You're right but that's my point; open sourcing, just like on-selling, is dependent on the constituent licenses. To hand wave about "impossible to open source" is equivalent to hand waving about "impossible to on-sell". It's nonsense. Anyway, it's always possible to open source the in-house developed stuff while leaving out the third party controlled stuff.
I was involved in similar discussions about open sourcing GEOS, and did actually work on its codebase and know the people who handled the licensing of 3rd party software.
In an OS, you're going to have lots of code licensed from other companies. Even if you're Microsoft, you just don't have the expertise to write EVERYTHING yourself. Any selling/licensing of the rights of the OS to another company will require that company to negotiate licenses with the holders of the original rights. Serenity Systems certainly had to deal with a lot of license negotiations. If they were lucky, some of IBM's rights were able to be passed on, but I'd doubt that's the case.
As for removing the licensed code, there's two problems. First, you're left with a system that doesn't work. Second, removing the licensed code takes a lot of man hours from engineers who know the project well. If they don't know it well, it's going to take a lot longer. Realistically, they're going to have to be working with lawyers looking over contracts to figure out what to do.
It is EXPENSIVE for IBM to release OS/2. It would probably cost them a few million dollars to do, and the end result would be a product that doesn't build and may or may not have much interest. And don't forget, once this code base is released, very few people are going to know what to do with it. Who knows if anyone that does will care.
No, maintaining their proprietary protocol is a completely unneeded IT expense. Now they have the power of the XMPP community behind them.
You're assuming that the XMPP community's desires correspond with AOL's desires. That's a pretty big assumption. Even if they do correspond, AOL is still going to have deadlines on when they need things implemented, which the community isn't going to care about.
And don't forget, AIM is a mature product. It's not like they need to do massive development on it.
There is a lot with AIM. File transfers don't work reliably behind firewalls. There's no voice and video support. Its client is archaic.
I think you're basing your opinion of AIM off the 3rd party clients. It's been years since I used the official client, but, back then if a file transfer failed, it asked if I wanted to do the transfer through their server instead of direct. It even gave a simple, clear explanation of the pros and cons of doing that.
Voice chat has been in there for almost a decade - I remember people in my dorm using it in '99. I vaguely remember GAIM Release Notes talking about not implementing video support, which I took to mean AIM had video support, but it's not something I've ever cared about so I don't know for sure.
In short, their legacy of being a dial-up information provider instead of an Internet Service Provider was weighing them down.
The difference between a well written game and a poorly written game is so obvious as to not even need pointing out. A poorly written game with beautiful graphics, is still likely to be a flop, while a graphically mediocre game with excellent writing can be a huge success.
A game with great gameplay and crap writing will be a far bigger success than a game with crap gameplay and great writing. Sure, if you're making an RPG, you're mostly relying on the storyline to drive your game. But for the vast majority of games, the writing just needs to be passable.
So it comes down to measuring the contribution to the final product. If the writing doubled the value of the project, the writer should get a larger share than if the writing was merely adequate.
Without the writing, you still have a game. For most games, it won't suffer much.
Remove the sound, you've still got a game, but it'll suffer rather noticeably.
Remove the graphics, programming, or gameplay design and you have no game.
I agree with your logic, and at the small scale we're looking at so far for the primaries (individual districts in very small states), it probably is demographic issues.
However, in the 2004 presidential election, we saw much larger discrepancies between the exit polls and the actual results than we've seen in the past. Why?
I never got that into QFG2. I wasn't fond of the typing interface, and navigating the city streets in that game was really frustrating.
QFG4 though... the voice acting in that game was amazing. You missed out on a LOT by playing without it. Excellent acting all around. The actors for the guys in the bar frequently broke off from the script with stuff a lot funnier than what was written.
Still love "The cute, innocent (well, maybe not so innocent) bunny was viciously slaughtered and now looks like road kill."
It never was scheduled for early or fall 2007. When the game was first announced a year before Wii's launch, it was announced as a launch title. Nothing was said about a release for the game after that until it was given a December '07 release.
I'm not sure if even Nintendo took the idea of having it for the Wii launch seriously, as when they said that, they hadn't yet actually talked to the producer of the previous games about making a new one or formed a dev team.
Second, require a high deductible on any insurance that a consumer has.
The problem with high deductibles is it encourages people to wait for treatment until things get really bad. Treatments tend to get more expensive and become less effective the longer your wait to do something.
You really don't want to encourage people to delay treatment on things that turn out to be cancer.
For the rest of your health care points, I don't see how that would help much. Sure, on things like joint replacements I could see the benefit of letting people research and compare. But for most things of significance, without a medical degree you're not going to know enough to do much of an evaluation. And in any sort of emergency situation, you're not going to have time to research and compare things.
Last time I looked (a few years ago), Consumer Reports reliability ratings were focused on the first few years of owning the car. Essentially, the warranty period. I'm much more interested in the reliability after the warranty period ends.
Using memory after releasing it is bad coding, however you spin it. It's also bad coding even if "most game developers" would do it. It's pretty much guaranteed that even the DOS version crashed because of this: it just didn't do it everytime. So the "works for six years" excuse is probably bollocks (although if you did QA for that version too, you probably know better:)).
I can tell you exactly what the error is, as I ran into the same bug in a project I worked on. It's easy to say it's a bug, but there really wasn't much wrong with doing things that way at the time.
When you used expanded memory on DOS, you allocated memory in 16 KB pages. There was a 64 KB buffer somewhere under 640 KB that the pages were mapped into. Programs would allocate pages, and ask the memory manager to swap the appropriate pages into the buffer. In a pure DOS environment (especially in the old days when expanded memory was found in addon cards rather than emulated), when you freed expanded memory, you were just telling the memory manager to put that page back into the list of free pages. If that memory happened to be in the page frame at the time, it was supposed to stay there until you requested otherwise.
Win9x changed this behavior - if you released an expanded memory page, not only did it free the handle, but it also immediately unmapped the page from the frame buffer. Is that behavior valid? I don't know. But it is different from every prior implementation of expanded memory.
In other words, if you don't see the sanctioned logo on front or back of the case on the actual paper inserts, odds are you have a DRM-laden disc.
Even before they started putting DRM on CDs, a large percentage of them didn't have the CD logo on the outside of the packaging. Basically, anything with elaborate cover art most likely doesn't have the CD logo on it.
You're right, there may be an optical drive bay. But if you have to install an internal job (especially on a case as tight to work in as a Shuttle case), it's not aimed at users looking for a simple computer.
Adding an external USB drive should be possible, but they tend to be really overpriced, negating the advantages of the system.
I think it's intended for the Linux geek looking for a small home server to stick in a corner.
Most slab allocators work by prepending a small header to the start of each malloc()'d block of memory that specifies the size of the block, so that free() will know how much it is expected to free. The implementation could use that to determine the size of any block of memory given by a pointer.
You can't rely on that. What if the string you received is contained within part of a larger data structure? That information you think is malloc data will really be random data used by the program.
No computer scientist should ever be worried about making something run 100% of the time, writing a "mission critical" program, or any of the other buzzwords. That's what software engineers do, and they're trained for it.
The vast majority of the business world doesn't make that distinction. If you want to do something with your degree, you're almost guaranteed to be handling both of those roles.
Another is JRPGs. Even Americans play these, believe it or not (duh, PS2 RPGs sell brilliantly here after all),
Actually, they don't sell very well here. The top handful of them do, but once you get past that, sales aren't very good. There is a reason the vast majority of them never see life outside of Japan.
Game developers and industry pushed the new consoles, consumers don't control these companies, they sped up the development and pushed the graphics angle faster and faster, the drove up their *own* dev costs. Gamers have absolutely no control over these companies, and the decision making.
You do realize that the Xbox 360 and Wii came out when they did because their predecessors stopped selling, right? Consumers said they had enough of the Xbox and GameCube, hence the need for the newer systems. The PS2 was the only console of the previous generation still selling. And it's still selling better than the PS3 is. So far, consumers have said they weren't ready for the PS3.
They actually remodeled the PS2 recently. In late November they announced a revision of the slim version that includes the power supply inside the unit.
I'd take that as a sign that they're still making them.
From what I gather, it's really a problem with the third party device manufacturers - they just don't implement the standards properly. The problem isn't with the cards themselves - cable companies have been mandated to use them in all cable boxes produced since July of last year. These boxes work. So the problem is obviously with the third party devices.
You get a box and CableCard from the same source and they work together fine. Get the two products from different sources and have issues. I'd take that to mean the problem could be on either end.
If you take a CableVision CableCard and stick it into a Comcast cable box, does everything work? How about the reverse? If you throw a random other box in the mix, what combos work?
My understanding has always been that the cable companies never wanted CableCards (or anything similar) and have been doing the minimum necessary to get the FCC off their backs.
The cost factor of Blu-Ray vs HD-DVD wasn't really on the consumer end. It's on the manufacturing end. Blu-Ray discs can't be made with the existing DVD making equipment. You need to buy all new equipment and replace your manufacturing plants. HD-DVD discs just require a relatively inexpensive upgrade to the DVD equipment.
Some of the movie studios didn't want to make the investment for Blu-Ray, hence the creation of HD-DVD.
Sony wins this war by getting enough Blu-Ray manufacturing plants built before HD-DVD gets a large userbase. HD-DVD wins if it can build a large userbase quickly.
The DS screens are small, but there are two of them. The bottom 1/4 - 1/3 of the screen in Starcraft was mainly status information, which can be handled on the 2nd screen. This port seems to be using popup overlays to handle the parts of it that you need to interact with.
If they scaled the art by 50% in each dimension, the DS screen would show approximately the same area as the original game did. I only briefly played Starcraft, and at that back when it was fairly new, so I don't know if the original art is large enough to make that a feasible option.
Whether you have 5 physical machines, or 1 physical with 5 virtuals, you still have 5 servers providing services, and you should have the appropriate number of staff members for that many servers.
With the physical machines, you have to worry about 5 sets of software failing and 5 sets of hardware failing. On the virtual setup, it's 5 sets of software and 1 set of hardware.
If you really only have 5 servers, it probably doesn't make a difference in the amount of staff you need. If you go from 500 servers to 100, you probably need less people than you did before.
That's the heart of the problem. Congress is authorized only to secure copyrights to creators ("Authors and Inventors") - not to employers, assignees, or heirs.
Corporations are essentially considered people as far as the law is concerned.
If you create something as an employee of a company, it's considered the company's work, not yours.
You'd have to essentially get rid of corporations to do what you want.
Slashdot Mirror
You're right but that's my point; open sourcing, just like on-selling, is dependent on the constituent licenses. To hand wave about "impossible to open source" is equivalent to hand waving about "impossible to on-sell". It's nonsense. Anyway, it's always possible to open source the in-house developed stuff while leaving out the third party controlled stuff.
I was involved in similar discussions about open sourcing GEOS, and did actually work on its codebase and know the people who handled the licensing of 3rd party software.
In an OS, you're going to have lots of code licensed from other companies. Even if you're Microsoft, you just don't have the expertise to write EVERYTHING yourself. Any selling/licensing of the rights of the OS to another company will require that company to negotiate licenses with the holders of the original rights. Serenity Systems certainly had to deal with a lot of license negotiations. If they were lucky, some of IBM's rights were able to be passed on, but I'd doubt that's the case.
As for removing the licensed code, there's two problems. First, you're left with a system that doesn't work. Second, removing the licensed code takes a lot of man hours from engineers who know the project well. If they don't know it well, it's going to take a lot longer. Realistically, they're going to have to be working with lawyers looking over contracts to figure out what to do.
It is EXPENSIVE for IBM to release OS/2. It would probably cost them a few million dollars to do, and the end result would be a product that doesn't build and may or may not have much interest. And don't forget, once this code base is released, very few people are going to know what to do with it. Who knows if anyone that does will care.
No, maintaining their proprietary protocol is a completely unneeded IT expense. Now they have the power of the XMPP community behind them.
You're assuming that the XMPP community's desires correspond with AOL's desires. That's a pretty big assumption. Even if they do correspond, AOL is still going to have deadlines on when they need things implemented, which the community isn't going to care about.
And don't forget, AIM is a mature product. It's not like they need to do massive development on it.
There is a lot with AIM. File transfers don't work reliably behind firewalls. There's no voice and video support. Its client is archaic.
I think you're basing your opinion of AIM off the 3rd party clients. It's been years since I used the official client, but, back then if a file transfer failed, it asked if I wanted to do the transfer through their server instead of direct. It even gave a simple, clear explanation of the pros and cons of doing that.
Voice chat has been in there for almost a decade - I remember people in my dorm using it in '99. I vaguely remember GAIM Release Notes talking about not implementing video support, which I took to mean AIM had video support, but it's not something I've ever cared about so I don't know for sure.
In short, their legacy of being a dial-up information provider instead of an Internet Service Provider was weighing them down.
No idea where that is coming from.
Why does the cable charge more for digital cable than analog if they want analog cable to go away?
Why do you still get charged for tone dialing on your land line?
Why do musicians get a percent taken out for breakage on downloaded music sales?
Once a fee is added to a bill, it's pretty rare to ever see it removed.
The difference between a well written game and a poorly written game is so obvious as to not even need pointing out. A poorly written game with beautiful graphics, is still likely to be a flop, while a graphically mediocre game with excellent writing can be a huge success.
A game with great gameplay and crap writing will be a far bigger success than a game with crap gameplay and great writing. Sure, if you're making an RPG, you're mostly relying on the storyline to drive your game. But for the vast majority of games, the writing just needs to be passable.
So it comes down to measuring the contribution to the final product. If the writing doubled the value of the project, the writer should get a larger share than if the writing was merely adequate.
Without the writing, you still have a game. For most games, it won't suffer much.
Remove the sound, you've still got a game, but it'll suffer rather noticeably.
Remove the graphics, programming, or gameplay design and you have no game.
I agree with your logic, and at the small scale we're looking at so far for the primaries (individual districts in very small states), it probably is demographic issues.
However, in the 2004 presidential election, we saw much larger discrepancies between the exit polls and the actual results than we've seen in the past. Why?
I never got that into QFG2. I wasn't fond of the typing interface, and navigating the city streets in that game was really frustrating.
QFG4 though... the voice acting in that game was amazing. You missed out on a LOT by playing without it. Excellent acting all around. The actors for the guys in the bar frequently broke off from the script with stuff a lot funnier than what was written.
Still love "The cute, innocent (well, maybe not so innocent) bunny was viciously slaughtered and now looks like road kill."
It never was scheduled for early or fall 2007. When the game was first announced a year before Wii's launch, it was announced as a launch title. Nothing was said about a release for the game after that until it was given a December '07 release.
I'm not sure if even Nintendo took the idea of having it for the Wii launch seriously, as when they said that, they hadn't yet actually talked to the producer of the previous games about making a new one or formed a dev team.
You're kind of forgetting that the game will most likely have all the characters of previous games
The comments for Lucas on the offical site have said that he's in the game instead of Ness, so that's one old character confirmed gone.
I've also gotten the impression that Ike is replacing Roy & Marth, although I don't think they've said anything concrete on that.
Once they start removing characters, I wouldn't be surprised to see others removed. I'd put clones, Mewtwo, and Game & Watch as the most likely to go.
Second, require a high deductible on any insurance that a consumer has.
The problem with high deductibles is it encourages people to wait for treatment until things get really bad. Treatments tend to get more expensive and become less effective the longer your wait to do something.
You really don't want to encourage people to delay treatment on things that turn out to be cancer.
For the rest of your health care points, I don't see how that would help much. Sure, on things like joint replacements I could see the benefit of letting people research and compare. But for most things of significance, without a medical degree you're not going to know enough to do much of an evaluation. And in any sort of emergency situation, you're not going to have time to research and compare things.
Last time I looked (a few years ago), Consumer Reports reliability ratings were focused on the first few years of owning the car. Essentially, the warranty period. I'm much more interested in the reliability after the warranty period ends.
Using memory after releasing it is bad coding, however you spin it. It's also bad coding even if "most game developers" would do it. It's pretty much guaranteed that even the DOS version crashed because of this: it just didn't do it everytime. So the "works for six years" excuse is probably bollocks (although if you did QA for that version too, you probably know better :)).
I can tell you exactly what the error is, as I ran into the same bug in a project I worked on. It's easy to say it's a bug, but there really wasn't much wrong with doing things that way at the time.
When you used expanded memory on DOS, you allocated memory in 16 KB pages. There was a 64 KB buffer somewhere under 640 KB that the pages were mapped into. Programs would allocate pages, and ask the memory manager to swap the appropriate pages into the buffer. In a pure DOS environment (especially in the old days when expanded memory was found in addon cards rather than emulated), when you freed expanded memory, you were just telling the memory manager to put that page back into the list of free pages. If that memory happened to be in the page frame at the time, it was supposed to stay there until you requested otherwise.
Win9x changed this behavior - if you released an expanded memory page, not only did it free the handle, but it also immediately unmapped the page from the frame buffer. Is that behavior valid? I don't know. But it is different from every prior implementation of expanded memory.
Isn't it the RIAA who actually represents the labels in court? That's the impression I've gotten.
I believe the RIAA represents them collectively in major things, while on the little things its the individual members.
Examples:
RIAA vs XM / Sirius / MP3.com / Random MP3 Player maker
Sony vs 90 Year Old Woman
Warner vs Mountain Hermit
EMI vs 10 year old girl
In other words, if you don't see the sanctioned logo on front or back of the case on the actual paper inserts, odds are you have a DRM-laden disc.
Even before they started putting DRM on CDs, a large percentage of them didn't have the CD logo on the outside of the packaging. Basically, anything with elaborate cover art most likely doesn't have the CD logo on it.
You're right, there may be an optical drive bay. But if you have to install an internal job (especially on a case as tight to work in as a Shuttle case), it's not aimed at users looking for a simple computer.
Adding an external USB drive should be possible, but they tend to be really overpriced, negating the advantages of the system.
I think it's intended for the Linux geek looking for a small home server to stick in a corner.
Actually, it's not. There's no optical drive bay in the system. So you can't watch a DVD or rip music.
Most slab allocators work by prepending a small header to the start of each malloc()'d block of memory that specifies the size of the block, so that free() will know how much it is expected to free. The implementation could use that to determine the size of any block of memory given by a pointer.
You can't rely on that. What if the string you received is contained within part of a larger data structure? That information you think is malloc data will really be random data used by the program.
No computer scientist should ever be worried about making something run 100% of the time, writing a "mission critical" program, or any of the other buzzwords. That's what software engineers do, and they're trained for it.
The vast majority of the business world doesn't make that distinction. If you want to do something with your degree, you're almost guaranteed to be handling both of those roles.
Another is JRPGs. Even Americans play these, believe it or not (duh, PS2 RPGs sell brilliantly here after all),
Actually, they don't sell very well here. The top handful of them do, but once you get past that, sales aren't very good. There is a reason the vast majority of them never see life outside of Japan.
Game developers and industry pushed the new consoles, consumers don't control these companies, they sped up the development and pushed the graphics angle faster and faster, the drove up their *own* dev costs. Gamers have absolutely no control over these companies, and the decision making.
You do realize that the Xbox 360 and Wii came out when they did because their predecessors stopped selling, right? Consumers said they had enough of the Xbox and GameCube, hence the need for the newer systems. The PS2 was the only console of the previous generation still selling. And it's still selling better than the PS3 is. So far, consumers have said they weren't ready for the PS3.
They actually remodeled the PS2 recently. In late November they announced a revision of the slim version that includes the power supply inside the unit.
I'd take that as a sign that they're still making them.
From what I gather, it's really a problem with the third party device manufacturers - they just don't implement the standards properly. The problem isn't with the cards themselves - cable companies have been mandated to use them in all cable boxes produced since July of last year. These boxes work. So the problem is obviously with the third party devices.
You get a box and CableCard from the same source and they work together fine. Get the two products from different sources and have issues. I'd take that to mean the problem could be on either end.
If you take a CableVision CableCard and stick it into a Comcast cable box, does everything work? How about the reverse? If you throw a random other box in the mix, what combos work?
My understanding has always been that the cable companies never wanted CableCards (or anything similar) and have been doing the minimum necessary to get the FCC off their backs.
The cost factor of Blu-Ray vs HD-DVD wasn't really on the consumer end. It's on the manufacturing end. Blu-Ray discs can't be made with the existing DVD making equipment. You need to buy all new equipment and replace your manufacturing plants. HD-DVD discs just require a relatively inexpensive upgrade to the DVD equipment.
Some of the movie studios didn't want to make the investment for Blu-Ray, hence the creation of HD-DVD.
Sony wins this war by getting enough Blu-Ray manufacturing plants built before HD-DVD gets a large userbase. HD-DVD wins if it can build a large userbase quickly.
The DS screens are small, but there are two of them. The bottom 1/4 - 1/3 of the screen in Starcraft was mainly status information, which can be handled on the 2nd screen. This port seems to be using popup overlays to handle the parts of it that you need to interact with.
If they scaled the art by 50% in each dimension, the DS screen would show approximately the same area as the original game did. I only briefly played Starcraft, and at that back when it was fairly new, so I don't know if the original art is large enough to make that a feasible option.
Whether you have 5 physical machines, or 1 physical with 5 virtuals, you still have 5 servers providing services, and you should have the appropriate number of staff members for that many servers.
With the physical machines, you have to worry about 5 sets of software failing and 5 sets of hardware failing. On the virtual setup, it's 5 sets of software and 1 set of hardware.
If you really only have 5 servers, it probably doesn't make a difference in the amount of staff you need. If you go from 500 servers to 100, you probably need less people than you did before.
That's the heart of the problem. Congress is authorized only to secure copyrights to creators ("Authors and Inventors") - not to employers, assignees, or heirs.
Corporations are essentially considered people as far as the law is concerned.
If you create something as an employee of a company, it's considered the company's work, not yours.
You'd have to essentially get rid of corporations to do what you want.