for them, but they can't charge for them if they get used if the consumer didn't request this.
In the US they're also like an installment plan, but you pay the installments whether or not you get a free phone out of the deal. It is a BIG ripoff, and it is the reason why I have a collection of perfectly fine old phones lying around - if I pay the same to keep the old phone or get a new one, why wouldn't I want the new one?
Agreed. There should also not be any early-termination fee for any aspect of the service other than the phone cost, and you shouldn't have that at all if you bring your own phone. Also, you should be able to pay out the remainder of your phone rental term without having phone service. So, if my phone fee is $10/month and my plan is $100 per month and I want to switch two months early at most it should cost me $20 (paid over the next two months).
And all phones should be unlocked - at most the company that you're renting them from should have a lein against the phone. I'm fine with them demanding the return of the phone if you fail to make payments for it, but if I cancel my deal with AT&T I should be able to use my existing phone with T-mobile or whatever as long as I pay for phone-only cost for the remainder of my AT&T plan.
Once this happens you could mostly deregulate phones, as they are a commodity. The plans are the part that are a natural monopoly.
It is only a free market if anybody is free to enter the market. That is certainly not the case with a phone company. How much capital do you think it would take to actually create a credible cell-phone provider?
Mobile phones are essentially a commodity.
Agreed - phones are a commodity.
Phone PROVIDERS, on the other hand are NOT. In the US there are really only four with any significant coverage, and only two with fairly wide coverage. That is not really a "free market."
And only T-mobile at this point has plans that don't involve phone subsidies, which makes the commodity status of phones an issue as well. What good does it do me to buy my own phone when the carrier is going to just charge me as if they had given me a subsidized one?
They should be required to separate the phone subsidy in the plan so that you can choose to get the plan with or without the subsidy.
If you cancel service early you could opt to pay the termination fee, or pay ONLY for the subsidy portion of the plan for the remainder of your contract (not the FULL plan). If you don't have this clause then the providers will play games with the subsidy value so that people still pay more.
Will it actually be reasonably supported in such a configuration? Google doesn't even support an open-source version of android that can actually make phone calls. They have lots of source code, but no releases or anything with QA behind it. Integrating proprietary drivers needed to actually make it work is a fairly manual process, and there are no guides/etc.
It seems like they're only willing to make it open-source to the extent that it doesn't cost them anything and it doesn't create viable competition.
From what I've seen Google open source projects don't really operate like normal open source projects. It is a proprietary cathedral model that publishes most of their source code, without much regard to whether the open source product is in itself something anybody would actually want to use. At least with mysql or whatever you can download the source for an actual release, and do a make install and get more-or-less exactly what you'd get if you just used their product.
From what I've seen on the web wave is open source in the same sense that android is open source. There are a bunch of files on a CVS server, with some build instructions that creates a program that does something, but it isn't something they actually intend anybody to actually use. In the case of android their open source code results in an OS that doesn't have access to the cell network or wifi, and which can't even blink the LED on the front from what I understand. Their build instructions are a real mess too as it requires a ton of manual effort to get the proprietary blobs added to it so that it works. It isn't like you can just put some proprietary bits in a directory, hit make, and end up with an image.
Plus, their only implementation runs on Java, and I don't really have half a gig of RAM to waste having that in memory all the time. Most servers that don't get much traffic can run in a lot less!
It isn't a bad concept, but I still don't see it taking off. Maybe if I ran a forum and the open-source version of wave were robust enough I might replace phpbb or whatever with it. However, most open-source forum systems are actually designed to be used by people, and they're not just intended for developers to contribute to one particular implementation of it.
Of course, who is to say what she actually said, but I agree - the word "love" is REALLY strong. Unless you're a teenager you shouldn't be using it to describe every little infatuation.
Relationships at work can be very messy - particularly when they involve somebody you work closely with. You need to take things carefully one step at a time.
Yup. I know somebody who was involved in some software that dealt with clinical trial data and they were ultra-paranoid about this sort of stuff (to the point where access to many fields in the database was need-to-know). A full date of birth was considered as sensitive as a name/address or ID code (and they didn't even track SSNs). Even the age or year-of-birth was considered sensitive if the age was above a threshold (how many 97-year-olds are there?).
They clearly didn't talk to a lawyer versed in this stuff before releasing this data.
You don't want to light up a radar or anything like that because it can be detected FAR beyond its own detection threshold unless you're very sure you need a positive firing solution and that's the way to get it.
Yup. Plus at longer ranges you get the radar return at about the same time as you get the incoming volley from your target who is firing down the bearing of your radar. By the time you fire back the entire area around you is being blasted at.
Now, if you can sneak in to what you have reason to believe is close range (negligible light travel time) then radar can be used pretty safely, since you'll be more prepared to make use of the data than they will.
I agree that drones are the way to go. In fact, missiles might actually become rockets with sensors and a laser on them. You fire them at the approximate area the enemy is in, and then the missile detects the target and fires at it when it is within range. Being a small object it could get in close, and it need not have energy for sustained maneuver or lots of shots. An x-ray laser on a rocket could be a really devastating weapon. However, the missile will need to be able to boost without giving away its trajectory, or it will be an easy target when it is ballistic.
I think the only reason the tank/dps/healer thing works is because of dumb AI. So, you're taking out a whole army of goblins or whatever in some cave. You come up to groups of 2-3 bad guys at a time, in constrained corridors so that your tank can actually block them from getting past.
There are a couple of obvious counters to this. The first group of 2-3 puts their biggest tanks in your way to slow you down, while the last guy does a Paul Revere and every goblin in the lair is headed your way. While they hold you up a bunch of them leave via their back door and circle around to come at you from behind. The goblins of course put their own tanks in front, and defend themselves from prepared positions. If you come across a long corridor you can bet that there will be archers chopping you to pieces from the far end.
However, the game AI just sends people at you in mobs of 2-3 to be chewed up one at a time.
In open terrain any larger group of enemies would certainly flank your tanks and take out your rear forces. They would never bunch up to allow an area-effect spell to do anything.
I do agree with some of the posters that a fantasy game shouldn't be completely realistic. In any REAL war with reasonably matched forces almost nobody makes it through multiple battles alive. Most people kill at most a few enemy soldiers and then are killed themselves. That doesn't make for good gameplay.
However, I am not big on MMORPGs because they are way too formulaic - the grind, roles, etc. I can't just go and do what I want to do unless I only want to experience a small part of the game or tick off my party/guildmates/whatever.
Singularity Sky (Charles Stross) has a similar depiction of space battle.
The bigger your weapons the longer your ranges get - since you can cover a bigger volume of space and the errors in targeting solutions matter less. At light-minute distances it would create an entirely new set of tactics.
Agreed - a big factor in space battles is going to be getting a firing solution on the enemy, and then blasting as much space as possible in the area where he is going to be. That might be a bullet every 10m, or laser fire directed across a cone, or whatever.
To detect the guy you need light to travel from him to you. To kill him you need your weapons to travel from you to him. That is at least twice the light-travel time for your weapon to get there, so as soon as the other guy's acceleration limits him to a volume of space you can target in twice the light travel time then he's dead. Stealth could be critical, as it could let a smaller vessel get closer to a big one and negate its ability to direct energy into a larger volume of space (plus the smaller vessel can probably accelerate more, increasing the volume of space that must be targeted).
I wouldn't be surprised if 99% of the actual shooting is fully automated. The only thing humans will be doing is telling the computer where to go and what to shoot at. The ship would be under constant maneuver with the computer doing its best to get to the general area it is asked to go to, while killing anything it can on the way.
Coordination of multiple ships could be really tricky - your communications also suffer travel delay due to the speed of light. Unless you direct your communications carefully they will also broadcast your position.
I do agree with your point that electronic warfare will be important. If you can trick out the enemy's radar then you greatly reduce his effective firing range, as the volume of space to be targeted goes way up.
In space combat that divergence angle will be essential - or you won't hit your target unless you're at very close range.
Suppose you open fire at a range of 10 light seconds? Do you think that a microscopic beam will have any chance of hitting the target at all?
I think that is the one thing that all this discussion is missing - the concept of VERY long range warfare in space.
When those guys from earth launch to head out to mars to conqueror it, they're going to be under observation and potential attack as soon as the light from the booster rocket travels to mars. Chances are the mars base will be under direct missile attack from earth as well. Now, at those long ranges no weapons will be effective, but I think that the ranges where you need to start worrying about incoming fire are going to be a LOT longer than people appreciate.
Ships will be destroyed when they reach a range where they are fired upon by a ship that can direct a lethal amount of energy into a volume of space large enough to contain most of the possible trajectories the target can take based upon the round-trip time of the sensors plus the weapons fire. Ships will constantly generate maximum thrust in semi-random directions to maximize the volume of space their opponent has to cover. They also have to guess when they are detected, since rocket thrust will undoubtedly broadcast their position at a longer range than a radar return will. Ships might even avoid using radar, as a radar gives your target a bearing to you that is twice as recent as the bearing you have on (since he sees you when the ping gets to him, and you see him when the ping returns).
Just think of what combat looks like when the travel time of light is noticeably wrong - that is what interplanetary space combat will look like. Now, in the near future there could certainly be some closer-range stuff before all the technology matures, but I think that ground-based platforms will have the advantage for a while since they can expend a LOT more energy.
On the forums for "search engine optimization", one discovers that ad clicks from Google search results tend to result in sales, while ad clicks from Google ads on non-Google sites (what Google euphemistically calls the "Google Content Network") don't. 50% of ad clicks come from 10% of the user base, and that 10% doesn't buy anything.
Actually, all that you know is that the people who view your ads on non-Google sites don't buy the items at that moment by clicking on the link and immediately buying it.
You have no idea how many people actually bought your product after VIEWING (not clicking) the ad.
By your argument a superbowl ad has zero impact, because not a single person clicked on their TV screen during the game and bought a Gilette razor or whatever.
I never really understood the obsession with click-through rates in the online advertising world. Sure, it is nice to have it, but I'd think that on TV or newspaper the rate would be even lower than online if such a concept even existed. The reason you advertise isn't so that some consumer drops what they're doing and decides that instead of reading the news they should be buying motor oil for their car. Instead, when they do go out to buy motor oil for their car you want them to be thinking, "I wonder if that Penzoil stuff is any better than what I'm using?"
However, I do agree that an advertiser would probably want to pay more for search ads than they pay for display ads. A search ad is much more likely to have an impact since you're targeting people who you KNOW are looking for your product. My objection is to the suggestion that non-search ads don't really have much value - they're at least as valuable as any other kind of advertising.
And just like in most air combat on earth - the guy with the longest effective weapon and detection ranges will kill the other guy every time.
I don't really see speed being an advantage in space combat. Acceleration certainly could be, as an incoming barrage of weapons needs to be directed into some area of space and acceleration gives you the ability to not be in that space. On the other hand, if accurate and powerful lasers are involved then detection range becomes most critical (unless combat is happening at ranges where ships can accelerate sufficiently to mess up a speed-of-light firing solution).
I think that Singularity Sky had a really good depiction of space combat. Fights happened at very long ranges, so detection was difficult, and speed-of-light delay factored into all the tactics. If it takes 10 minutes for a laser to reach a target, and you see where the target was 10 minutes ago, that has a big impact on your fighting.
I don't see fancy dogfights going on in spacecraft. Honestly, I don't see them going on too much longer in aircraft either (once we get automated weapons that can fire in any direction you don't need to maneuver). If the guy is "on your tail" the computer blows him up with a laser pointed backwards at him, or the ship just flips around to fire at him. There is no lift vector in space. The only reason that there is a lot of maneuver in aircraft fights is that weapons are directed forwards only (although modern missiles are getting close to shooting in any direction), and a flying aircraft is subject to thrust, lift, and weight and they're all huge forces relative to the aircraft's mass all the time. In space you only have thrust, and compared to the ship's orbital velocity it probably won't be all that big.
I think the main advantages of spacecraft in combat will be:
1. Ability to avoid detection.
2. High weapon speed - ie lasers/etc. At long ranges weapon velocity has a huge impact on effective range. If your weapon reaches the target faster, the space that it can be in from time of aiming is smaller.
3. Ability to direct as much energy as possible into a given area - whether laser or kinetic. If you can fire a killing shot into a volume of 1km^3 you can open fire at a much longer range than if you can only kill something in a volume of 1m^3. The target is maneuvering, so you need to cover as much of its possible trajectories as possible. Needless to say, aiming will be automated - the battle could easily be determined by who shoots first. A human will at most indicate intent to destroy a target.
4. Ability to accelerate. If you can increase your distance from the center of the aim point by x meters, the volume of space the laser or flak spread has to cover increases by x^3.
5. Fuel. If you can maintain constant and apparently random acceleration then the guy shooting at you needs to cover a large area with his weapons. On the other hand, maneuvering will most likely give off EM energy aiding detection.
In a nutshell, space combat really bears little resemblance to combat on earth.
I think the design of these sorts of things is basically like steam.
The theater downloads the movie several days or maybe even a week before release. It is probably an enormous file, so it greatly eases distribution if people can do it well in advance. However, the file is encrypted and the theaters don't get the keys.
Then at midnight on opening day the keys are made available to people who are allowed to get them. Now you have 10,000 theaters all hitting the server for keys, but they're only a kilobyte or whatever so no big deal.
I suspect that there might have been a glitch in the servers that provide the keys. Without them, the theaters have a whole bunch of random-looking data.
Now, I have no doubts that in addition to this minimal amount of security they have all kinds of DRM software out the wazoo and of course that means that anything could go wrong.
I don't think there are centuries of relevant court rulings.
Uh, the whole idea of a contract is that you agree on something and then it is a done deal. Further changes require the consent of both parties. Contracts require a "meeting of the minds" and you can't have a meeting when one party decides something and just tells the other to live with it.
If there were then these terms wouldn't be in the contract.
I don't see how the one follows the other. There is a clear understanding in law that slavery is illegal, but I can write up a contract for you to sell yourself into slavery, and you can sign it if you'd like to. It isn't enforceable at all, but anybody can write whatever they want on paper.
Lots of contract terms are ruled illegal by courts. That rarely keeps them out of future contracts. Look at how many contracts disavow liability for even gross negligence. At best those kinds of terms are designed to discourage people from seeking redress in the courts (and it probably works).
If you want to know what your legal rights are with regard to a contract you've signed, call a lawyer that YOU pay. Don't call the guy who wrote the contract. They're under NO obligation to tell you that their own lawyers laughed while they were writing it up since it was a joke.
Now, the problem is that while you have the theoretical freedom to get out of your deal (assuming that AT&T does make a unilateral change), the fact is that AT&T could make life very difficult on you. First, they could just refuse to cancel your contract and demand an ETF. You could of course refuse to pay it (if they have a credit card number for you be sure to cancel the account or be sure you can reverse the charge later, and heaven help you if you have EFT set up). They could then turn you over to collections and write a nasty gram to the credit agencies.
You could of course write a letter explaining the situation. The fact is that this letter won't do you much good - some guy lending your money is going to be just as upset about you actually holding a company to the letter of their agreement as they would be about you not paying your bills. After all, maybe they'll want to screw you over in the future and they can't very well do that if their customers don't bend over.
The only way you're going to get that record expunged is to wait 7 years, or to sue in court, which is very expensive.
On the other hand, AT&T will never be able to actually collect that ETF. To do that they would need to get to court and get a declaration of judgment, and no court would side with them. However, they already have lots of leverage by messing with your credit history. They know that most consumers will just capitulate.
And this is why credit agencies need more regulation. They're basically people paid to gossip about you, and you have VERY little recourse if those records contain incorrect information. In fact, companies can discriminate against you even if you take actions that are perfectly within your legal rights.
Lovely. Too bad that centuries of court rulings say that they can't do that - regardless of what is written in the contract.
Of course, the way credit reporting companies work, it is effectively up to the consumer to try to fix their own credit damage when the phone company claims an ETF and the consumer refuses to pay it. It isn't like the credit bureaus care who is in the right.
Really? At the moment people are having a hell of a time rooting the Motorola Droid, even though many working on it know their way around Linux as well as anybody.
Last I heard it has been rooted already - it only needs to happen once per phone/firmware/etc. Gee, it took all of a week or two. How many people proficient at such things even had their hands on it in the first week on the market?
Besides, there are tools for mounting ext2/3 file systems under windows, or browsing them like a zip file. All somebody needs to do is write a guide about using such a tool, and then everybody knows.
Yup, kind of like rooting your phone. Do you think that the 10,000 people running Cyanogenmod on their G1s all figured out a local root exploit on their own?
I'll admit that the copy protection system Android uses is hopelessly lame, and is basically already broken beyond repair.
Uh, the whole concept of copy protection is broken beyond repair.
To run a program, the CPU needs to be able to read it in the clear. To copy a program, the CPU needs to be able to read it in the clear. How exactly are you going to do one without the other?
The only way you can prevent copy-protection is to retain physical control over the hardware that it runs on. Sometimes you can do that by obfuscating the hardware (smartcard concept), but that is never 100% effective. The only certain way to do it is to run your app over the web or something so that the code never leaves your hands, and then secure the servers REALLY well.
In my thinking the token security on android is enough to be able to say that they gave it a shot at least. Why add stuff that will slow down the phone to that?
Simple create a file system as a file on the SD card. The filesystem will be encrypted and loopback mounted using the standard Linux facilities for this.
Seems kind of silly.
If the apps-on-SD solution runs on ext2/3/4 or whatever, then 99.999% of phone users won't be able to figure out how to mount it anyway. 99.999% of those who could mount an ext2/3/4 will figure out how to root the phone and bypass absolutely any DRM scheme they come up with.
Your threat model is somebody who knows how to mount a unix filesystem, but who does not know how to follow a local root exploit script posted on the internet. Certainly more could be done to obfuscate things, but that's about the limit of it.
So, why burden the already-sluggish CPU with AES?
Yeah, and still only 9 judges for 330,000,000 people and they never have time to hear many important cases and decide those cases as narrowly as possible? FAIL.
Actually, that bit actually makes sense. The courts of appeals (and of course the district courts) are supposed to get it right the first time.
What else can you do, have 14 "supreme courts"? And what happens when various ones of those issue conflicting rulings? Now do we have a "super-supreme" court? At best we could have an extra layer, but that makes things even more astronomically expensive for anybody who actually has a case that needs to go through all the appeals.
One thing that would help is if the court could issue general guidances to the lower courts before a case reaches it. Those should be kept simple, and wouldn't have the full weight of actual decisions. However, if they can see that half the appeals courts are doing it one way and half are doing it another, they could just say - do it like these guys are doing it.
The problem with the current system is that in order to even get an argument before the court you need to be involved in a suit or charged with a crime. That is a huge risk to get yourself into.
Suppose congress passes some unconstitutional law that says I can't do X, but I'd rather do X anyway. Suppose I even think that the supreme court will rule that X is unconstitutional. If I do X and get caught I'll space time in prison, I'll spend money on bail, I'll spend a fortune on lawyers, and I'll spend half of my life for a few years in courtrooms. As soon as the first court rules against me I might be filing appeals from prison, or I'll have paid fines that I don't get interest back on if the ruling is reversed. And, there is a risk that the court will uphold my sentence and I'll get nothing back. So, everybody just keeps their heads down and follows the unconstitutional law.
Then think about how much work goes into creating the perfect "test case" that some special interest wants to go to the court, so that a ruling comes out that is most skewed in their direction. Why not just have the court issue guidances in the absence of a specific case, and then they can still hear appeals if they need to refine things?
The problem with the supreme court isn't size/scale - it is the over-reliance on case law.
Saving one's bacon is a very strong motivator to getting something done. We need to do more of it. Or do you think that we won't have any equipment problems as we scale up our space activities?
By that logic we should tell somebody to design a better repair robot, or we'll kill them after executing their coworkers one at a time as deadlines are missed. Saving one's bacon is a strong motivator and all that...
Sure, being in a dangerous situation in space does motivate one to get home safely. However, that isn't a reason to put people in dangerous situations in space.
And, many of those situations could have been worked out on earth much more safely. You could people in a bubble and pretend that they are enroute to mars for two years. If something breaks they have to fix it with the stuff on-hand. It would be a serious simulation and maybe we'd even let the situation get almost dangerous. However, once some threshold is met the doors could be popped open and if necessary people could be whisked away to a hospital. It would be a whole lot cheaper than an actual mission, and if the crew came up with a brilliant idea but it didn't completely work out, then we'd actually know about it rather than it being lost when the crew perished without a working radio.
Well, I tend to doubt that any court would uphold an EULA term that prevented you from fixing a defect in software that you paid for, provided that you didn't redistribute it. What would the damages be? Even if you distributed a patch, if that patch were just a binary diff then you're not redistributing anything.
Big companies usually get code escrow agreements in place when they spend lots of money on software, but that usually doesn't help ordinary consumers or small businesses. Even the big companies rarely take advantage of the escrow.