I guess that's why computer programs in the future don't have bugs: they have Heisenbug compensators!
:-)
Re:Article has some valid points
on
Is Linux Dead?
·
· Score: 2
Second, Linux still has not gained any major inroads in the personal computer world. Yes, I know WalMart sells Linux-able PCs...
...
The point? Take this article in stride, and take its criticisms to heart - Linux has failed in 10 years to make any strong inraods into the personal computer market, commercially speaking.
It seems to me that some (perhaps many) are underestimating the importance of WalMart selling PCs with Linux preinstalled.
I believe this is possibly one of the most important in-roads Linux has made in the desktop world since its inception, because:
It immediately eliminates the vast majority of the configuration issues people tend to have with Linux. Most such issues are related to the initial setup, but that has already been done.
It immediately eliminates the greatest advantage Microsoft had over Linux: preinstalls.
It puts Linux in front of the average person more than anything else could. I can't think of any faster way to get the Linux user experience to improve than to put it in front of the average person and deal with the feedback.
I think the fact that WalMart is selling PCs with Linux preinstalled is a huge commercial inroad, something few people would have even dreamed of a couple of years ago. Don't be so quick to dismiss it.
For instance, how much money is actually spent on software development and acquisition per year right now? If it's, say, $120 billion, then that $60 billion figure is a very high number, an additional 50% that's spent on top of the acquisition and development costs. But if $6 trillion per year is spent on software (not likely, of course), then that $60 billion per year represents only 1% of the money actually spent on software.
Then, of course, you have to account for the price of quality. It is possible to write extremely high quality software: the guys who write software for the space shuttle do so, for instance. But such robust software costs a fortune and (perhaps more importantly) takes a great deal of time to develop.
In reality, we have nobody to blame but ourselves for the poor quality of software (open source software aside, though it's not immune from poor quality). If people in general were willing to pay for greater reliability for software, those who provide better quality would be doing better than those who don't, and other software houses would take notice of this and start to produce better quality software themselves.
Rather the opposite has happened, of course. And it's not just limited to software, either. Look at the goods sold at such places as Wal-Mart: such goods are often of inferior quality to other similar goods, but they often sell better regardless.
When are we finally going to figure out that people want what's barely good enough for the cheapest price possible? Quality only matters when it's low enough that the product in question falls into the "not good enough" category. Warpage of the market by monopolies aside, the market has decided that the quality of software available is good enough. Otherwise, that software wouldn't sell.
Things might be different if people were thoughtful enough and had enough knowledge to determine the total cost of ownership, or to even care about such a thing. But they're not, probably due in large part to the dumbing down of the U.S. population.
The Congress shall have Power To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries
-- United States Constitution
The purpose of copyright is to promote progress, to entice authors and inventors to release their works and discoveries to the public.
But that is not an end unto itself. The true end is the benefit to society that the release of such works brings.
Now, remember that the whole incentive here, the entire reason for granting the monopoly privilege of copyright, is to allow the originators of works to make money from their works, which in turn (theoretically) gives them incentives to release their works to the public.
When you publish something on the web, you're publishing your works for free, unless you go to the extra trouble of implementing some kind of access control. The Wayback Machine won't work on a site that has access control, so all it ends up archiving is stuff that was published for free public consumption.
So the real question is: if a work has already been released for free to the general public, how would letting authors restrict the republication of that work after the fact bring greater benefit to society than not letting the author impose such restrictions?
My opinion is that it is much more beneficial to society as a whole if the release of a work for free public consumption automatically implied that members of the public have the right to redistribute that work. So if an author doesn't want people in the general public to be able to redistribute his work, he has to control who receives the work and who doesn't. Certainly requiring payment for the work in question is sufficient to meet the requirement of controlling access. But whatever method the author chooses, it should be one that makes it clear that the work in question is not being released for free to the public.
I'm seeing lots of comments about how software isn't "engineering" or a "science." Well, I hate to break it to you guys but engineering is just as much hackery as anything else. Yes, there is some science to figuring out how much stress a material can take, but it's not like you can mathematically prove that an airplane is safe.
Well, yes, you actually could, but to do so you'd have to run it through a computer simulation, which is likely to have bugs...:-)
Seriously, though, engineering isn't the same as programming computers, and I'll tell you why: because every engineering field ultimately has to answer to the laws of physics. Those laws are well-known and well-understood. When you engineer something, you work backwards from the laws of physics. When an engineer designs a bridge, he takes the requirements (the problem specification) and selects a basic design which fits within the problem specifications. The limitations of various designs are well known because the laws of physics dictate those limitations. The design specifies the form of the individual pieces, but the laws of physics determine the necessary characteristics (strength, weight, etc.) of those pieces. Only if the physical characteristics of one or more of those pieces are beyond our ability to manufacture within the cost constraints is the design unworkable, and that is something that can easily be determined ahead of time. If the bridge breaks in use, it's either because the engineer (probably accidentally) ignored the laws of physics or because the engineer didn't understand the problem space well enough.
The laws of physics ultimately dictate everything about a physical object. The individual pieces of the bridge in the example above are designed and constructed in precisely the same way that the bridge is: the smaller parts which comprise a bridge part are all subject to the same laws of physics that the bridge itself is subject to, so their suitability can be analyzed the same way.
And so as a result, you can know with a great deal of certainty whether or not a physical design will break, so it's only when the initial problem specification is wrong, when the laws of physics are not fully recognized and dealt with, or when there are errors made during the actual construction that a manufactured object will unexpectedly fail in use.
Software is very different: there are no laws of physics that govern the functioning of a computer program. In fact, from the program's point of view, the only limitations it has are imposed either by the computer hardware or by the operating system (and the operating system, being a program itself, has no limitations placed on it other than those imposed by the hardware).
While this means that the software designer has a lot more freedom, it also means that the software designer has no consistent way to determine whether or not his design functions properly: it's entirely possible for his design to meet the requirements of the problem at hand but to still malfunction when subject to inputs that are unanticipated.
Which leads to another distinction: the kinds of input to a physical object are limited by the same laws of physics which determine how that object will behave when subject to those inputs. The elements of the bridge are subject to the force of gravity, to the force of the wind, and to the force of the loads put on it by its users, but all of those forces are tightly bound: the force of gravity is a constant, we know that the chance that the force of the winds will exceed some large value is so low that it's sufficient to design the bridge with that value in mind, and the forces that can be placed on it by its users are limited by the density of the materials being pushed across the bridge. And because the forces on the bridge are tightly bound, the forces the individual pieces will undergo are also tightly bound. The laws of physics are unrelenting, so input "bounds checking" is automatic and complete.
But software is completely different in that regard. The software itself is what creates and polices the limits on its input, and because the software itself is responsible for enforcing many of its own limits, it follows that input "bounds checking" is not automatic and complete, but is itself subject to the same possible flaws that the rest of the software is subject to!
As a result of all this, creating a piece of software is more akin to creating an entire universe, complete with a set of governing laws, than to creating a bridge or other engineered object, because not only must the creator define the design, he must also define the rules under which it will operate.
One other thing: because the laws of physics are unchanging, it's possible to construct "standard" objects which can be used to build larger objects. And because the engineering on those component objects has already been done, the engineer can safely rely on their behavior. But the software architect's choices aren't so clear-cut. As I said, software not only must solve the problem it was designed to solve, but must also define the set of rules under which it is to operate. So a "pre-engineered" software component not only defines how it behaves, it defines the rules under which it will operate. Both of these things must be considered when the architect is deciding whether or not to use it, and the latter part can often have a huge impact on the entire project, because the entire project may have to operate under those rules, too. This is why in software it's often easier to just "roll your own" components rather than rely on components which have already been written.
Imagine where we'd be if most electronics shops decided to design and construct their own base components (resistors, capacitors, etc.) rather than use standard off-the-shelf ones. Imagine if most bridge designers had to design and construct the beams, trusses, cables, etc., used to build a bridge, rather than relying on standard components for those things. Things would be a lot more expensive to design and build, and there would probably be a lot more errors made.
Perhaps that's an argument for standardized components in software. Nice as that sounds though, I don't think you're going to see that happening, because the curse software designers operate under is also a blessing: the ability to define your own rules means that anything is possible. In the real world, the laws of physics always act to limit what we can do and how we can do it. If we want the freedom to do anything we want in software, we must accept that the price of that freedom is the greater possibility of error.
Any boss will respect you for asking what you're worth. If you don't - you're the naive nerd at the back of the room.
You're not hired because of your pretty eyes, loyalty and what not, you're hired because of your skills. If you deliver, you're certainly not the first to go when things get tough.
If that's how the boss really thinks, then you should definitely stay -- because your boss is exceptional.
Most middle managers understand nothing but the short term bottom line. That means if you cost more to retain than someone else who claims to have roughly the same set of skills but who is/would be paid half of what you are (especially if that other person already works for the company), then you'll be outta there as soon as the boss can arrange it.
Jesus, haven't you people learned anything by now? In the Real World, quality doesn't matter. Ability doesn't matter. The only things that matter are expedience, appearance, and cost. Why do you think there's so much crap in the computer world (and in the rest of the world, for that matter. Why else do you think Wal-Mart is so successful)? If ability were valued more than expedience, appearance, and cost, there'd be a lot less crap that barely works out there.
So if the boss doesn't give a shit about how well the product works, why do you think he'll give a shit about how good you are at what you do?
So in theory all three will logically head toward the equalibrum point just a bit over their cost. Given that all three have been playing this game for a while, they probably have a good idea of where that equilibrum point is for new products.
Yeah, but they'll all end up at the same price point only if their internal costs are identical.
How likely do you think that is?
That's why I suspect it is price fixing (or, at least, collusion).
I disagree. I think we'll hear just as much whining from Microsoft and the "pro business" crowd no matter who renders the judgement. The whining we'll hear, both the amount and the type, will be based strictly on the judgement itself and will have nothing to do with who renders it.
In other words, anyone who would render a judgement against Microsoft is going to be considered "biased" by both Microsoft and the "pro business" crowd you referred to no matter what lengths the person rendering judgement went to in order to be fair.
After all, the whole purpose of the whining is to reduce or eliminate the effects of the judgement, so why would it matter who rendered it?
Re:Perhaps OSS Zealots shouldn't piss off Blizzard
on
Warcraft III Gone Gold
·
· Score: 2
However, certain factions, namely the Linux zealots eager to boycott and bitch and try to steal intellectual property and server code and processes they DON'T OWN are rocking the damn boat for the majority of Linux/non-Windows/non-Mac people who wouldn't mind a Blizzard game.
Yeah. We all know it's perfectly reasonable for Blizzard to sue bnetd.org, because reverse engineering Blizzard's Ultra Secret game protocol so that you don't have to be connected to the internet to play multiplayer Starcraft is so wrong that it should be a crime. I mean, only criminals would reverse engineer a protocol in order to get something they legally own to work the way they think it should.
Blizzard's most likely simply bothered by the fact that the existence of bnetd decreases the intrinsic value of Battle.Net. Perhaps they have/had some plans in the wings to move Battle.Net over to a for-pay service; if they only allow their games to talk to *their* community (Battle.Net), then they've got a clear, guaranteed way to at least make some money.
Yeah. Because everyone knows that there's no way to make any money by selling games. I mean, if you want to make enough cash to buy a Ferrari, you clearly have to do something else.
And poor Blizzard is so cash strapped that they don't have the money to develop a new game, their only hope is to milk Battle.Net.
I'm glad that when Microsoft FINALLY gets their Official Spanking from the court, we won't have to put up with as much whining from MS and the "pro-business" crowd (note the quotes) about how the judge was biased, the trial was unfair, etc.
Nonsense. They'll be making that claim no matter what.
People (e.g. Ralph Nadar) say this all the time. What good would it actually do?
It depends on how it's done.
As far as I'm concerned, dissolution of Microsoft's business charter should be done by (a) selling off all their assets and (b) taking all the resulting proceeds plus the money in the bank and distributing it equally amongst all working citizens of the United States, except for those who own stock in the company.
The reason for doing it that way? To make it clear to everyone that if your company breaks the law too much, it will die and the people who have invested in it will lose everything they've invested.
Seems to me that it's the only way to keep companies from behaving like Microsoft: make it clear that doing so will, in the long run, be so unprofitable that the people running the show won't even think about it.
Instead, go out and buy a cheap server. You can get a AtlonXP 1800 w/ 512MB of RAM and 100 GB of disk for around 200-400 dollors if you put it together yourself.
I agree, except that if you're going to be setting up a server for people to store their data on, it has to be extremely reliable.
So don't skimp when you're building it. Use a P4-based system (because the P4 will throttle down automatically to prevent itself from overheating, so if the CPU fan dies the system will still run) and build a RAID-5 or mirrored disk subsystem. You might also consider building a standby system to which the data gets mirrored every night, so that you have a fallback in case the first system dies.
An experiment with using laser propulsion on aircraft saw a serious setback today.
The experiment involves using a laser to heat up a droplet of water on the surface of the aircraft. The conversion of the water to a gas is what drives the aircraft.
But today the prototype aircraft was destroyed.
"We fired the laser and POOF, the airplane went up in a puff of smoke!" said the aircraft's chief designer, Takashi Yabe.
Traditionally, lasers have been associated with shooting aircraft down, but this research is attempting to prove that they can be used to propel them as well.
"Unfortunately, today, we managed to prove that lasers really can be used to shoot airplanes down. At least paper ones!" Yabe said.
The researchers are all set to try again. "Whoever decided that we should use paper was obviously a moron", Yabe said, but the other researchers only nodded and Yabe turned red with embarrassment immediately after saying that.
A spokesperson for the U.S. Department of Defense had no comment on today's incident, but word from our sources is that the President will attempt to convince Russia and Europe to start building airplanes out of paper sometime this week.
It is still your responsibility to navigate, using the proper charts.
I believe I said that already. Despite that, the fact remains that when you're being vectored around by ATC is, in general, exactly the most difficult time to keep track of your position relative to terrain.
Well, maybe it isn't for you bumbling civilians but it is for those of us in the military.
No, it is for us "bumbling civilians", too. But unlike you military guys, us "bumbling civilians", at least in general aviation, don't have the benefit of having a person on board (the navigator) whose sole purpose in life is to figure out where we are or, in the case of single pilot operations (which is what most of us general aviation civilians engage in), a terrain following radar.
Oh, and lest you get too cocky, remember that it was a military guy flying the B-17 that crashed into the Empire State Building way back in the day.
The ATC / Control Tower is never responsible for controlling the aircraft, nor actually guiding aircraft between hills/etc.
Uh, not quite.
What you say is true for aircraft operating under VFR (Visual Flight Rules), but not for aircraft operating under IFR (Instrument Flight Rules). When you're operating IFR, if the controller gives you a vector, you follow the vector.
Now, that said, it's still the pilot's responsibility to clear terrain, but without GPS-based terrain avoidance it's damned near impossible to guarantee terrain avoidance when you're IFR, in the soup, and on vectors unless you're intimately familiar with the terrain and your own location.
The bummer of it is that it's when you're relatively low (thus closer to terrain), in relatively busy airspace (you're talking to approach control, who's usually talking to a bunch of different airplanes, and it's a party line so you have to continuously monitor the ATC frequency listening for your callsign), and are having to do a lot of different things (like listen to the ATIS [Automated Terminal Information Service] at the destination airport to get weather, runway info, etc., brief the approach you'll be making into the destination airport, set up the radios and the navigation instruments for the approach, etc., all while listening to the ATC party line frequency for instructions) that it's most likely that you'll be receiving vectors from ATC. That means that if you're in the soup, you have very little time to handle terrain avoidance, which is why GPS-based terrain avoidance gear is such good stuff to have.
Bottom line is that ATC can vector planes into terrain, and it actually has done so occasionally. It doesn't happen often. But it happens. And that's with all the equipment working, at that.
Yeah, it could be a lot better. But the FAA governs aviation with an iron fist, so you're only allowed to use technology that it deems suitable. And the FAA moves at a glacial pace when it comes to allowing new technology in airplanes. It's why modern general aviation airplanes are still using aircooled piston engines that were literally designed back in the 1940's, and why everyone is still using AM radios for communications.
While I agree there are a few things I missed, my main point remains: the costs which dominate the operation of a network provider do not significantly vary with the amount of bandwidth being provided. Expertise costs money no matter whether it's a T-1 that's being used or a pure fiber link. The type of expertise may vary but what reason is there to believe that the amount of labor required to put up a link is directly proportional to the amount of bandwidth represented by that link? The man-hour cost of the labor certainly isn't proportional to the bandwidth required.
My understanding (misinformed as it may be) is that a very large portion of the costs of bandwidth are related to the construction of the links themselves. That those costs are so high that most players can't even get into the game because of them, which is why large, monopolistic companies who already own a great deal of telecommunications infrastructure are really the only guys left. If that's the case, then there's plenty of bandwidth left to be taken advantage of, because there's a lot of dark fiber that remains to be lit up.
In the meantime, like I said, if a network provider is having bandwidth problems, it probably means that the provider is oversubscribed, and that's his problem. He can take advantage of that situation by raising his prices to all his customers, and I think this is exactly what we're seeing.
Making customers pay for some amount of bandwidth usage over some fixed amount is certainly one way to raise the price, but don't make the mistake of believing that the cost of bandwidth really is proportional to the amount of bandwidth used: it isn't, and any such proportional price structure is strictly artificial.
Look, a lot of you guys seem to be saying "well, bandwidth costs so people who use more should pay more".
But bandwidth isn't the same as other things at all.
For instance, it makes sense to pay more for power if you use more. The reason is that the power you use ultimately translates to fuel expended. Fuel costs money, so the more fuel you use, the more you have to pay to offset the costs.
But bandwidth? It's not the same at all. Let's look at the costs:
Running lines. This is a fixed cost. It's why there's a lot of "dark fiber" out there right now: if you're going to take the time to run a line, you may as well run a lot of it. Most of the expense is in the labor to run the line, and that's a one-time cost. Yes, there's maintenance as well, but that doesn't change based on the amount of bandwidth the lines represent, either.
Routers. Fixed expense. Yes, the more capable equipment costs more, but let's face it: routers are subject to Moore's law just as all other computing equipment is. So routers should be getting cheaper per unit bandwidth over time, right? In any case, routing equipment probably doesn't even come close to dominating the expense side of the equation.
Labor. This varies, but not by bandwidth usage. Rather, it varies based on the number of subscribers. The more customers you have, the more labor you have to expend in order to service them. This is in the form of technical support, billing, and maintenance.
Property leases. This, too, is independent of bandwidth.
Electricity and other consumable items. This may vary by bandwidth a little, but not much. It probably takes less electricity to run a fiber connection than it does to run a T-1 of the same length.
I don't think I missed anything important, but if I did, please let me know.
So what's the point? Simple: bandwidth itself isn't what costs money. What costs money is the labor and equipment used to provide that bandwidth.
And that is why it doesn't, in general, make sense to charge more for people who use more bandwidth: those people aren't costing the provider any more money at all or, if they are, it's only because the provider was stupid enough to sign peering agreements in which they pay for the bandwidth they use instead of a flat fee. Instead, if the ISP is undercharging for their services (i.e., can't pay the bills based on the money they get from their subscribers), they should either cut their costs or raise their prices. But before doing either one, they'd better have a good handle on where they're spending their money first.
It's only if a few select subscribers are causing quality of service issues that are, in turn, substantially raising the amount of labor required to keep the operation going that charging those subscribers more may make sense. But I would argue that, in that case, those subscribers are either abusing the service (true only if they're using a substantial amount of bandwidth to initiate DOS attacks against others) and therefore should have their service terminated, or (more likely) that the service itself is oversubscribed. The latter isn't the customers' problem, it's the provider's problem, and charging based on bandwidth used is an entirely inappropriate response, in my opinion.
Back when RedHat was a privately owned company, I'd have had little trouble believing that any software patents they acquired would be used for White Hat purposes only.
But RedHat is a publicly-traded company now, so it has the same "duty to the shareholders" that every other large corporation in the U.S. has. Hence, I have a lot less faith in the company regarding the use of stuff like software patents only for Good. In short, I won't be surprised at all if they use these patents to smack down other commercial Linux distributions, all for Profit at Any Cost.
Only if a Good Guy retains a controlling share of the company would that not apply.
Patents and copyrights
on
Fair IP Laws?
·
· Score: 2
My thoughts on how patents should be:
The length of time for the patent might be fine as it is. But only if the following rules are applied:
A patent application must be made on something completely unique. It must not "extend" an existing technique.
When you submit a patent application, you get something from the patent office proving when you filed it.
A patent application gets held for 5 years by the patent office. The patent will not be granted during that period of time.
While the patent is being held, any other patent applications or other material that show up that are essentially the same as the process being patented will cause all equivalent patent applications to be rejected as being "too obvious".
Only if the patent office receives no other patent applications for the "same" technique during the 5-year period of time will it issue the patent.
The use of the method being patented during or prior to the 5 year holding period is an absolute defense against patent infringement suits. Anyone who sues for patent infringement and loses must pay not only the defendant's entire legal costs but must also pay damages at least equal to those legal costs.
So: you'd better be damned sure that your patent is solid gold if you want to sue someone for patent infringement.
My thoughts on how copyrights should be:
They're non-transferable: only the original author of a work gets to hold the copyright. The copyright gives him the ability to license the work to others for distribution purposes, but only the originator of the work may decide who gets to redistribute it. The originator may change his terms at any time.
The copyright must be renewed each year, with the fee doubling every year (others thought of this but I think it's a great idea, so I mention it here).
The copyright expires the first year that the registration fee fails to be paid. No exceptions.
Since people, not corporations, are the originators of works, people, not corporations, get to hold the copyrights. The copyright to a work created by many people is held simultaneously by those people. But regardless of how much or little was contributed by an individual, each individual may separately negotiate distribution rights. End result: a publisher has to keep all the copyright owners of a work happy if it wants to be able to continue to distribute the work exclusively. Copyright is a contract between the people and creators of works, and it's not the people's problem if the creators of a work can't negotiate with each other to decide who should redistribute the work in question.
Copyright allows a holder to prevent one thing and one thing only: mass distribution of a work. This includes current P2P methods, of course, since that is mass distribution whether you like to admit it or not. But copyright infringement is limited only to those doing the actual distributing. There would be no such thing as "contributory" infringement as there is now, since I regard that concept as being complete nonsense. So P2P network operators themselves would be in the clear, but the actual end nodes would not be.
The copyright holder is allowed copyright only on condition that he releases his work to the copyright office in an unencumbered format -- no access controls allowed -- and that format must be the highest quality form the author and any publisher intends to use. Where there is source material involved, the author must release that to the copyright office as well. The copyright office will release the work and all related materials to the public once the copyright expires (i.e., when the copyright fee stops being paid)
Neither the copyright holder nor any of his publishers are allowed to use any means to prevent the fair use of a work. If a means of copy prevention would in any way prevent or hamper any type of fair use of a work ("hamper" includes quality degradation), then that means is forbidden. Failure to follow this rule will cause immediate expiration of the copyright.
Derivative works are allowed, but the creator of a derivative work is not entitled to copyright, and must be able to prove that he attempted (and failed) to negotiate in good faith with the originator to create and release the derivative work. Because the circumstances can vary so greatly, the courts will be left decide whether or not "good faith" was maintained.
End result: there will be a strong separation between creators of works and publishers of works, and creators will be able to reap the full benefits of creation, as long as they're not assholes about allowing derivative works.
When are we finally going to see reasonably high-speed, high-capacity, semi-permanent, random-access, directly-writable media?
The closest thing I've ever seen is magneto-optical, and that hasn't taken off at all.
Instead, what we have is random-read, serial-write media like CD-R that requires that you build a filesystem image in memory or on a hard disk before you write it out to CD (or DVD).
But what I'd really like to see is removable media with the same read/write characteristics as a hard disk (so that I can create and use a filesystem on it) but which is much more "permanent" (like CD-R).
So why are we currently only getting one but not the other? Why can't we get both in the same package?
Remember, people, the same people who write their password on a piece of paper and stick it on their monitor are the ones who form the majority of Yahoo's user base. They're relatively unsophisticated and, more importantly, simply don't consider their contact information, purchasing preferences, etc., to be terribly private information.
The reason people don't care about this is that they behave as they always have: they don't care about something until it affects them directly. This is why we're slowly (or perhaps not so slowly) but surely losing our rights over time: the majority of people aren't affected by the loss. For instance, they don't have anything terribly earth-shattering to say, so restrictions on speech aren't an issue to them.
Should people care about their loss of privacy? Absolutely. The reason is primarily because of the ultimately evil things that can be done with that loss, namely the police state tactics that can be used to control the population once the data is gathered and accessible to those in power. But very few people think along those lines. Most people are sheep, and will therefore act like sheep.
Which is exactly why oppressive police states happen to begin with.
$ gdb diablo (gdb) break compute_defense if strcmp(defense_item.name, "Ironic Iron Plate Mail of the Damned") == 0 Breakpoint 1 at 0x80483f6: file defense.c, line 36. (gdb) run Starting program: diablo
Breakpoint 1, compute_defense (attack="DMCA", defense_item=0xbffff854) at defense.c:36 36 if (strcmp(attack, defense_item.bonus_defense_attack) == 0) { (gdb) print defense_item.bonus_defense_attack $1 = 0xbffff99b "DCMA" (gdb) n 44 debug_printf("Sorry, no defense bonus against %s\n", attack);
Hmm...sometimes, it pays to know that the Digital Millenium Copyright Act is abbreviated DMCA and not DCMA.:-)
You know, none of this would even be an issue if people took the appropriate amount of responsibility for what they did and held others to the same standard and no higher.
What's "appropriate"? Simple: if something you did causes harm to someone else, then you should do your best, within reason, to remedy the situation. What's "within reason"? Simple: if it's within your power and it's not going to break you (cause "undue hardship"), it's "within reason" (and note that this should not be considered in isolation, but in context of all the other liability cases that may arise from the harmful act. So it should not be possible to kill a company or to destroy an individual through a multitude of liability lawsuits).
The problem is that here in the U.S., you can be held liable even if you make a best effort to remedy the problem. So, for instance, if a bug in your software causes problems for someone else, then fixing the bug in a reasonable amount of time (in other words, a short enough period of time that the bug has no significant additional effect on the victim after you've been notified of it) and giving the bug fix to the victim should be considered sufficient effort in many cases. But the way liability cases seem to go here in the U.S., it wouldn't be nearly enough.
That's because here in the United States, it seems that the jurors often take the stance that the "victim" has no responsibility whatsoever for what happens. For instance, it doesn't seem to matter whether or not the victim researched the alternatives, spoke with others about their experiences with the product, read the manual, etc. -- the victim is considered blameless regardless. And to make matters worse, in the U.S. there's this idea called "joint and severable liability" which, in essence, seems to mean that even if you're responsible for only 2% of the damage, you can be made to pay for 100% of it.
Now, in Microsoft's case, it's often that they do not make a best reasonable effort to fix the bugs in their software and, when they do, they often charge extra for them (a.k.a., software upgrades). Microsoft is by no means the only company that does this (in fact, many software companies do the same thing), and it's only Microsoft's immense market penetration that makes them notable here.
I could go on for some time, but the bottom line is that liability in this country is so screwed up that I'm not convinced that it's possible to write a reasonable law, except perhaps for one that completely dispenses with the notion of "joint and severable" liability, and perhaps one which forbids suit against someone who has already faced a lawsuit on the same liability issue.
Slashdot Mirror
It seems to me that some (perhaps many) are underestimating the importance of WalMart selling PCs with Linux preinstalled.
I believe this is possibly one of the most important in-roads Linux has made in the desktop world since its inception, because:
I think the fact that WalMart is selling PCs with Linux preinstalled is a huge commercial inroad, something few people would have even dreamed of a couple of years ago. Don't be so quick to dismiss it.
Then, of course, you have to account for the price of quality. It is possible to write extremely high quality software: the guys who write software for the space shuttle do so, for instance. But such robust software costs a fortune and (perhaps more importantly) takes a great deal of time to develop.
In reality, we have nobody to blame but ourselves for the poor quality of software (open source software aside, though it's not immune from poor quality). If people in general were willing to pay for greater reliability for software, those who provide better quality would be doing better than those who don't, and other software houses would take notice of this and start to produce better quality software themselves.
Rather the opposite has happened, of course. And it's not just limited to software, either. Look at the goods sold at such places as Wal-Mart: such goods are often of inferior quality to other similar goods, but they often sell better regardless.
When are we finally going to figure out that people want what's barely good enough for the cheapest price possible? Quality only matters when it's low enough that the product in question falls into the "not good enough" category. Warpage of the market by monopolies aside, the market has decided that the quality of software available is good enough. Otherwise, that software wouldn't sell.
Things might be different if people were thoughtful enough and had enough knowledge to determine the total cost of ownership, or to even care about such a thing. But they're not, probably due in large part to the dumbing down of the U.S. population.
The purpose of copyright is to promote progress, to entice authors and inventors to release their works and discoveries to the public.
But that is not an end unto itself. The true end is the benefit to society that the release of such works brings.
Now, remember that the whole incentive here, the entire reason for granting the monopoly privilege of copyright, is to allow the originators of works to make money from their works, which in turn (theoretically) gives them incentives to release their works to the public.
When you publish something on the web, you're publishing your works for free, unless you go to the extra trouble of implementing some kind of access control. The Wayback Machine won't work on a site that has access control, so all it ends up archiving is stuff that was published for free public consumption.
So the real question is: if a work has already been released for free to the general public, how would letting authors restrict the republication of that work after the fact bring greater benefit to society than not letting the author impose such restrictions?
My opinion is that it is much more beneficial to society as a whole if the release of a work for free public consumption automatically implied that members of the public have the right to redistribute that work. So if an author doesn't want people in the general public to be able to redistribute his work, he has to control who receives the work and who doesn't. Certainly requiring payment for the work in question is sufficient to meet the requirement of controlling access. But whatever method the author chooses, it should be one that makes it clear that the work in question is not being released for free to the public.
Well, yes, you actually could, but to do so you'd have to run it through a computer simulation, which is likely to have bugs ... :-)
Seriously, though, engineering isn't the same as programming computers, and I'll tell you why: because every engineering field ultimately has to answer to the laws of physics. Those laws are well-known and well-understood. When you engineer something, you work backwards from the laws of physics. When an engineer designs a bridge, he takes the requirements (the problem specification) and selects a basic design which fits within the problem specifications. The limitations of various designs are well known because the laws of physics dictate those limitations. The design specifies the form of the individual pieces, but the laws of physics determine the necessary characteristics (strength, weight, etc.) of those pieces. Only if the physical characteristics of one or more of those pieces are beyond our ability to manufacture within the cost constraints is the design unworkable, and that is something that can easily be determined ahead of time. If the bridge breaks in use, it's either because the engineer (probably accidentally) ignored the laws of physics or because the engineer didn't understand the problem space well enough.
The laws of physics ultimately dictate everything about a physical object. The individual pieces of the bridge in the example above are designed and constructed in precisely the same way that the bridge is: the smaller parts which comprise a bridge part are all subject to the same laws of physics that the bridge itself is subject to, so their suitability can be analyzed the same way.
And so as a result, you can know with a great deal of certainty whether or not a physical design will break, so it's only when the initial problem specification is wrong, when the laws of physics are not fully recognized and dealt with, or when there are errors made during the actual construction that a manufactured object will unexpectedly fail in use.
Software is very different: there are no laws of physics that govern the functioning of a computer program. In fact, from the program's point of view, the only limitations it has are imposed either by the computer hardware or by the operating system (and the operating system, being a program itself, has no limitations placed on it other than those imposed by the hardware).
While this means that the software designer has a lot more freedom, it also means that the software designer has no consistent way to determine whether or not his design functions properly: it's entirely possible for his design to meet the requirements of the problem at hand but to still malfunction when subject to inputs that are unanticipated.
Which leads to another distinction: the kinds of input to a physical object are limited by the same laws of physics which determine how that object will behave when subject to those inputs. The elements of the bridge are subject to the force of gravity, to the force of the wind, and to the force of the loads put on it by its users, but all of those forces are tightly bound: the force of gravity is a constant, we know that the chance that the force of the winds will exceed some large value is so low that it's sufficient to design the bridge with that value in mind, and the forces that can be placed on it by its users are limited by the density of the materials being pushed across the bridge. And because the forces on the bridge are tightly bound, the forces the individual pieces will undergo are also tightly bound. The laws of physics are unrelenting, so input "bounds checking" is automatic and complete.
But software is completely different in that regard. The software itself is what creates and polices the limits on its input, and because the software itself is responsible for enforcing many of its own limits, it follows that input "bounds checking" is not automatic and complete, but is itself subject to the same possible flaws that the rest of the software is subject to!
As a result of all this, creating a piece of software is more akin to creating an entire universe, complete with a set of governing laws, than to creating a bridge or other engineered object, because not only must the creator define the design, he must also define the rules under which it will operate.
One other thing: because the laws of physics are unchanging, it's possible to construct "standard" objects which can be used to build larger objects. And because the engineering on those component objects has already been done, the engineer can safely rely on their behavior. But the software architect's choices aren't so clear-cut. As I said, software not only must solve the problem it was designed to solve, but must also define the set of rules under which it is to operate. So a "pre-engineered" software component not only defines how it behaves, it defines the rules under which it will operate. Both of these things must be considered when the architect is deciding whether or not to use it, and the latter part can often have a huge impact on the entire project, because the entire project may have to operate under those rules, too. This is why in software it's often easier to just "roll your own" components rather than rely on components which have already been written.
Imagine where we'd be if most electronics shops decided to design and construct their own base components (resistors, capacitors, etc.) rather than use standard off-the-shelf ones. Imagine if most bridge designers had to design and construct the beams, trusses, cables, etc., used to build a bridge, rather than relying on standard components for those things. Things would be a lot more expensive to design and build, and there would probably be a lot more errors made.
Perhaps that's an argument for standardized components in software. Nice as that sounds though, I don't think you're going to see that happening, because the curse software designers operate under is also a blessing: the ability to define your own rules means that anything is possible. In the real world, the laws of physics always act to limit what we can do and how we can do it. If we want the freedom to do anything we want in software, we must accept that the price of that freedom is the greater possibility of error.
If that's how the boss really thinks, then you should definitely stay -- because your boss is exceptional.
Most middle managers understand nothing but the short term bottom line. That means if you cost more to retain than someone else who claims to have roughly the same set of skills but who is/would be paid half of what you are (especially if that other person already works for the company), then you'll be outta there as soon as the boss can arrange it.
Jesus, haven't you people learned anything by now? In the Real World, quality doesn't matter. Ability doesn't matter. The only things that matter are expedience, appearance, and cost. Why do you think there's so much crap in the computer world (and in the rest of the world, for that matter. Why else do you think Wal-Mart is so successful)? If ability were valued more than expedience, appearance, and cost, there'd be a lot less crap that barely works out there.
So if the boss doesn't give a shit about how well the product works, why do you think he'll give a shit about how good you are at what you do?
Yeah, but they'll all end up at the same price point only if their internal costs are identical. How likely do you think that is?
That's why I suspect it is price fixing (or, at least, collusion).
In other words, anyone who would render a judgement against Microsoft is going to be considered "biased" by both Microsoft and the "pro business" crowd you referred to no matter what lengths the person rendering judgement went to in order to be fair.
After all, the whole purpose of the whining is to reduce or eliminate the effects of the judgement, so why would it matter who rendered it?
Yeah. We all know it's perfectly reasonable for Blizzard to sue bnetd.org, because reverse engineering Blizzard's Ultra Secret game protocol so that you don't have to be connected to the internet to play multiplayer Starcraft is so wrong that it should be a crime. I mean, only criminals would reverse engineer a protocol in order to get something they legally own to work the way they think it should.
Right?
Yeah. Because everyone knows that there's no way to make any money by selling games. I mean, if you want to make enough cash to buy a Ferrari, you clearly have to do something else.
And poor Blizzard is so cash strapped that they don't have the money to develop a new game, their only hope is to milk Battle.Net.
Yeah, my heart bleeds for those poor bastards...
Nonsense. They'll be making that claim no matter what.
It depends on how it's done.
As far as I'm concerned, dissolution of Microsoft's business charter should be done by (a) selling off all their assets and (b) taking all the resulting proceeds plus the money in the bank and distributing it equally amongst all working citizens of the United States, except for those who own stock in the company.
The reason for doing it that way? To make it clear to everyone that if your company breaks the law too much, it will die and the people who have invested in it will lose everything they've invested.
Seems to me that it's the only way to keep companies from behaving like Microsoft: make it clear that doing so will, in the long run, be so unprofitable that the people running the show won't even think about it.
I agree, except that if you're going to be setting up a server for people to store their data on, it has to be extremely reliable.
So don't skimp when you're building it. Use a P4-based system (because the P4 will throttle down automatically to prevent itself from overheating, so if the CPU fan dies the system will still run) and build a RAID-5 or mirrored disk subsystem. You might also consider building a standby system to which the data gets mirrored every night, so that you have a fallback in case the first system dies.
An experiment with using laser propulsion on aircraft saw a serious setback today.
The experiment involves using a laser to heat up a droplet of water on the surface of the aircraft. The conversion of the water to a gas is what drives the aircraft.
But today the prototype aircraft was destroyed.
"We fired the laser and POOF, the airplane went up in a puff of smoke!" said the aircraft's chief designer, Takashi Yabe.
Traditionally, lasers have been associated with shooting aircraft down, but this research is attempting to prove that they can be used to propel them as well.
"Unfortunately, today, we managed to prove that lasers really can be used to shoot airplanes down. At least paper ones!" Yabe said.
The researchers are all set to try again. "Whoever decided that we should use paper was obviously a moron", Yabe said, but the other researchers only nodded and Yabe turned red with embarrassment immediately after saying that.
A spokesperson for the U.S. Department of Defense had no comment on today's incident, but word from our sources is that the President will attempt to convince Russia and Europe to start building airplanes out of paper sometime this week.
I believe I said that already. Despite that, the fact remains that when you're being vectored around by ATC is, in general, exactly the most difficult time to keep track of your position relative to terrain.
No, it is for us "bumbling civilians", too. But unlike you military guys, us "bumbling civilians", at least in general aviation, don't have the benefit of having a person on board (the navigator) whose sole purpose in life is to figure out where we are or, in the case of single pilot operations (which is what most of us general aviation civilians engage in), a terrain following radar.
Oh, and lest you get too cocky, remember that it was a military guy flying the B-17 that crashed into the Empire State Building way back in the day.
Uh, not quite.
What you say is true for aircraft operating under VFR (Visual Flight Rules), but not for aircraft operating under IFR (Instrument Flight Rules). When you're operating IFR, if the controller gives you a vector, you follow the vector.
Now, that said, it's still the pilot's responsibility to clear terrain, but without GPS-based terrain avoidance it's damned near impossible to guarantee terrain avoidance when you're IFR, in the soup, and on vectors unless you're intimately familiar with the terrain and your own location.
The bummer of it is that it's when you're relatively low (thus closer to terrain), in relatively busy airspace (you're talking to approach control, who's usually talking to a bunch of different airplanes, and it's a party line so you have to continuously monitor the ATC frequency listening for your callsign), and are having to do a lot of different things (like listen to the ATIS [Automated Terminal Information Service] at the destination airport to get weather, runway info, etc., brief the approach you'll be making into the destination airport, set up the radios and the navigation instruments for the approach, etc., all while listening to the ATC party line frequency for instructions) that it's most likely that you'll be receiving vectors from ATC. That means that if you're in the soup, you have very little time to handle terrain avoidance, which is why GPS-based terrain avoidance gear is such good stuff to have.
Bottom line is that ATC can vector planes into terrain, and it actually has done so occasionally. It doesn't happen often. But it happens. And that's with all the equipment working, at that.
Yeah, it could be a lot better. But the FAA governs aviation with an iron fist, so you're only allowed to use technology that it deems suitable. And the FAA moves at a glacial pace when it comes to allowing new technology in airplanes. It's why modern general aviation airplanes are still using aircooled piston engines that were literally designed back in the 1940's, and why everyone is still using AM radios for communications.
My understanding (misinformed as it may be) is that a very large portion of the costs of bandwidth are related to the construction of the links themselves. That those costs are so high that most players can't even get into the game because of them, which is why large, monopolistic companies who already own a great deal of telecommunications infrastructure are really the only guys left. If that's the case, then there's plenty of bandwidth left to be taken advantage of, because there's a lot of dark fiber that remains to be lit up.
In the meantime, like I said, if a network provider is having bandwidth problems, it probably means that the provider is oversubscribed, and that's his problem. He can take advantage of that situation by raising his prices to all his customers, and I think this is exactly what we're seeing.
Making customers pay for some amount of bandwidth usage over some fixed amount is certainly one way to raise the price, but don't make the mistake of believing that the cost of bandwidth really is proportional to the amount of bandwidth used: it isn't, and any such proportional price structure is strictly artificial.
But bandwidth isn't the same as other things at all.
For instance, it makes sense to pay more for power if you use more. The reason is that the power you use ultimately translates to fuel expended. Fuel costs money, so the more fuel you use, the more you have to pay to offset the costs.
But bandwidth? It's not the same at all. Let's look at the costs:
I don't think I missed anything important, but if I did, please let me know.
So what's the point? Simple: bandwidth itself isn't what costs money. What costs money is the labor and equipment used to provide that bandwidth.
And that is why it doesn't, in general, make sense to charge more for people who use more bandwidth: those people aren't costing the provider any more money at all or, if they are, it's only because the provider was stupid enough to sign peering agreements in which they pay for the bandwidth they use instead of a flat fee. Instead, if the ISP is undercharging for their services (i.e., can't pay the bills based on the money they get from their subscribers), they should either cut their costs or raise their prices. But before doing either one, they'd better have a good handle on where they're spending their money first.
It's only if a few select subscribers are causing quality of service issues that are, in turn, substantially raising the amount of labor required to keep the operation going that charging those subscribers more may make sense. But I would argue that, in that case, those subscribers are either abusing the service (true only if they're using a substantial amount of bandwidth to initiate DOS attacks against others) and therefore should have their service terminated, or (more likely) that the service itself is oversubscribed. The latter isn't the customers' problem, it's the provider's problem, and charging based on bandwidth used is an entirely inappropriate response, in my opinion.
But RedHat is a publicly-traded company now, so it has the same "duty to the shareholders" that every other large corporation in the U.S. has. Hence, I have a lot less faith in the company regarding the use of stuff like software patents only for Good. In short, I won't be surprised at all if they use these patents to smack down other commercial Linux distributions, all for Profit at Any Cost.
Only if a Good Guy retains a controlling share of the company would that not apply.
The length of time for the patent might be fine as it is. But only if the following rules are applied:
So: you'd better be damned sure that your patent is solid gold if you want to sue someone for patent infringement.
My thoughts on how copyrights should be:
End result: there will be a strong separation between creators of works and publishers of works, and creators will be able to reap the full benefits of creation, as long as they're not assholes about allowing derivative works.
The closest thing I've ever seen is magneto-optical, and that hasn't taken off at all.
Instead, what we have is random-read, serial-write media like CD-R that requires that you build a filesystem image in memory or on a hard disk before you write it out to CD (or DVD).
But what I'd really like to see is removable media with the same read/write characteristics as a hard disk (so that I can create and use a filesystem on it) but which is much more "permanent" (like CD-R).
So why are we currently only getting one but not the other? Why can't we get both in the same package?
The reason people don't care about this is that they behave as they always have: they don't care about something until it affects them directly. This is why we're slowly (or perhaps not so slowly) but surely losing our rights over time: the majority of people aren't affected by the loss. For instance, they don't have anything terribly earth-shattering to say, so restrictions on speech aren't an issue to them.
Should people care about their loss of privacy? Absolutely. The reason is primarily because of the ultimately evil things that can be done with that loss, namely the police state tactics that can be used to control the population once the data is gathered and accessible to those in power. But very few people think along those lines. Most people are sheep, and will therefore act like sheep.
Which is exactly why oppressive police states happen to begin with.
Hmm...somehow I don't think that'll work...
Hmm...sometimes, it pays to know that the Digital Millenium Copyright Act is abbreviated DMCA and not DCMA.What's "appropriate"? Simple: if something you did causes harm to someone else, then you should do your best, within reason, to remedy the situation. What's "within reason"? Simple: if it's within your power and it's not going to break you (cause "undue hardship"), it's "within reason" (and note that this should not be considered in isolation, but in context of all the other liability cases that may arise from the harmful act. So it should not be possible to kill a company or to destroy an individual through a multitude of liability lawsuits).
The problem is that here in the U.S., you can be held liable even if you make a best effort to remedy the problem. So, for instance, if a bug in your software causes problems for someone else, then fixing the bug in a reasonable amount of time (in other words, a short enough period of time that the bug has no significant additional effect on the victim after you've been notified of it) and giving the bug fix to the victim should be considered sufficient effort in many cases. But the way liability cases seem to go here in the U.S., it wouldn't be nearly enough.
That's because here in the United States, it seems that the jurors often take the stance that the "victim" has no responsibility whatsoever for what happens. For instance, it doesn't seem to matter whether or not the victim researched the alternatives, spoke with others about their experiences with the product, read the manual, etc. -- the victim is considered blameless regardless. And to make matters worse, in the U.S. there's this idea called "joint and severable liability" which, in essence, seems to mean that even if you're responsible for only 2% of the damage, you can be made to pay for 100% of it.
Now, in Microsoft's case, it's often that they do not make a best reasonable effort to fix the bugs in their software and, when they do, they often charge extra for them (a.k.a., software upgrades). Microsoft is by no means the only company that does this (in fact, many software companies do the same thing), and it's only Microsoft's immense market penetration that makes them notable here.
I could go on for some time, but the bottom line is that liability in this country is so screwed up that I'm not convinced that it's possible to write a reasonable law, except perhaps for one that completely dispenses with the notion of "joint and severable" liability, and perhaps one which forbids suit against someone who has already faced a lawsuit on the same liability issue.