Excuse the parent poster - he doesn't understand that we don't live in a vacuum.
Most moderately intelligent people understand that what we buy or participate in is very heavily dependent on what our friends, neighbors or families are buying. Folks like him who believe that people participate in a market without influence from trends are delusional.
In other words, you know what you're talking about. Parent poster is lost.
You're crazy. By the time somebody like bruce is involved, upper management has gotten wind of the problems and just wants honest professional opinion by somebody with a history and experience in properly evaluating the status of large products.
This kind of assessment is specifically for people not entrenched in the project - it is for people with the power to prevent any furthur entrenchment. As another poster noted, if the intended audience would dismiss such an assessment, than no degree of 'professionalism' would have changed their mind.
What bruce delivered is intellectually honest - and it's rather telling that he's the consultant, not you, so I would rather believe that he knows how to do his job better than you. Every assertion he made was clearly explained, and each was accompanied by an example.
Instead, it's a fait accompli for them, or people will leave in droves, IMO.
To what? Apple would be making OSes, not phones and MP3 players if selling their OS on beige intel boxes didn't constitute corporate suicide. People talk about Linux, but its just not viable. Its an awesome OS but it will *never* be the dominant OS for the same reason that perfectly awesome products which are cheap rarely outsell the well marketed corperate 'pass the buck' options. Linux is much much more friendly to anybody that lifts the hood. It took me 4 hours today to crawl through the hundreds of thousands of registry keys *on a new installation of windows* to figure out some super ultra obscure python file association bug. But nothing can beat taking a portion of your budget, paying a company, and when things go wrong, being able to point out that you're spending support money on the same corporate behemoth that wrote the thing. It is just too important in business for people to be able to say you've got a support contract with the folks who wrote the damn thing. I'm ignoring the fact that it is not sound logic - its just how business works, and a great deal of end users too.
I really don't get why people aren't beating down the doors to the government to ask why an OS as super ultra fucking pleasurable as OS X is unable to compete in the marketplace requires such a price point premium on hardware and is locked down. Man, I am not going to pay for an Apple machine, but I would absolutely kill to hand over my money to buy OS X. BSD under the hood, the best GUI front-end ever - there is something seriously wrong that both unix nerds and end users aren't wondering why the market has apparently decided it's so boutique and special. It's not the market - Microsoft is just really untouchable. They're in such a powerful position in the computer world that its basically unfathomable that they lose massive market share. It'd be too painful for everyone.
I doubt that anyone will ever state that Ray's singularity has been achieved, because we'd perceive that whatever development, even to ourselves, was going on was entirely within the scope of natural development. People look at this like its technology superseding biology, but since these kinds of conjectures are always framed within the context of genetic engineering or biological engineering, it seems to me that the 'hardware' will inevitably be biological. (Nevermind the fact that it took millions of years for us to learn how to build something that could beat us at a board game with a very simple set of rules.) It took millions and millions of years for humans to become sufficiently developed, but I don't care how self-recursive our technological skills become - we'll have accepted those developments as natural and normal to whatever necessary degree. A singularity is an impossibility because it implies that the goal of technology is to create better technology - it never was, never has been, and never will be. It will always be about solving problems. It doesn't matter how 'fast' we get at it; in fact, the very tenets of technology state that new technology gives rise to new problems. I almost laugh and think that were this singularity occurr, we'd be working our assess of as fast as possible to figure out a way of slowing it down. The obsession on speed of technological progress and knowledge gathering is disingenuous - history is rife with examples of working our asses off to create something new as soon as possible, and then suddenly faced with the realization that we did it so fast, we created a new problem that we now have to solve even faster! I saw his keynote at GDC'08, and it just left me feeling really empty. Not because I don't think his vision can come to pass; precisely because I think its impossible to define what he's alluding to without a lot of hand waving and misnomers.
I mean, if you were a species that hadn't really changed in millions of years (which of course they are) would you be looking at humans and going, "HOLY CRAP, HORSES TO CARS IN 150 YEARS... A BEAUTIFUL FUTURE AWAITS!" Eventually, we'll die off or evolve, but I don't understand how its really relevant to anything beyond the fact that if it's so inevitable, its practically moot. Nobody on earth has any clue what it will feel like - or even if 'feeling' is an evolutionary misstep or over complication that we'll purposely phase out.
I'm much more interested in what makes life so complicated and strange than some vision of complete dominance over our environment and computational capabilities.
To put it another way - we're a hell of a lot more intelligent than cats. But would you really think its healthy and/or even particularly relevant for a cat to really really wish it could do trigonometry?
The point is still valid within the context of the idea that whatever input required for a human to develop healthily is far beyond staring at a chess board and moving some pieces for 100 years.
We'd need robots who could design equipment for themselves, to scale mountains. To invent instruments. To scour the depths of the ocean. The point is still valid in the sense that people are a product of their environment, and what makes the human experience so unique is that we're constantly attempting to gain more access to more input. Presumably, any old brain in a box placed in a single room, unable to move, would cease being healthy after awhile, and probably even recognizably human after years because I would have to imagine that some part of the programming of the human mind requires or in the very least infers the ability to alter and modify our environment to a satisfactory degree.
Left joins are about the last thing an uneducated user is going to do. Personally, as long as the people accessing the database arn't "programmers" I'd be very comfortable permitting read access even if they could create monster statements that would hose the server.
The only fucked up, wrong SQL statements I've ever seen came from programmers. And I'm sure you can limit performance usage and grant granular enough access to make it a non issue, even if the customer has some shitty programmer.
I don't get the thrust of the article. You never need to say no to a client; you say yes, with provisions.
The term is in widespread use today specifically because it was used by conservatives to rile people up. Think of it as a social troll that people fall for millions of times a day on internet forums. He wasn't trying to be PC, he was just being civil. Of course the world isn't simple, of course you're going to say some things that are true that offend people.. but you'll find that it's a lot easier to get people to see things the way you do when you demonstrate a sensitivity to their beliefs. That's not being weak, that's about increasing your chances of having your viewpoints respected. If you don't care about whether people might be able to see things from your side, then why open your mouth in the first place?
They work well if you're ten people. They feel like sweatshops when you're 80. They're loud, lack privacy, and its too easy for people to yell across the room or walk up to your desk instead of forcing them to think about whether they really need to initiate the communication in the first place or if its something they can figure out/live without in the first place.
Open Concepts are music to a companies' ears. They're cheap as hell. Designers/artists/loud people love them. But engineers who can't do math while listening to music on headphones rightfully hate them.
Fine for tactical games - but consider that players of Quake3 rocket arena frequently resort to rocketjump-to-360 rail gun shots only matches purely for fun. Flat out impossible on a controller. I fell in love with Call of Duty 4 via the 360 demo, but I probably would not have bought it were it not available on PC. Fast paced FPSes just can't be exploited properly on consoles. Those who champion Halo, etc.. look, its not a terrible game, but its just not what a true FPS is supposed to be in my book. It's all about conservation of your aim (ie, being in the right spot before you need to be) rather than pure reflex and aiming.
Why not? The market spoke pretty hard here - their shit went out of business, and while its nice when the law can protect people from their own shortsightedness, I'm not convinced DRM is such a technology that needs regulation. Yet?
Lots of companies have backed off or weakened their DRM to more reasonable levels due to consumer backlash (sony, anyone?) As big a cynic of the 'correctness' of the invisible hand as I am, I really havn't seen DRM as an example of a technology that restricts customers to the point where they can't reply via the market.
Be reasonable. Lots of people start fires with cigarettes inside houses, so its not news. When its a 360 power supply, it is - given that it seems to be the first time it's caused (or its owner) a fire.
Lots of things get hot, but maybe not hot enough to start a fire. Thats where the interest level comes from - now we know. If you work hard enough, you can start a house fire with a 360 power supply. Nobody needs to be reminded that using a bbq indoors might cause a fire, its patently obvious.
The mistake is thinking that in a first world nation, power comes from might. Theres a reason why the 1st is based on allowing people to communicate, and the second is based on allowing people to shoot.
That's all I'm saying. I think the 1st is the reason the US citizens are empowered and happy, not the 2nd. When you look at incarceration rates and violence data, it seems quite clear to me that some countries have figured out that the 1st is VASTLY more important than the 2nd.
Seriously, if US citizenry ever had to rise up, they'd use the guns given to them by the government. It's not like it'd be the US army versus the homeland. It seems quite obvious to me that the right to personally bear arms would have absolutely no value to an armed insurrection. It'd be completely incidental. Gun violence in the states, by accident, or by crime, is very high. It seems to me that it'd be astute to downplay the cultural affinity for gun ownership, because at the end of the day, it doesn't seem to improve day to day live for citizens, and its unlikely to factor into a situation where citizens feel they need to rebel.
Government is in the pockets of big business. My recommendation is to demand that the public demand more from government to demand more from big business. You pretty much make my point for me.
Short of governments bail out companies because they're well connected, and a need to retain some level of stability (like, if every airline in the states went bankrupt at the same time, would you advocate the collective economic punishment of no functional airlines? I'd think my tax dollars are better spent on at least maintaining some level of commercial air service.)
You're saying the government regulates on the corporate side. I'm saying we should also demand regulation to protect consumers.
In a lawless world, the power just goes to the most powerful. Regulation should be demanded to make credit scores a publicly available piece of information. I really don't believe in a totally laissez faire system - maybe the market ultimately selects wisely, but it takes a lot of pain and suffering to move the invisible hand.
I work on engines (ie, the middleware) not games. So using a competitor's product that we've already deemed unsuitable for our needs doesn't really deal with the key issue of whether its more annoying to abstract directx/sony/nintendo
Actually, as a long time Unix programmer and current game console developer, DirectX really is a lot more pleasurable than most of the alternatives. I'm not a big MS fan, but DirectX is nice, and the documentation is very good.
Developing for Nintendo and Sony requires a bit of a taste for the black arts.
I think you're missing my point. I agree with your sentiment. I'm saying better regulation is needed because its clear that in a free market where you credit rating is not
The individual about whom the information is collected has the right to have it verified. False information is libel or defamation.
Couldn't agree more. The reality is that the demand isn't there to consider erroneous credit scores libel or defamation. But given that it seems to me credit ratings are simply a summary of what banks, loaners, and bill givers 'think' of you. Its hard to say a credit rating is libel since its an artificial rating system to begin with. That's a large can of worms. Its why I'd advocate regulation of the industry to force transparency and consistency. I don't really understand how privacy enters into the equation, since if you're seeking a service from somebody and they check your credit rating, you'd already given them your name, etc. The original complaint was about why does the individual have such a hard time checking the number, but companies have very little problems. To me, its just supply and demand, until government steps in and says, now wait.. if you're going to discriminate against borrowers based on their history (which to me I accept as being necessary for a healthy economy) at least pay a tax on making sure this data is available to the borrowers even if theres no real business case to do so.
You're saying the government 'enforces' the corporate side. I'm saying, without a government, I'd be very surprised if you would be able to gain access to your credit rating at all. Regulation exists to try and make markets more transparent because the free market doesn't gravitate towards even playing fields.
If I have an object that should last the lifetime of an app, and a reference from that to another object, which has a list of other objects, to another object, to another object, to another object, to anther object, to a list of objects with an object that contains a list that is never cleared but just added too every time you hit 'built', thats bad architecture. But its also a reality in some environments.
A language is only as good as its debugging tools are. A really awesome GC does nothing in with a 5 year old application with 100,000 lines of code when pertaining to a leak you inherited from a programmer who left 2 years ago. 'Most programmers' could mean not programming memory leaks in C++ app even in the absence of smart pointers. In reality, you will have programmers that create leaks, no matter how awesome the GC is. Because its not really a GC issue. GCs clean shit you don't need anymore. Programmers write code that makes the interpreter think you still need something.
What? No one uses reference counting in modern garbage collectors, and cycles are comfortably handled in virtually all modern GC'd languages.
Pray tell, how does a modern GC handle a case where you have an object thats still alive by programming standards through a maze of instances, but not still needed by 'design' standards? I know Python's GC sucks, but its not like Python is not in use in many production environments. I'm not under the impression that the Python GC is built on totally outdated architecture. I'm having a great deal of trouble troubleshooting leaks introduced by long gone programmers in a very large codebase. Its not all reference counting obviously - the object has to be accessible from the root, but our application is so large that python code just to analyze the heap is prohibitively impractical. At this point, I'm just trying to backport an extension that can analyse the object heap in C, and not add in the additional complication of actually affecting the refcount.
Thats a credit economy - given a free market, the demand isn't going to come from people who need reports on themselves, the demand is going to come from people who want to check your credit. So its no surprise that the bias the market dictates is for people who want access to other peoples' credit, not the credit owners themselves.
Without regulation, it seems rather natural to me that lenders and scammers could supplant any revenue that might be had from people asking for their own credit reports by offering money to credit reporting companies because its in their best interest for people not to know their own credit rating.
If your refcount doesn't reach 0, then you have a leak, GC or not. So sure, GCs are nice for certain tasks, convenient, if you will, but theres no way in hell I'd put a newb programmer on coding a hits per second kind of massive function GC or no GC. GC is convenient for programmers who know what they're doing, not some kind of magical solution for programmers who don't understand where the memory is going.
No fancy language features used - no template metaprogramming, no pointers, no virtual inheritance. The kind of stuff a new C++ programmer might write after reading Stroustrop.
I think most programmers use the STL when its obvious and readable. My experience is that people will skip 'accumulate' and iterate to be more explicit. The STL is a great path for mental masturbation and lots of problems. The STL is useful like a rope is. You can either use it for where its very clear you're not going to hang yourself.. or you can hang yourself.
I love the container classes and the like, but I have a very hard time thinking thats what a new C++ programmer is going to write if I ask him or her to write a function that averages ints.
C++ and STL can certainly offer you tons of ways of writing obscure bugs.. give me a language that doesn't. We could sit here for days and voley obscure "spot the bug" arguments, but I don't think it makes a very good case for the unsuitability of a language.
Slashdot Mirror
It's about knowing things I do not.
As in, you don't know that there can be more than one person who knows things you do not.
Excuse the parent poster - he doesn't understand that we don't live in a vacuum.
Most moderately intelligent people understand that what we buy or participate in is very heavily dependent on what our friends, neighbors or families are buying. Folks like him who believe that people participate in a market without influence from trends are delusional.
In other words, you know what you're talking about. Parent poster is lost.
You're crazy. By the time somebody like bruce is involved, upper management has gotten wind of the problems and just wants honest professional opinion by somebody with a history and experience in properly evaluating the status of large products.
This kind of assessment is specifically for people not entrenched in the project - it is for people with the power to prevent any furthur entrenchment. As another poster noted, if the intended audience would dismiss such an assessment, than no degree of 'professionalism' would have changed their mind.
What bruce delivered is intellectually honest - and it's rather telling that he's the consultant, not you, so I would rather believe that he knows how to do his job better than you. Every assertion he made was clearly explained, and each was accompanied by an example.
Instead, it's a fait accompli for them, or people will leave in droves, IMO.
To what? Apple would be making OSes, not phones and MP3 players if selling their OS on beige intel boxes didn't constitute corporate suicide. People talk about Linux, but its just not viable. Its an awesome OS but it will *never* be the dominant OS for the same reason that perfectly awesome products which are cheap rarely outsell the well marketed corperate 'pass the buck' options. Linux is much much more friendly to anybody that lifts the hood. It took me 4 hours today to crawl through the hundreds of thousands of registry keys *on a new installation of windows* to figure out some super ultra obscure python file association bug. But nothing can beat taking a portion of your budget, paying a company, and when things go wrong, being able to point out that you're spending support money on the same corporate behemoth that wrote the thing. It is just too important in business for people to be able to say you've got a support contract with the folks who wrote the damn thing. I'm ignoring the fact that it is not sound logic - its just how business works, and a great deal of end users too.
I really don't get why people aren't beating down the doors to the government to ask why an OS as super ultra fucking pleasurable as OS X is unable to compete in the marketplace requires such a price point premium on hardware and is locked down. Man, I am not going to pay for an Apple machine, but I would absolutely kill to hand over my money to buy OS X. BSD under the hood, the best GUI front-end ever - there is something seriously wrong that both unix nerds and end users aren't wondering why the market has apparently decided it's so boutique and special. It's not the market - Microsoft is just really untouchable. They're in such a powerful position in the computer world that its basically unfathomable that they lose massive market share. It'd be too painful for everyone.
It's depressing that you could link to, and presumably read that, and not understand what it means.
It's kind of scary to think you may hold opinions and beliefs you take as truth based on such obvious misinterpretations.
I doubt that anyone will ever state that Ray's singularity has been achieved, because we'd perceive that whatever development, even to ourselves, was going on was entirely within the scope of natural development. People look at this like its technology superseding biology, but since these kinds of conjectures are always framed within the context of genetic engineering or biological engineering, it seems to me that the 'hardware' will inevitably be biological. (Nevermind the fact that it took millions of years for us to learn how to build something that could beat us at a board game with a very simple set of rules.) It took millions and millions of years for humans to become sufficiently developed, but I don't care how self-recursive our technological skills become - we'll have accepted those developments as natural and normal to whatever necessary degree. A singularity is an impossibility because it implies that the goal of technology is to create better technology - it never was, never has been, and never will be. It will always be about solving problems. It doesn't matter how 'fast' we get at it; in fact, the very tenets of technology state that new technology gives rise to new problems. I almost laugh and think that were this singularity occurr, we'd be working our assess of as fast as possible to figure out a way of slowing it down. The obsession on speed of technological progress and knowledge gathering is disingenuous - history is rife with examples of working our asses off to create something new as soon as possible, and then suddenly faced with the realization that we did it so fast, we created a new problem that we now have to solve even faster! I saw his keynote at GDC'08, and it just left me feeling really empty. Not because I don't think his vision can come to pass; precisely because I think its impossible to define what he's alluding to without a lot of hand waving and misnomers.
... A BEAUTIFUL FUTURE AWAITS!" Eventually, we'll die off or evolve, but I don't understand how its really relevant to anything beyond the fact that if it's so inevitable, its practically moot. Nobody on earth has any clue what it will feel like - or even if 'feeling' is an evolutionary misstep or over complication that we'll purposely phase out.
I mean, if you were a species that hadn't really changed in millions of years (which of course they are) would you be looking at humans and going, "HOLY CRAP, HORSES TO CARS IN 150 YEARS
I'm much more interested in what makes life so complicated and strange than some vision of complete dominance over our environment and computational capabilities.
To put it another way - we're a hell of a lot more intelligent than cats. But would you really think its healthy and/or even particularly relevant for a cat to really really wish it could do trigonometry?
The point is still valid within the context of the idea that whatever input required for a human to develop healthily is far beyond staring at a chess board and moving some pieces for 100 years.
We'd need robots who could design equipment for themselves, to scale mountains. To invent instruments. To scour the depths of the ocean. The point is still valid in the sense that people are a product of their environment, and what makes the human experience so unique is that we're constantly attempting to gain more access to more input. Presumably, any old brain in a box placed in a single room, unable to move, would cease being healthy after awhile, and probably even recognizably human after years because I would have to imagine that some part of the programming of the human mind requires or in the very least infers the ability to alter and modify our environment to a satisfactory degree.
Left joins are about the last thing an uneducated user is going to do. Personally, as long as the people accessing the database arn't "programmers" I'd be very comfortable permitting read access even if they could create monster statements that would hose the server.
The only fucked up, wrong SQL statements I've ever seen came from programmers. And I'm sure you can limit performance usage and grant granular enough access to make it a non issue, even if the customer has some shitty programmer.
I don't get the thrust of the article. You never need to say no to a client; you say yes, with provisions.
modern PC climate?
.. but you'll find that it's a lot easier to get people to see things the way you do when you demonstrate a sensitivity to their beliefs. That's not being weak, that's about increasing your chances of having your viewpoints respected. If you don't care about whether people might be able to see things from your side, then why open your mouth in the first place?
time to learn
The term is in widespread use today specifically because it was used by conservatives to rile people up. Think of it as a social troll that people fall for millions of times a day on internet forums. He wasn't trying to be PC, he was just being civil. Of course the world isn't simple, of course you're going to say some things that are true that offend people
Count me out for Open Designs.
They work well if you're ten people. They feel like sweatshops when you're 80. They're loud, lack privacy, and its too easy for people to yell across the room or walk up to your desk instead of forcing them to think about whether they really need to initiate the communication in the first place or if its something they can figure out/live without in the first place.
Open Concepts are music to a companies' ears. They're cheap as hell. Designers/artists/loud people love them. But engineers who can't do math while listening to music on headphones rightfully hate them.
You need to look into the history of copyright law. It was enacted for some good reasons, but not the reasons you seem to think.
Fine for tactical games - but consider that players of Quake3 rocket arena frequently resort to rocketjump-to-360 rail gun shots only matches purely for fun. Flat out impossible on a controller. I fell in love with Call of Duty 4 via the 360 demo, but I probably would not have bought it were it not available on PC. Fast paced FPSes just can't be exploited properly on consoles. Those who champion Halo, etc .. look, its not a terrible game, but its just not what a true FPS is supposed to be in my book. It's all about conservation of your aim (ie, being in the right spot before you need to be) rather than pure reflex and aiming.
Why not? The market spoke pretty hard here - their shit went out of business, and while its nice when the law can protect people from their own shortsightedness, I'm not convinced DRM is such a technology that needs regulation. Yet?
Lots of companies have backed off or weakened their DRM to more reasonable levels due to consumer backlash (sony, anyone?) As big a cynic of the 'correctness' of the invisible hand as I am, I really havn't seen DRM as an example of a technology that restricts customers to the point where they can't reply via the market.
Be reasonable. Lots of people start fires with cigarettes inside houses, so its not news. When its a 360 power supply, it is - given that it seems to be the first time it's caused (or its owner) a fire.
Lots of things get hot, but maybe not hot enough to start a fire. Thats where the interest level comes from - now we know. If you work hard enough, you can start a house fire with a 360 power supply. Nobody needs to be reminded that using a bbq indoors might cause a fire, its patently obvious.
I have a 360, but I find punching myself in the face is a suitable replacement for Halo 3.
The mistake is thinking that in a first world nation, power comes from might. Theres a reason why the 1st is based on allowing people to communicate, and the second is based on allowing people to shoot.
That's all I'm saying. I think the 1st is the reason the US citizens are empowered and happy, not the 2nd. When you look at incarceration rates and violence data, it seems quite clear to me that some countries have figured out that the 1st is VASTLY more important than the 2nd.
Seriously, if US citizenry ever had to rise up, they'd use the guns given to them by the government. It's not like it'd be the US army versus the homeland. It seems quite obvious to me that the right to personally bear arms would have absolutely no value to an armed insurrection. It'd be completely incidental. Gun violence in the states, by accident, or by crime, is very high. It seems to me that it'd be astute to downplay the cultural affinity for gun ownership, because at the end of the day, it doesn't seem to improve day to day live for citizens, and its unlikely to factor into a situation where citizens feel they need to rebel.
Government is in the pockets of big business. My recommendation is to demand that the public demand more from government to demand more from big business. You pretty much make my point for me.
Short of governments bail out companies because they're well connected, and a need to retain some level of stability (like, if every airline in the states went bankrupt at the same time, would you advocate the collective economic punishment of no functional airlines? I'd think my tax dollars are better spent on at least maintaining some level of commercial air service.)
You're saying the government regulates on the corporate side. I'm saying we should also demand regulation to protect consumers.
In a lawless world, the power just goes to the most powerful. Regulation should be demanded to make credit scores a publicly available piece of information. I really don't believe in a totally laissez faire system - maybe the market ultimately selects wisely, but it takes a lot of pain and suffering to move the invisible hand.
I work on engines (ie, the middleware) not games. So using a competitor's product that we've already deemed unsuitable for our needs doesn't really deal with the key issue of whether its more annoying to abstract directx/sony/nintendo
The only problem with font size is that it conflicts with the desire to avoid line wrap, have as many lines of code on screen at one time, etc.
I use a high resolution, I find that taking a 10 minute break every hour helps immensely.
Actually, as a long time Unix programmer and current game console developer, DirectX really is a lot more pleasurable than most of the alternatives. I'm not a big MS fan, but DirectX is nice, and the documentation is very good.
Developing for Nintendo and Sony requires a bit of a taste for the black arts.
I think you're missing my point. I agree with your sentiment. I'm saying better regulation is needed because its clear that in a free market where you credit rating is not
.. if you're going to discriminate against borrowers based on their history (which to me I accept as being necessary for a healthy economy) at least pay a tax on making sure this data is available to the borrowers even if theres no real business case to do so.
The individual about whom the information is collected has the right to have it verified. False information is libel or defamation.
Couldn't agree more. The reality is that the demand isn't there to consider erroneous credit scores libel or defamation. But given that it seems to me credit ratings are simply a summary of what banks, loaners, and bill givers 'think' of you. Its hard to say a credit rating is libel since its an artificial rating system to begin with. That's a large can of worms. Its why I'd advocate regulation of the industry to force transparency and consistency. I don't really understand how privacy enters into the equation, since if you're seeking a service from somebody and they check your credit rating, you'd already given them your name, etc. The original complaint was about why does the individual have such a hard time checking the number, but companies have very little problems. To me, its just supply and demand, until government steps in and says, now wait
You're saying the government 'enforces' the corporate side. I'm saying, without a government, I'd be very surprised if you would be able to gain access to your credit rating at all. Regulation exists to try and make markets more transparent because the free market doesn't gravitate towards even playing fields.
If I have an object that should last the lifetime of an app, and a reference from that to another object, which has a list of other objects, to another object, to another object, to another object, to anther object, to a list of objects with an object that contains a list that is never cleared but just added too every time you hit 'built', thats bad architecture. But its also a reality in some environments.
A language is only as good as its debugging tools are. A really awesome GC does nothing in with a 5 year old application with 100,000 lines of code when pertaining to a leak you inherited from a programmer who left 2 years ago. 'Most programmers' could mean not programming memory leaks in C++ app even in the absence of smart pointers. In reality, you will have programmers that create leaks, no matter how awesome the GC is. Because its not really a GC issue. GCs clean shit you don't need anymore. Programmers write code that makes the interpreter think you still need something.
What? No one uses reference counting in modern garbage collectors, and cycles are comfortably handled in virtually all modern GC'd languages.
Pray tell, how does a modern GC handle a case where you have an object thats still alive by programming standards through a maze of instances, but not still needed by 'design' standards? I know Python's GC sucks, but its not like Python is not in use in many production environments. I'm not under the impression that the Python GC is built on totally outdated architecture. I'm having a great deal of trouble troubleshooting leaks introduced by long gone programmers in a very large codebase. Its not all reference counting obviously - the object has to be accessible from the root, but our application is so large that python code just to analyze the heap is prohibitively impractical. At this point, I'm just trying to backport an extension that can analyse the object heap in C, and not add in the additional complication of actually affecting the refcount.
Thats a credit economy - given a free market, the demand isn't going to come from people who need reports on themselves, the demand is going to come from people who want to check your credit. So its no surprise that the bias the market dictates is for people who want access to other peoples' credit, not the credit owners themselves.
Without regulation, it seems rather natural to me that lenders and scammers could supplant any revenue that might be had from people asking for their own credit reports by offering money to credit reporting companies because its in their best interest for people not to know their own credit rating.
Programmers will code memory leaks. Always.
If your refcount doesn't reach 0, then you have a leak, GC or not. So sure, GCs are nice for certain tasks, convenient, if you will, but theres no way in hell I'd put a newb programmer on coding a hits per second kind of massive function GC or no GC. GC is convenient for programmers who know what they're doing, not some kind of magical solution for programmers who don't understand where the memory is going.
No fancy language features used - no template metaprogramming, no pointers, no virtual inheritance. The kind of stuff a new C++ programmer might write after reading Stroustrop.
.. or you can hang yourself.
.. give me a language that doesn't. We could sit here for days and voley obscure "spot the bug" arguments, but I don't think it makes a very good case for the unsuitability of a language.
I think most programmers use the STL when its obvious and readable. My experience is that people will skip 'accumulate' and iterate to be more explicit. The STL is a great path for mental masturbation and lots of problems. The STL is useful like a rope is. You can either use it for where its very clear you're not going to hang yourself
I love the container classes and the like, but I have a very hard time thinking thats what a new C++ programmer is going to write if I ask him or her to write a function that averages ints.
C++ and STL can certainly offer you tons of ways of writing obscure bugs