To me the missing element in these competitions has always been the selection of language. The programs are expected to communicate in English (or maybe some other human language) which is very hard to do. It's unnecessarily hard to do.
What they should do is create a symbolic language that has only a few hundred (or maybe a few thousand elements), thereby constraining the concept space. Then write programs that can "speak" and "understand" that symbolic language. Compare those programs to a human who is only allowed to use the symbols provided, and we'll see if any of the programs can fool the judges.
Bottom line: It's not a fair contest for the computer. Level the playing field!!!
Does anybody know why cell phones never receive telmarketing calls? Are they simply prohibited from calling certain number blocks? If so, that means they have the capability to remove blocks or numbers...
Beauty is, of course, in the eye of the beholder. X-Plane is a really neat simulator, and you're saying that Austin should be forced to give it away for free? I'm sure you LIKE getting stuff for free, but it has to be freely given, not forcibly taken. So I'd say that from the standpoint of a commercial author, it makes quite a lot of sense for software to be NON-free.
Considering the state of the national airsystem right now, simulators may be the best bet for a lot of people out there. The vast majority of private pilots are not permitted to fly at the moment; enhanced class B airspaces are still closed to VFR flight. This may enter a gradual period of reduced constraint over the next few weeks.
I don't think I'm going to dignify this idiocy with much of a response, other than to say:
Software is different You don't need huge investments to create software. The industry was very prosperous on its own for decades before patents were introduced.
The real issue here is whether we should corrupt world-wide standards with an obviously corrupt patent process. The answer is that we should not.
Companies holding patents seek to legitimize those patents by having standards make use of them. I think that a patent on a technique is automatic grounds for rejection of that technique as a basis for a world-wide standard. If a company chooses to patent a technique and does not assign that patent to the W3C, they can forget about it ever becoming a standard, period.
That's a pretty crappy thing to say. Where's your donation, fuckhead? I'd be surprised and shocked if they were doing anything other than sending the exact amount directly to the red cross. Go shove your head up your ass, and leave it there.
It's the best thing I've seen so far in this category...it's a monthly pay service, but is far better put together than most products of its type. It supports phones, PDAs, PCs, and even provides you a web desktop to all that information.
I guess the real question is, why is R&D so important? I heard a few months ago on NPR that the drug companies play extensive games with their accounting to make it LOOK incredibly expensive to do drug R&D. When you think about it, it's absolutely in the best interests of the drug companies to do this. The figures being tossed around were (for some drug I can't remember) a drug company figure of $800 million for R&D, and a contrary figure of $125 million by a watchdog group's accounting.
Other things that add to the cost include the vast cost of litigation in this country, and incredibly destabilizing effects of outlandish juried settlements.
Actually, ActiveX IS a gaping security hole. The company that signs the ActiveX control usually THINKS they know what it does, and then it turns out they're wrong. It's all about points of failure. In Java applets you've got a single point of failure: The Java security manager and sandbox. A piece of code downloaded off the net can do just about anything it wants in that sandbox and it has a very low probability of getting out.
And then there's ActiveX. A company (even Microsoft) can release something and find that there are security holes that they didn't even know that they had, but the black hats of the world are only too happy to keep to themselves. Downloadable code that has unrestricted access to the OS, where there isn't even any security applied at the OS level, is just plain stupid. But that's what we're stuck with.
There is exactly ONE nice thing about ActiveX. They install cleanly and quickly, with only a single button click.
COM blows, because it's got no security. That doesn't mean it couldn't have any; just that it doesn't right now.
Nope. Modern garbage collectors certainly don't walk around memory collecting things (and neither do modern heap allocators). All the tabular stuff that indexes memory and structures is usually contained in a small set of pages. Those refer to blocks elsewhere. No page swapping is required.
Java's class packaging is considerably more than a small step in the right direction. It supports a universal naming convention, based on the internet's naming systems, that can underlie local and remote code. D's modules are a primitive mechanism at best, similar to Delphi's. They're OK for a single organization, but problematic for integrating code on a wider basis.
Dude, thanks! I gotta figure out how to write our own custom classloader for our app:) I always wondered how they got that crap to come in so damn fast. Time to break out the JProbe again, and see what can be done.
Good UI can't be done by just anybody. You really need to throw your best and brightest at it.
So all those thousands of people who've written Delphi components very successfully over the VCL base classes are just smoking crack, and really didn't do that at all?
Sometimes you need to modify the VCL, but it's pretty rare. Every once in a while something you need is buried in a private section. I suggest you educate yourself a little more on the VCL and what it can do for you. There's probably another way to do whatever it is that you're trying to do.
So you feel that design by interface isn't practical? Millions of Java programmers would like to disagree with you. You almost never need MI if you have interfaces.
There's behavioral inheritance and structural inheritance. Structural inheritance just isn't very necessary, and needing it generally indicates that you did something wrong.
What you've learned is that shitty programming can run slow anywhere. A well-written Java UI can fly. Do yourself a favor and download JBuilder from Borland. That's a nicely written Java UI, and it's pretty much the same speed as a native UI.
It's unfair to compare a Java UI directly with a native UI. How well does that native UI run on other platforms? Oh yeah, it can't. How well does it run from a web page? It doesn't.
Properly written Java code can approach the speed of pure C, be done in a tenth the time, and be significantly more maintainable and portable.
Did he really market and sell the product to US citizens? As far as I can tell the product was simply sold on the internet. It wasn't "marketed" to US citizens any more or less than it was marketed to people elsewhere in the world. That a few Americans happened to buy it is circumstantial, at best. Merely using a US payment service shouldn't mean that the product is directed at the US market.
Imagine that you've gotten lucky and you've photographed and created an e-book of swimsuit models. You sell your e-book online, and somebody from Iran (amidst 100 sales to people from elsewhere in the world) buys your book with a Visa card; they download the e-book immediately. Five months later you're flying to Moscow but your flight stops in Iran. You're arrested.
The nature of the internet is what's really in question here -- does making your product available for sale on the internet constitute active participation in the rules and structure of every country the internet reaches?
These undoubtedly contrast!!!
You know, I thought the same thing initially. But here's an interesting way to think about it: You didn't write a commercial perl interpreter that calls perl to do its interpretation. You aren't writing a gzip program that calls gzip. You are creating substantial, differing functionality; the GPL expressly allows you to call them when you're doing that. I should think that it applies for shells, dynamic, or static linking.
It seems to me that Vidomi has released a video editor that calls VirtualDub, a GPL video editor, to do its work.
It shouldn't stand; it looks like a fairly clear violation to me.
You got that right. It's frickin' hard to make an app (or a cooperating network of apps) that can survive the laundry list of possible failures. I've seen deployment environments that are so far away from optimal that it's just scary. You gotta have an X-files attitude about distributed software: Trust no-one. Trust no process, no socket, no network. If it could go wrong, it will.
There's another great leap in computation waiting out there for us -- peer to peer, distributed, fault-tolerant, scalable, indexed, transactional information storage. Don't say database, because it needs to be able to store any kind of data, at any time. This thing is to information what the internet is to packets. Route around the damage, converge on optimality. It's a very hard problem. Whoever solves it first, best, is gonna be extra rich.
The essential idea is this: Was the US plane in international flight space? Probably. But why was it so close? The article points out a basic difference in Chinese viewpoints vs. American viewpoints. Americans feel that it's OK to do anything you want, as long as you don't "cross the line". The line in this case is the law. Americans view the law as being a tightly defined concept. Chinese culture, on the other hand, maintains that you shouldn't even approach the line; that you shouldn't give the appearance of an offense.
This was enlightening to me. The Chinese are not really seeking an apology for the actual collision. What they can't understand is why we are running surveillance aircraft so close to their airspace? Even if it is not technically against international law, why do we insist on provoking them?
All the posturing on "fault" for the accident is a smokescreen for this basic cultural difference. Are there solid military reasons for doing the surveillance? Yes. But we must balance that against the fact that even if we're not breaking the law, we are definitely too close for Chinese comfort.
Uh...they don't need your caller id. They already know exactly who you are b/c your Tivo needs to log in to the remote network using (probably) the serial number of your Tivo...otherwise they'd have no way of knowing whether you're a "real" Tivo or not. Each Tivo is registered to a person, who needs to give over the $10 each month it costs for the service.
How the heck can Tivo NOT be making money? They get my $10 each month. Isn't that enough? How many subscribers do they have anyway?
In an ideal world, all pixel computation would take place in floating point, period. You'd choose color that way. 24/32 bit is entirely inadequate for sophisticated image processing. It's not about what gets on the display; it's about the computations that got it there.
There is no innovation here whatsoever; it is merely the application of existing techniques to a larger scale problem. So what if they have a large number of potential versions and targets. The problem has been solved before (game program updates, for example, use precisely this technique).
You think they should be able to get a patent on doing more? That's like saying that every patented process has another patent waiting when somebody uses it on twice as many objects!
Game companies have been doing patches like this forever, and certainly prior to 1998. A game release schedule will often have an initial 1.0, followed by a 1.01, 1.02, 1.03, etc. Sooner or later they'll do a 1.1. They'll often make patches for 1.01 to 1.02, but they'll make a single patch that goes from 1.0 to 1.1, for example. A prior art search should start there!
The thing that bugs me is that there is advertising email that I want to get. I want advertising that tells me about products that I'm interested in, and new things that are happening. I don't want anything else. Since I can't block out the masses, I just delete everything that remotely looks like advertising.
Wouldn't it be cool if we could all have open source tivo-like boxes. Tivo's good (and it's already based on Linux), but anything can be improved. The main question would be the necessity of having hardware-based encoding. Tivo's cheap for what it does. But as processor power increases...it's possible.
We'd need a thousand squirrels typing in television schedules.
Slashdot Mirror
To me the missing element in these competitions has always been the selection of language. The programs are expected to communicate in English (or maybe some other human language) which is very hard to do. It's unnecessarily hard to do.
What they should do is create a symbolic language that has only a few hundred (or maybe a few thousand elements), thereby constraining the concept space. Then write programs that can "speak" and "understand" that symbolic language. Compare those programs to a human who is only allowed to use the symbols provided, and we'll see if any of the programs can fool the judges.
Bottom line: It's not a fair contest for the computer. Level the playing field!!!
Does anybody know why cell phones never receive telmarketing calls? Are they simply prohibited from calling certain number blocks? If so, that means they have the capability to remove blocks or numbers...
Beauty is, of course, in the eye of the beholder. X-Plane is a really neat simulator, and you're saying that Austin should be forced to give it away for free? I'm sure you LIKE getting stuff for free, but it has to be freely given, not forcibly taken. So I'd say that from the standpoint of a commercial author, it makes quite a lot of sense for software to be NON-free.
Considering the state of the national airsystem right now, simulators may be the best bet for a lot of people out there. The vast majority of private pilots are not permitted to fly at the moment; enhanced class B airspaces are still closed to VFR flight. This may enter a gradual period of reduced constraint over the next few weeks.
I don't think I'm going to dignify this idiocy with much of a response, other than to say:
Software is different
You don't need huge investments to create software. The industry was very prosperous on its own for decades before patents were introduced.
The real issue here is whether we should corrupt world-wide standards with an obviously corrupt patent process. The answer is that we should not.
Companies holding patents seek to legitimize those patents by having standards make use of them. I think that a patent on a technique is automatic grounds for rejection of that technique as a basis for a world-wide standard. If a company chooses to patent a technique and does not assign that patent to the W3C, they can forget about it ever becoming a standard, period.
That's a pretty crappy thing to say. Where's your donation, fuckhead? I'd be surprised and shocked if they were doing anything other than sending the exact amount directly to the red cross. Go shove your head up your ass, and leave it there.
It's the best thing I've seen so far in this category...it's a monthly pay service, but is far better put together than most products of its type. It supports phones, PDAs, PCs, and even provides you a web desktop to all that information.
FusionOne
I guess the real question is, why is R&D so important? I heard a few months ago on NPR that the drug companies play extensive games with their accounting to make it LOOK incredibly expensive to do drug R&D. When you think about it, it's absolutely in the best interests of the drug companies to do this. The figures being tossed around were (for some drug I can't remember) a drug company figure of $800 million for R&D, and a contrary figure of $125 million by a watchdog group's accounting.
Other things that add to the cost include the vast cost of litigation in this country, and incredibly destabilizing effects of outlandish juried settlements.
And then there's ActiveX. A company (even Microsoft) can release something and find that there are security holes that they didn't even know that they had, but the black hats of the world are only too happy to keep to themselves. Downloadable code that has unrestricted access to the OS, where there isn't even any security applied at the OS level, is just plain stupid. But that's what we're stuck with.
There is exactly ONE nice thing about ActiveX. They install cleanly and quickly, with only a single button click.
COM blows, because it's got no security. That doesn't mean it couldn't have any; just that it doesn't right now.
Java's class packaging is considerably more than a small step in the right direction. It supports a universal naming convention, based on the internet's naming systems, that can underlie local and remote code. D's modules are a primitive mechanism at best, similar to Delphi's. They're OK for a single organization, but problematic for integrating code on a wider basis.
Good UI can't be done by just anybody. You really need to throw your best and brightest at it.
Sometimes you need to modify the VCL, but it's pretty rare. Every once in a while something you need is buried in a private section. I suggest you educate yourself a little more on the VCL and what it can do for you. There's probably another way to do whatever it is that you're trying to do.
D = Java + structures - (class library+remote code)
There's behavioral inheritance and structural inheritance. Structural inheritance just isn't very necessary, and needing it generally indicates that you did something wrong.
It's unfair to compare a Java UI directly with a native UI. How well does that native UI run on other platforms? Oh yeah, it can't. How well does it run from a web page? It doesn't.
Properly written Java code can approach the speed of pure C, be done in a tenth the time, and be significantly more maintainable and portable.
Did he really market and sell the product to US citizens? As far as I can tell the product was simply sold on the internet. It wasn't "marketed" to US citizens any more or less than it was marketed to people elsewhere in the world. That a few Americans happened to buy it is circumstantial, at best. Merely using a US payment service shouldn't mean that the product is directed at the US market. Imagine that you've gotten lucky and you've photographed and created an e-book of swimsuit models. You sell your e-book online, and somebody from Iran (amidst 100 sales to people from elsewhere in the world) buys your book with a Visa card; they download the e-book immediately. Five months later you're flying to Moscow but your flight stops in Iran. You're arrested. The nature of the internet is what's really in question here -- does making your product available for sale on the internet constitute active participation in the rules and structure of every country the internet reaches? These undoubtedly contrast!!!
It seems to me that Vidomi has released a video editor that calls VirtualDub, a GPL video editor, to do its work.
It shouldn't stand; it looks like a fairly clear violation to me.
You got that right. It's frickin' hard to make an app (or a cooperating network of apps) that can survive the laundry list of possible failures. I've seen deployment environments that are so far away from optimal that it's just scary. You gotta have an X-files attitude about distributed software: Trust no-one. Trust no process, no socket, no network. If it could go wrong, it will. There's another great leap in computation waiting out there for us -- peer to peer, distributed, fault-tolerant, scalable, indexed, transactional information storage. Don't say database, because it needs to be able to store any kind of data, at any time. This thing is to information what the internet is to packets. Route around the damage, converge on optimality. It's a very hard problem. Whoever solves it first, best, is gonna be extra rich.
The essential idea is this: Was the US plane in international flight space? Probably. But why was it so close? The article points out a basic difference in Chinese viewpoints vs. American viewpoints. Americans feel that it's OK to do anything you want, as long as you don't "cross the line". The line in this case is the law. Americans view the law as being a tightly defined concept. Chinese culture, on the other hand, maintains that you shouldn't even approach the line; that you shouldn't give the appearance of an offense.
This was enlightening to me. The Chinese are not really seeking an apology for the actual collision. What they can't understand is why we are running surveillance aircraft so close to their airspace? Even if it is not technically against international law, why do we insist on provoking them?
All the posturing on "fault" for the accident is a smokescreen for this basic cultural difference. Are there solid military reasons for doing the surveillance? Yes. But we must balance that against the fact that even if we're not breaking the law, we are definitely too close for Chinese comfort.
How the heck can Tivo NOT be making money? They get my $10 each month. Isn't that enough? How many subscribers do they have anyway?
In an ideal world, all pixel computation would take place in floating point, period. You'd choose color that way. 24/32 bit is entirely inadequate for sophisticated image processing. It's not about what gets on the display; it's about the computations that got it there.
You think they should be able to get a patent on doing more? That's like saying that every patented process has another patent waiting when somebody uses it on twice as many objects!
Game companies have been doing patches like this forever, and certainly prior to 1998. A game release schedule will often have an initial 1.0, followed by a 1.01, 1.02, 1.03, etc. Sooner or later they'll do a 1.1. They'll often make patches for 1.01 to 1.02, but they'll make a single patch that goes from 1.0 to 1.1, for example. A prior art search should start there!
The thing that bugs me is that there is advertising email that I want to get. I want advertising that tells me about products that I'm interested in, and new things that are happening. I don't want anything else. Since I can't block out the masses, I just delete everything that remotely looks like advertising.
We'd need a thousand squirrels typing in television schedules.
The whiteboard edition is all you need to do serious work. The pay versions are even better.