We will never be rid of stereotype unless we evolve away the instinct. We can just try to reduce the effect through social pressure.
Absolutely. Stereotyping is simply a form of inductive reasoning. The rabbit cannot afford to think, "Well, maybe this coyote will be friendly." Refraining from acting upon prejudice is a conscious act.
If you were congenitally missing an eye, and your other eye could see in the dark to compensate, it would be a positive result, but also a genetic defect.
It would be a genetic defect because there is a pathological consequence--loss of stereoscopic vision.
This is where prejudice really lies: in making judgments of individuals based on what you believe you know about the statistical characteristics group. Many hope to reduce prejudice by insisting that real statistical differences do not exist, but this avoids the real problem--it is still prejudice even if the difference is real. People are individuals, not statistics, and if you judge a person on a statistical basis, without troubling to learn what that person is like as an individual, then you are engaging in prejudice.
Moreover, trying too hard to deny such statistical differences can actually encourage prejudice, because it conveys the message, "If this difference were real, then it would be OK to be prejudiced."
The thing is that intelligence is not a single dimensional quantity like height is. We pretend it is by assigning an IQ value to some measurement of it, but even scientists who study it will tell you that a major problem in the field is a lack of understanding of just what intelligence is.
It is better to think of IQ, not as intelligence per se, but rather as a measure that has a substantial correlation with other measures of intelligence.
Suppose that you were trying to measure height, but for some reason were unable to do so directly. You might instead measure the length of the forefinger. Now length of forefinger is not height, and you will find some short people with long fingers and some tall people with short fingers. Still, the correlation is almost certainly good enough that you could learn a lot about the inheritance of height by measuring fingers.
Einstein's brain was actually the product of a genetic defect.
A genetic variation does not necessarily constitute a "defect." Are blue eyes a defect? Generally, the term defect is used when there is some kind of significant pathological consequence.
I'm sure everyone has heard the phrase "I don't want to know". Sometimes I wonder if there are some things that scientists should avoid studying. For example, what if it could be scientifically proven, without a doubt, that race A was in many ways superior to race B?
It is worth noting that people may also be harmed by the decision not to know. Information about the genetic basis of intelligence could lead to improvements in the treatment of genetic disease. Indeed, a suggestion in the paper is that the same genes may play a role in genetic diseases in this population. This could be very important for treatment.
If Apple choose to add a big chunk of markup for 'branding reasons' onto their Intel PCs then that puts them in a rather unique position on the Intel platform where customers are just paying for a nice case and and a piece of firmware which allows OS X to be booted.
This would be a legitimate complaint if Apple planned to sell "bare" Mac compatible computers, but it seems more likely that they will be offering them, as they have in the past, in a package with OSX and a bunch of other Apple software. Besides, even in the Intel world, you pay more for a brand name computer than for a "noname" clone.
Like XPostFacto you mean ?
A bit like XPostFacto, in that XPostFacto is used mainly by people who can't afford to buy an new Mac, and want to extend the lifespan of a older Mac, usually at home. It's not used much in a business environment. There are a variety of little problems, but they may be worth it for somebody who is willing to spend a little time working around them. And while XPostFacto needs to be revised for major system revisions, it typically comes through routine updates fine. After all, Apple doesn't have a reason to intentionally break XPostFacto. A hack to run Mac OSX on Wintel boxes will be viewed by Apple more like the hacks to strip the DRM off of iTunes songs, which break with every iTunes update.
I'm sure that somebody will figure out how to hack OS X to run on a generic Wintel box. It won't affect anything. Nobody will use it except for a handful of hackers, and not for any serious purposes, because it will be too much of a pain to maintain, with the patches breaking every time Apple releases a system update.
I read somewhere that Aqua largely depends on the AltiVec execution unit for the heavy graphical stuff. It's a totally uneducated guess, but I think the Intel based macs - if based on Intels current offering - will have some kind of coprocessor to support 128 bit vectors. This might make it very difficult to run OS-X on non-Apple hardware.
OS X Tiger offers the ability to offload the "heavy graphical stuff" to the graphics card, so this is unlikely to be an issue.
So even after this move, switching from whichever OS you are running now to OSX will entail purchasing Apple hardware.
Yes, I think the ability of Linux to run on generic Intel boxes, combined with the practical philosophical appeal of open source, will keep it lively.
However, if Apple manages to come out with reasonably priced Intel boxes, I'm not sure why anybody would buy anything else. After all, InteliMac(TM) owners will potentially have access to all Mac software and all Windows software. And Linux as well.
Want to bet that someone gets it running on generic hardware within a week of it hitting the shelves?
I'll bet it takes a bit longer than that. There are hacks to get Mac OS X to run on older, unsupported Mac hardware, but it clearly has taken quite a bit of work.
And what do you want to bet that every Apple security update will break the generic hardware patches?
And with an open source kernel, how, exactly, do they plan to stop people from hacking OS X onto commodity PC hardware?
The open source kernel doesn't have much to do with it. After all, the kernel already can be run on Wintel hardware. Still, I imagine that somebody will figure out how to hack it to run on generic Wintel boxes. After all, people have managed to hack OS X to run on older "unsupported" Mac hardware. Of course, Apple will doubtless take a sterner view of hacking OS X to run on Wintel boxes. Apple has dealt with this before, with Apple II clones. They dealt with it by lawsuits (Apple's litigiousness is by no means a recent phenomenon), and by making every OS revision incompatible with clones. I imagine that they will do the same, with every security update also updating the code that checks for "nonauthorized" hardware.
Actually, I found that the 68K to PPC transition went pretty smoothly. Virtually everything still ran. On the first generation of PowerPC's, 68K applications ran perceptibly slower, but not by much--about like having a 68030 instead of a 68040. By the second generation, even the 68K applications were faster than on 68040.
And it is likely that this transition will probably go even more smoothly: Early versions of the PPC MacOS still were running a lot of 68K OS code in emulation; it is a safe bet that the Intel OS X will be 100% native code. And there is less hand-tweaked assembly code running around, so it will be easier for developers to simply recompile. Most major applications are already cross-platform, so developers already know what to tweak to enhance Intel processor performance.
My guess is that the transition will be smoother than the PPC transition, and much smoother than the OS X transition.
Financially, this is going to be a big bump for Apple. I'm certainly not going to order any more new Macs until the Intel systems are available. This may be one reason why they chose to do it now, when the success of the iPod will carry them through.
It may be the best decision for Apple, but I still think that it would have been better if they'd been able to reach a deal with IBM to develop the PPC further. I would much rather have seen multicore PPC's.
The question of whether the Intel OS X will run on generic Intel hardware seems to still be open. I'd guess not, but then I didn't believe that they'd switch to Intel in the first place.
Unless you think your (reclaimed) desk space is worth 1000$/sq ft, or that you think LCD is better in a interior-designer standpoint, in which case I'll grant you it's a better buy for you.
People put a high value on their desk space. Now that flat panels are available, I can't give a 17" CRT away. It kills me to throw away a perfectly good CRT monitor, but nobody wants them.
This isn't over by a long shot. The MPAA took a gamble, based on what they thought they had in Congress, and lost. They won't make the same mistake twice. Look for subtle changes in the "new and improved" DMCA, COPA and its children, and other roundabout ways to implement the same thing.
However, it looks like they are going to miss out on the time window for implementing this particular method of making consumer digital video more inconvenient and unreliable. By the time they get their act together, there will likely be too much broadcast flag-free equipment out there.
Darwinian selection should cause those societies which encourage/permit such uses to become extremely fragile and likely to become infected and wiped out thru... oh, I don't know, HN51 avian bird flu let's say... nah, that would never happen, biology adapting to weed out the less diverse and less resistant groups, would it?
Considering the extent of cross-mating made possible by modern transportation, combined with medical technology that enables genotypes to survive that probably would have been lethal in many times and places, human populations probably have greater genetic diversity than at any time in human history. A small fraction of the population screening out a handful of alleles is not going to make much of a dent.
The parents I've meet with Downs and autistc children have commented on how rewarding it is.
Rewarding != desirable. Hardly anybody with a Downs or autistic child wants to give it up. Parents love their children and raising even a seriously disabled child is a rewarding experience. But offer any prospective parent the rewarding experience or raising a child with autism or Downs syndrome vs. the rewarding experience of raising a child without these conditions, and guess which they'll choose?
Genetic testing will probably cause more harm than good - we need to have it screened for medical uses only, such things as fatal diseases, not What's Hot This Week
And how are you going to do that? The technology is not that complicated. Are you going to prohibit people from owning a PCR machine (which is really just a precisely controlled hotplate)? Or make thermostable polymerase an illicit substance? Pretty soon, anybody will be able to test for any gene sequence they choose.
Yes, if you interupted the last game and instead of leaving the figures where they were now have to play the smae moves agian.
Ah, but you see, in chess you are not required to play the same moves again. There is enough variability in the gameplay that it is almost never the same twice. If you have to interrupt a chess game in progress, you can leave the game set up, just as you can leave a videogame on pause.
However, in story based games I think there's also an issue here. You're basically saying that people are wrong to want to use bookmarks in a book. That if the book is good, they won't mind restarting it every time they have to put it down until they can get through it in one sitting. Well, save points are more analogous to chapters, but still.
I pretty much agree with that. Most people will try to time their reading so that they can get to a "stopping point." And it is common to go back and read a few pages if you are forced to interrupt your reading in the middle of a scene. I'm sure that if there were a way for authors to require readers to begin and end at scene boundaries, they would do so. One of the reasons why movies are often more enjoyable in a theater is that there is not the temptation to pause it to answer a phone or get a snack (although I often wish there was an intermission in the middle, as was once the case for long movies).
I think there is fundamentally a difference between a story based game and a reflex based game. Golf, tetris, jezzball, those games don't make me want a save, as it's not really repetitive to start over.
If a game is story-based to this extent, then it should be a movie, not a game. And it is not merely reflex-based games that can have intrinsically enjoyable gameplay. Is it repetitive to play another game of chess?
But I also don't want to work through the last 5 minutes of dialogue, a cut scene, and a number of minor battles in say the latest Final Fantasy game if I die fighting the boss.
Exposition in story based games should always be easily skipped after the first time through. But if you don't enjoy replaying the minor battles before the boss, then either the minor battle gameplay is not enjoyable, the boss is too hard relative to other gameplay, or the save point is poorly positioned.
Many people don't like being told "this is where you should end your gaming today" for any number of reasons. Maybe you want to play for another 10 minutes before going to sleep. Maybe your mom/spouse/kid/friend just interrupted you to do something important. You need to quit NOW, but who wants to lose the last 10-30 or more minutes of play?
So pause it. Or turn it off. If the game play itself is actually enjoyable, then you'll enjoy doing it again. If you don't, it's not a problem with save, it's a problem with the fundamental game design.
Maybe you just completed something you think was very difficult, and you don't want to have to do it again.
Again, this indicates a deficiency with game design. Games should be challenging, rather than difficult. So replaying a section should be an enjoyable opportunity to improve your gameplay, not a chore.
You and the developer won't always agree on what warrents a save.
The point is that well-placed save points improve the game experience. Real-life games are designed this way. When you play golf, if you blow the last few holes, you don't cross them off your scorecard, go back a few holes and do them over. You just play another game. The game is constructed such that (a) the "chunks" are small enough to doing a whole game over is not a burden, and (b) the gameplay is intrinsically enjoyable enough that it is actually fun to play again, and you always believe that you have a realistic prospect of doing better the second time around. The single biggest problem with videogames is that it is possible to keep people playing by offering novel rewards--something new to see--as a reward for continued play. This leads designers to neglect the gameplay. So instead of the game being intrinsically fun to play, playing the game becomes a "fee" that you have to pay to see what comes next.
I could half the content and then have both areas loaded, but what happens when you come to a T-junction and have a choice? Should I cut the content delivery down to a third of what it would otherwise be?
Make the arms of the T long enough to load the content at either end. If necessary, segment the content so that only the immediately accessible area beyond the door is loaded first.
"Cinematic" camera angles. No, thank you. Understand that we need to see what my character sees. As soon as you start panning the camera around Mario for no better reason than to see the pretty sunset on the distance, we lose control. And here's another tip: If you have a single level where the player's character is required to run toward the camera, send the fucker back for more programming because you're not done yet.
I can't agree with either of these. Chase cameras get boring in a hurry. Many of the games I've most enjoyed had cinematic cameras with camera angles well thought out. If the camera angle is bad, you're doing the wrong thing. And for games with a puzzle aspect, being able to look around is important. I've also seen a number of games with effective "run toward the camera" scenes, usually with something chasing me. I only object if there are surprise pits and the like that my character should be able to see and avoid, but I can't because I can't see them coming up. In other words, it's OK not to let me see ahead, but don't use it to artificially add difficulty.
The analogy sort of breaks down there, but the point is we shouldn't ever see a "save point" in a game again. Limited saves were invented for consoles that didn't have the memory to let you "quicksave" (where you can save at any time, any where, with one keystroke like on a PC).
The need for a "Save anywhere" feature is inversely related to the quality of the game. Wanting to save at arbitrary locations is a symptom, not a problem. In a well-constructed game, defined save points add to excitement and suspense.
Why would I want arbitrary saves? Because
1) The only thing exciting is the boss battles, and everything leading up to them is drudgery. (poor game design)
2) The game play completely sucks, and the only reason to play is to see what cool graphics the next level has (even worse game design).
3) Losing to the boss requires me to repeat a puzzle I've already solved, which is boring now that I know the answer (poorly positioned save points).
4) The boss battles are so insanely difficult that it takes two dozen tries to even begin to figure out how to beat him. I'm so frustrated that I have no patience for anything except fighting the boss again, but I have to do a bunch of other stuff first (poorly positioned save point, and possibly poor design, depending upon whether fighting the boss over and over is actually enjoyable).
Slashdot Mirror
We will never be rid of stereotype unless we evolve away the instinct. We can just try to reduce the effect through social pressure.
Absolutely. Stereotyping is simply a form of inductive reasoning. The rabbit cannot afford to think, "Well, maybe this coyote will be friendly." Refraining from acting upon prejudice is a conscious act.
If you were congenitally missing an eye, and your other eye could see in the dark to compensate, it would be a positive result, but also a genetic defect.
It would be a genetic defect because there is a pathological consequence--loss of stereoscopic vision.
This is where prejudice really lies: in making judgments of individuals based on what you believe you know about the statistical characteristics group. Many hope to reduce prejudice by insisting that real statistical differences do not exist, but this avoids the real problem--it is still prejudice even if the difference is real. People are individuals, not statistics, and if you judge a person on a statistical basis, without troubling to learn what that person is like as an individual, then you are engaging in prejudice.
Moreover, trying too hard to deny such statistical differences can actually encourage prejudice, because it conveys the message, "If this difference were real, then it would be OK to be prejudiced."
The thing is that intelligence is not a single dimensional quantity like height is. We pretend it is by assigning an IQ value to some measurement of it, but even scientists who study it will tell you that a major problem in the field is a lack of understanding of just what intelligence is.
It is better to think of IQ, not as intelligence per se, but rather as a measure that has a substantial correlation with other measures of intelligence.
Suppose that you were trying to measure height, but for some reason were unable to do so directly. You might instead measure the length of the forefinger. Now length of forefinger is not height, and you will find some short people with long fingers and some tall people with short fingers. Still, the correlation is almost certainly good enough that you could learn a lot about the inheritance of height by measuring fingers.
Einstein's brain was actually the product of a genetic defect.
A genetic variation does not necessarily constitute a "defect." Are blue eyes a defect? Generally, the term defect is used when there is some kind of significant pathological consequence.
I'm sure everyone has heard the phrase "I don't want to know". Sometimes I wonder if there are some things that scientists should avoid studying. For example, what if it could be scientifically proven, without a doubt, that race A was in many ways superior to race B?
It is worth noting that people may also be harmed by the decision not to know. Information about the genetic basis of intelligence could lead to improvements in the treatment of genetic disease. Indeed, a suggestion in the paper is that the same genes may play a role in genetic diseases in this population. This could be very important for treatment.
If Apple choose to add a big chunk of markup for 'branding reasons' onto their Intel PCs then that puts them in a rather unique position on the Intel platform where customers are just paying for a nice case and and a piece of firmware which allows OS X to be booted.
This would be a legitimate complaint if Apple planned to sell "bare" Mac compatible computers, but it seems more likely that they will be offering them, as they have in the past, in a package with OSX and a bunch of other Apple software. Besides, even in the Intel world, you pay more for a brand name computer than for a "noname" clone.
Like XPostFacto you mean ?
A bit like XPostFacto, in that XPostFacto is used mainly by people who can't afford to buy an new Mac, and want to extend the lifespan of a older Mac, usually at home. It's not used much in a business environment. There are a variety of little problems, but they may be worth it for somebody who is willing to spend a little time working around them. And while XPostFacto needs to be revised for major system revisions, it typically comes through routine updates fine. After all, Apple doesn't have a reason to intentionally break XPostFacto. A hack to run Mac OSX on Wintel boxes will be viewed by Apple more like the hacks to strip the DRM off of iTunes songs, which break with every iTunes update.
I'm sure that somebody will figure out how to hack OS X to run on a generic Wintel box. It won't affect anything. Nobody will use it except for a handful of hackers, and not for any serious purposes, because it will be too much of a pain to maintain, with the patches breaking every time Apple releases a system update.
I read somewhere that Aqua largely depends on the AltiVec execution unit for the heavy graphical stuff. It's a totally uneducated guess, but I think the Intel based macs - if based on Intels current offering - will have some kind of coprocessor to support 128 bit vectors. This might make it very difficult to run OS-X on non-Apple hardware.
OS X Tiger offers the ability to offload the "heavy graphical stuff" to the graphics card, so this is unlikely to be an issue.
So even after this move, switching from whichever OS you are running now to OSX will entail purchasing Apple hardware.
Yes, I think the ability of Linux to run on generic Intel boxes, combined with the practical philosophical appeal of open source, will keep it lively.
However, if Apple manages to come out with reasonably priced Intel boxes, I'm not sure why anybody would buy anything else. After all, InteliMac(TM) owners will potentially have access to all Mac software and all Windows software. And Linux as well.
Want to bet that someone gets it running on generic hardware within a week of it hitting the shelves?
I'll bet it takes a bit longer than that. There are hacks to get Mac OS X to run on older, unsupported Mac hardware, but it clearly has taken quite a bit of work.
And what do you want to bet that every Apple security update will break the generic hardware patches?
And with an open source kernel, how, exactly, do they plan to stop people from hacking OS X onto commodity PC hardware?
The open source kernel doesn't have much to do with it. After all, the kernel already can be run on Wintel hardware. Still, I imagine that somebody will figure out how to hack it to run on generic Wintel boxes. After all, people have managed to hack OS X to run on older "unsupported" Mac hardware. Of course, Apple will doubtless take a sterner view of hacking OS X to run on Wintel boxes. Apple has dealt with this before, with Apple II clones. They dealt with it by lawsuits (Apple's litigiousness is by no means a recent phenomenon), and by making every OS revision incompatible with clones. I imagine that they will do the same, with every security update also updating the code that checks for "nonauthorized" hardware.
Macintouch quotes Phil Schiller, an Apple VP as saying, "We will not allow running Mac OS X on anything other than an Apple Mac."
Actually, I found that the 68K to PPC transition went pretty smoothly. Virtually everything still ran. On the first generation of PowerPC's, 68K applications ran perceptibly slower, but not by much--about like having a 68030 instead of a 68040. By the second generation, even the 68K applications were faster than on 68040.
And it is likely that this transition will probably go even more smoothly: Early versions of the PPC MacOS still were running a lot of 68K OS code in emulation; it is a safe bet that the Intel OS X will be 100% native code. And there is less hand-tweaked assembly code running around, so it will be easier for developers to simply recompile. Most major applications are already cross-platform, so developers already know what to tweak to enhance Intel processor performance.
My guess is that the transition will be smoother than the PPC transition, and much smoother than the OS X transition.
Financially, this is going to be a big bump for Apple. I'm certainly not going to order any more new Macs until the Intel systems are available. This may be one reason why they chose to do it now, when the success of the iPod will carry them through.
It may be the best decision for Apple, but I still think that it would have been better if they'd been able to reach a deal with IBM to develop the PPC further. I would much rather have seen multicore PPC's.
The question of whether the Intel OS X will run on generic Intel hardware seems to still be open. I'd guess not, but then I didn't believe that they'd switch to Intel in the first place.
Unless you think your (reclaimed) desk space is worth 1000$/sq ft, or that you think LCD is better in a interior-designer standpoint, in which case I'll grant you it's a better buy for you.
People put a high value on their desk space. Now that flat panels are available, I can't give a 17" CRT away. It kills me to throw away a perfectly good CRT monitor, but nobody wants them.
This isn't over by a long shot. The MPAA took a gamble, based on what they thought they had in Congress, and lost. They won't make the same mistake twice. Look for subtle changes in the "new and improved" DMCA, COPA and its children, and other roundabout ways to implement the same thing.
However, it looks like they are going to miss out on the time window for implementing this particular method of making consumer digital video more inconvenient and unreliable. By the time they get their act together, there will likely be too much broadcast flag-free equipment out there.
Darwinian selection should cause those societies which encourage/permit such uses to become extremely fragile and likely to become infected and wiped out thru ... oh, I don't know, HN51 avian bird flu let's say ... nah, that would never happen, biology adapting to weed out the less diverse and less resistant groups, would it?
Considering the extent of cross-mating made possible by modern transportation, combined with medical technology that enables genotypes to survive that probably would have been lethal in many times and places, human populations probably have greater genetic diversity than at any time in human history. A small fraction of the population screening out a handful of alleles is not going to make much of a dent.
The parents I've meet with Downs and autistc children have commented on how rewarding it is.
Rewarding != desirable. Hardly anybody with a Downs or autistic child wants to give it up. Parents love their children and raising even a seriously disabled child is a rewarding experience. But offer any prospective parent the rewarding experience or raising a child with autism or Downs syndrome vs. the rewarding experience of raising a child without these conditions, and guess which they'll choose?
And how are you going to do that? The technology is not that complicated. Are you going to prohibit people from owning a PCR machine (which is really just a precisely controlled hotplate)? Or make thermostable polymerase an illicit substance? Pretty soon, anybody will be able to test for any gene sequence they choose.
Yes, if you interupted the last game and instead of leaving the figures where they were now have to play the smae moves agian.
Ah, but you see, in chess you are not required to play the same moves again. There is enough variability in the gameplay that it is almost never the same twice. If you have to interrupt a chess game in progress, you can leave the game set up, just as you can leave a videogame on pause.
However, in story based games I think there's also an issue here. You're basically saying that people are wrong to want to use bookmarks in a book. That if the book is good, they won't mind restarting it every time they have to put it down until they can get through it in one sitting. Well, save points are more analogous to chapters, but still.
I pretty much agree with that. Most people will try to time their reading so that they can get to a "stopping point." And it is common to go back and read a few pages if you are forced to interrupt your reading in the middle of a scene. I'm sure that if there were a way for authors to require readers to begin and end at scene boundaries, they would do so. One of the reasons why movies are often more enjoyable in a theater is that there is not the temptation to pause it to answer a phone or get a snack (although I often wish there was an intermission in the middle, as was once the case for long movies).
I think there is fundamentally a difference between a story based game and a reflex based game. Golf, tetris, jezzball, those games don't make me want a save, as it's not really repetitive to start over.
If a game is story-based to this extent, then it should be a movie, not a game. And it is not merely reflex-based games that can have intrinsically enjoyable gameplay. Is it repetitive to play another game of chess?
But I also don't want to work through the last 5 minutes of dialogue, a cut scene, and a number of minor battles in say the latest Final Fantasy game if I die fighting the boss.
Exposition in story based games should always be easily skipped after the first time through. But if you don't enjoy replaying the minor battles before the boss, then either the minor battle gameplay is not enjoyable, the boss is too hard relative to other gameplay, or the save point is poorly positioned.
Many people don't like being told "this is where you should end your gaming today" for any number of reasons. Maybe you want to play for another 10 minutes before going to sleep. Maybe your mom/spouse/kid/friend just interrupted you to do something important. You need to quit NOW, but who wants to lose the last 10-30 or more minutes of play?
So pause it. Or turn it off. If the game play itself is actually enjoyable, then you'll enjoy doing it again. If you don't, it's not a problem with save, it's a problem with the fundamental game design.
Maybe you just completed something you think was very difficult, and you don't want to have to do it again.
Again, this indicates a deficiency with game design. Games should be challenging, rather than difficult. So replaying a section should be an enjoyable opportunity to improve your gameplay, not a chore.
You and the developer won't always agree on what warrents a save.
The point is that well-placed save points improve the game experience. Real-life games are designed this way. When you play golf, if you blow the last few holes, you don't cross them off your scorecard, go back a few holes and do them over. You just play another game. The game is constructed such that (a) the "chunks" are small enough to doing a whole game over is not a burden, and (b) the gameplay is intrinsically enjoyable enough that it is actually fun to play again, and you always believe that you have a realistic prospect of doing better the second time around. The single biggest problem with videogames is that it is possible to keep people playing by offering novel rewards--something new to see--as a reward for continued play. This leads designers to neglect the gameplay. So instead of the game being intrinsically fun to play, playing the game becomes a "fee" that you have to pay to see what comes next.
I could half the content and then have both areas loaded, but what happens when you come to a T-junction and have a choice? Should I cut the content delivery down to a third of what it would otherwise be?
Make the arms of the T long enough to load the content at either end. If necessary, segment the content so that only the immediately accessible area beyond the door is loaded first.
"Cinematic" camera angles. No, thank you. Understand that we need to see what my character sees. As soon as you start panning the camera around Mario for no better reason than to see the pretty sunset on the distance, we lose control. And here's another tip: If you have a single level where the player's character is required to run toward the camera, send the fucker back for more programming because you're not done yet.
I can't agree with either of these. Chase cameras get boring in a hurry. Many of the games I've most enjoyed had cinematic cameras with camera angles well thought out. If the camera angle is bad, you're doing the wrong thing. And for games with a puzzle aspect, being able to look around is important. I've also seen a number of games with effective "run toward the camera" scenes, usually with something chasing me. I only object if there are surprise pits and the like that my character should be able to see and avoid, but I can't because I can't see them coming up. In other words, it's OK not to let me see ahead, but don't use it to artificially add difficulty.
The analogy sort of breaks down there, but the point is we shouldn't ever see a "save point" in a game again. Limited saves were invented for consoles that didn't have the memory to let you "quicksave" (where you can save at any time, any where, with one keystroke like on a PC).
The need for a "Save anywhere" feature is inversely related to the quality of the game. Wanting to save at arbitrary locations is a symptom, not a problem. In a well-constructed game, defined save points add to excitement and suspense.
Why would I want arbitrary saves? Because
1) The only thing exciting is the boss battles, and everything leading up to them is drudgery. (poor game design)
2) The game play completely sucks, and the only reason to play is to see what cool graphics the next level has (even worse game design).
3) Losing to the boss requires me to repeat a puzzle I've already solved, which is boring now that I know the answer (poorly positioned save points).
4) The boss battles are so insanely difficult that it takes two dozen tries to even begin to figure out how to beat him. I'm so frustrated that I have no patience for anything except fighting the boss again, but I have to do a bunch of other stuff first (poorly positioned save point, and possibly poor design, depending upon whether fighting the boss over and over is actually enjoyable).