Sounds like a good idea. Cache is a very good concept, that has proved itself across and in different systems.
Basically in an environment where size matters less than speed, cache brings benefits.
i.e. IMO, cache servers would do Internet good as a whole, provided they are:
1. Enviromentally competitive - draw little power, so they do not contribute to a whole lot of watts Internet contributes to already. I am talking about inexpensive systems with lots of cheap hard drive space. A lot of cache server systems to 'speed up' Internet means a lot of power draw, so obviously a factor.
2. Legal P2P content? Oh, lets play police again shall we? 90% of p2p traffic is illegal, so how caching 10% of the part that is legal is a solution? Besides, is encrypted traffic illegal? Speaking in Apache terms, is it an 'allow-deny' or a 'deny-allow' rule?
Perhaps use routers with a hard drive. That way it is a polymorph system, that acts as a good old TCP/IP router, but may retrieve content from cache, if the connection allows it. Replacing old routers with such upgraded systems can be done over time.
A cheap harddrive is what, a 5W? It only needs to be faster than cache-miss data retrieval equivalent, right?
I think Mr. Gates should finally step off his self-raised piedestal of fame and grandeure.
Somehow it is so human and unfortunately, in my opinion, completely wrong, to assume a company is success because of one single reason, or a small set of key reasons, that a chairman consciously can express. There is a whole lot of history behind Microsoft, and some things that happened in the past, at the right time, and the right place, helped it to become what it is. No more no less. Granted Microsoft started a whole market of consumer computerism, but it quickly monopolized it like a big predator that just killed a big meaty prey. A wise person would not make hasty conclusions, and instead, tell Microsoft for what it is - an opportunist, a scavenger, a conman, yet with a charm and blessing of efficiency, modest competency and brains. Last being only the merit of smart geeks in search of a job. Many people would do anything for food. Why not sit in the office at Redmond and type away chunks of IE code? It's not peer reviewed anyway, and everyone around cheers you up telling you how good programmer you are and how you contribute to the better IT of the planet. Talk about sandbox-career.
The fact is, today, Microsoft is by no means the most innovative, successful, inspiring, and competent company. Multitude of recent facts related to them stand as proof. Things like OOXML vote rigging, Vista- disaster (and for people who actually took time to read Microsoft blogs years prior to the actual release, a disaster even worse), all their semi-tech talk about features, that end up vaporware. It has, and is a big circus. The sad fact is anyone can succeed, given enough programmers, and a aggressive strategy. It's business. And the rules of engagement in business allow and encourage such strategies. It does not mean products are any good. Given the absence of choice, they might be. But look at IE for instance. Self-crowned jewel of Internet Surfing for Average Joe, until some people decided to show the world how bad it really was - inventing Firefox.
Consumerism appreciates Microsoft. Awareness and modern market does not. A market where software is peer-reviewed, criticized, praised, compared to, dissected as source code, and what not. In the midst of it all, Microsoft tries to play it like it is still 80's. Where you buy a floppy disk worth 100$ of software, bring it home and plug it into your MS-DOS, typing away obscure stuff at command line, on which you read in a equally expensive "Learn Computers" book.
Sadly, now a whole lot of young aspiring managers heed to the Message of Gates, an aging billionaire with supposed infinite knowledge of IT. I doubt he is one. He is just one hell of a CEO.
True innovation, inspiration and science are whole another field. But then again, let us not confuse science with good old marketing, which Microsoft excels at. In fact it should have been called Marketsoft. What does Micro- stand for again??
And of course, i have no idea how to market stuff. But a better product needs less marketing. And I am talking about science.
REALLY good point!
The only downside is human factor, of course, - will enough people be willing to spend precious minutes of their life to do take this kind of procedure?
Telling apart animals is CONSIDERABLY different than telling apart symbols, in my opinion. Given sufficiently low resolution, sufficiently bad lightning and focus, even you will have a problem telling apart a cat from a dog. Just my opinion.
What if the service retrieves images by google search for "animal photo" and does post-processing on them, within boundaries of relatively easy human recognition. Google indexes images from real websites, where real people put and label photos, so this would make for a highly unpredictable set of images, given they are thus altered, even if a bot does index these same images too, since storing them is of no use. The key factor then becomes to comparing images from bot's index to service image, and given a sheer amount of images, i don't think the bot can be successful, even if it can compute a "similiarity percentage". Additionally, computers have trouble comparing visual data of low resolution. On top of that, since google presents images of different resolution, the bot would have to resize them. All this would be a quite a complex system for people behind spam. When spammers are outlawed enough, this throat should get really thin IMHO.
The only defense against this sort of attack would be to be constantly adding new images and removing old ones, but that would take more time than most people are willing to spend. It is absolutely no use adding and removing images to any database, or hash table. Most captcha images you see do not come from an image database, but are generated on the fly by software on demand randomly, which would make an nearly infinite hash-table, or at least a completely useless database, since even for any possible usual 4-letter character combination, the mathematical set of functions the software uses to distort and affect the image makes for a nearly infinite amount of variants.
Your hash table would have to be a distributed storage, and it would have to be an IPv6 network;-) or maybe not, i havent exactly computed the size of such database hehe
You are comparing apples to oranges. Spaniards were not to the Aztec, what emerging music distribution is to old fashioned music label corporations. Please do not make us weep for the poor record label suits who just want to put bread on their table;) Spaniards were aggressive and cared little for the native population compared to how much they cared for their gold which contributed greatly in their eyes to saving and promoting their empire (which was at war most of the time, hence needed all resources it could get its hands on). Eventually their worst side took over, and Aztec people were no more. Independent modern idealist musicians just want to make their little wages, drink beer and make music, while it is in vital interests of music publishers to find and exploit such people, offering little value to the deal themselves, apart from promised distribution, which many musicians do not even care about, yet having little choice of alternative, sign under. I see how compelling it is to draw parallels, but its fantasyland.
As to vanishing jobs, how about getting a real job. Desire [to make money doing abstract business] is not an occupation. Everyone can be a businessman given the right head, but business is so abstract of a concept, same people can put their talent to use in businesses of very different flavours and of different objectives.
Good blacksmiths, although fewer left, were never so valued as at this time. Handcraft is always in demand, only it's market changes.
As to other jobs, like chimney sweeping, well, it is the way life goes. You are thinking in terms of markets and jobs, but unfortunately, priorities and demands shift, and there is no reason to keep chimneysweepers around and paying them only because they can apparently do no other thing.
So, as you said, to the dustbin of history with all of it. Impermanence prevails.
Not sure what you mean by "information wants to be free" but RatioFaker appears to be ad-ware supported. Which means it infests the host with unwanted software that has nothing to do with it's own operation or goals. I would not touch anything like that with a 10-feet pole.
Why not just write 0s or 1s all over writeable area? I mean each and every sector on each track on each platter. Why all the grinding and shredding? Unless it is somehow possible to recover WIPED data, it should not be neccessary...
People who, for one reason or another like or know enough to do a driver update without smashing their machine to pieces, will always prefer PCs because PCs were and will stay to be the bleeding edge of hardware that drives all these games today. It is perhaps appropriate to call the whole PC gaming a sort of testing grounds for the future of gaming, and every 5 years or so, some manufacturer or another (MS, Sony, Nintendo at this time) decide to cement the testing grounds into a stable, non-volatile gaming platform that one can owe for more than a year and play games at without thinking about at least, graphic driver update. Nevertheless, the testing grounds that PCs are will remain, because there is a purpose to it. Another advantage is that since it is all testing, it is all bleeding edge, and most hardcore gamers breathe bleeding edge. Ever seen a 15 year old who knows everything about NVidia's roadmap for two years ahead, yet has hardly ever been intim with a female? I have.
You just can't expect computers to die as a gaming platform, because no matter how nice it is to have a non-changing console development platform that you don't have to update drivers for, and with which you can just have fun developing games, without worrying about drivers and funky crashes, version conflicts etc, it is still not an option to expect the gaming hardware market (which as most historians of the field know kick started and fueled 3d mathematics and algorithms, plus GPU design since abouit 1995 with the advent of 3dfx Voodoo, Riva and Rage chips) to freeze every 5 years, so that little kids can play their shiny little white PSx that site under their TV.
It is simply two parallel markets, and the only thing they share is the game industry.
You on the other hand, are truly gifted in the imagination department:-)
Because, in reality, you don't accidentally check in debug code in any builds, unless (1) you started programming yesterday or have lowest IQ of any programmer on Earth. Things like C/C++ preprocessor that help you with "#define DEBUG" (Most compilers, builders, and IDEs do it for you when you create building profiles, which most programmers do, if not - see (1)) and things like "#ifdef DEBUG" make it REALLY hard to check out debug builds. I understand though you were being sarcastic weren't you? It truly requires imagination to believe that excuse you wrote...;)
I agree. In any case, if you debug, it probably is faster for later checking to just write these usernames and passwords to a file on your local filesystem. Debugging where debug information goes to an email account is something entirely different an agenda.
The guy is a fraud. And whoever put up the message on the site knows it and pretends innocense. Weird, all this..
Has this guy John Terry never heard of compiling profiles, and a possibility of doing a '#define DEBUG' and testing if this is a debug scenario build? In C/C++ these facilities have been around for decades, and if MS thinks.NET applications are the future, I hope they have similiar functionality for.NET languages and or developing environments as well. Kind of childish of John Terry promising all this wonderful functionality yet lacking basic developer skills.
For me it rather looks like he had a debug and release profile, and it was malicious intent. You have to really be an amateur to not use #ifdef DEBUG for cases like this, and if he knew how to API GMail service, he appears to be smart enough to not accidentally have forgotten a debug code in the source.
As for the solution, and to all who argue whether we should likewise trust the likes of Mozilla's Thunderbird, Firefox etc, perhaps a logical step forward in innovation and an end to all this plain text password juggling would be a single OS-wide facility that will receive, and transport/delagate password and other sensitive information, as server that will expose it's API to client (software that accepts user password input). This way you won't have to speculate where does your password go after you type it in, a some form of manifest could lock software to letting only this OS-wide facility manage passwords on behalf of a user application that is of convenience to a user, and display a warning if the software violates it's security manifest by doing anything itself with decoded or plain-test input password.
The downside is that the facility will have to hold off attacks on itself, attacks that will try to compromise is, it being a trusted module, but we do trust a Linux kernel, why can't we trust a password delegation facility?
Granted both Linux and Windows have authentication facilities like this, but software DOES NOT have to use it, and user's do not know when their favourite software uses on or not. In Linux one has to review the source code (and do so for every MD5 signature change, to be sure nothing is compromised yet again), but users can't read source code, so the problem boils down to human availability, or to be more precise, availability of a developer per software version, who can assure you the software you use uses a generally trusted facility.
To sum up, we are still in stone age of information systems. It's amazing how much trust users learn to have to a piece of software they have never seen, how naive they must be to think anything they download and install from Internet is safe for their persona in one way or another. Is it perhaps that they trust their OS that much, they think it will protect them? It doesnt. And there is no reason to blame the users, the communication user-developer is really almost as bad as it was when computers were the size of a living room...Except everyone knows now what a 'blog' is, what is MySpace and Facebook, and what it means to be 'online':)
Stop talking nonsense, this issue happened to my brother yesterday, and I had to coach him by the telephone to do a System Restore. Its a goddamn mess, this Vista thing. Don't you tell me about a REAL thing. It was real for me.
How is that for a second hand account?
Of course, we are all retarded here, and you are one smart ass that has got it all figured out - Vista is great, Microsoft is misunderstood, and Mozilla Firefox developers hate to admit bugs. Riiiight.
I think you misunderstood me. The connection problem and all its implications can clearly be solved on the OS' part by falling back to IPv4. That would make a truly good OS. I said "for users" because indeed to them it's just an error, and I consider it blaming on the users if a lazy piece of software spits out something like "IPv6 host not found.", which makes no sense to at least half of the users. And all that stands between such error messages and experiences that just function is something about 15 to 50 lines of C code. Granted, "missing functionality" as rather ambiguous expression, as it is something very debated by those implementing software, but i belive this case calls for exactly that expression. I mean company like MS, with all those funds, and the money it spends on most incredible useless things, you GOT to agree they could have done it - an experience that just works. Only this at least, this little thing. Don't take me out of context here, but the answer to your question in this very case is YES.
I think if both protocol stacks are present, with IPv6 doing the job first, they really have no choice. It is either fallback or connection error. And connection error is for users a missing functionality.
There is not timeout, as nothing needs to time out, simply a missing address record error, so no need to really _wait_ for anything.
Besides, if the users do not know what a protocol is, much less how to disable IPv6, they have even less options. Which is why a consumer friendly company would program a fallback and not discuss it too much. Users do not care, remember?
I think you disregard the fact that Microsoft, unlike whatever it's PR machine and spokespeople say, does NOT care much for it's customers, as unlogical as it sounds. The company is too big and has grown to include too many influential facets in commercial investors of all important sorts, lobbyists, and just people wanting to make a fortune off the software giant. At this point, Microsoft is ruled and directed by those who have power, consumers come second. Vista is a good display of that. In the beginning a lot of features were promised for the consumer, that indeed made sense, like WinFS. This was done to attact attention of users, and keep their interest while Vista was under development all those years. It was like a concept car at a car show. It felt real, but not really available, as the manufacturer was still in the process of making it. When the promised release date, after promising too much and too long, started to hang above our heads, Microsoft had to unveil the real Vista - that which left out the more important features, left in all that is eye candy, and all the features that were lobbyied in by various power players in the industry. Money had been invested, hands were shaken at meetings, and all this time the people who Vista depended on for survival and prosperity - users - were watching a show where the star turned out to be quite a different person they thought it to be. IPv6, regardless of it's non commercial nature as a protocol and idea (survival of the Internet as we know it), relates to a whole lot of industry sectors, with considerable sums of money invested by Cisco for example. Routers have to be manufactures, and those who manufacture them deliver the Internet backbone, and all the money they sit on - it has to flow. 96 or so percent of OS market means that money, properly invested in Vista, is money VERY WELL spent, and router manufacturers wanting to sell new hardware know that. If Microsoft wanted (with 1500 or so good software engineers, albeit at a wrong job IMO, i am more than sure it could) with about 15-50 lines of code make IPv6 connection timeout fallback to IPv4 lookup, it would. But it does not really want to does it? After all, if all hosts that answer to Vista connection requests of those 96 percent of users, can do away with IPv4, when will IPv6 come ? According to ISP's and Internet hardware manufacturers, not soon enough. Unfortunately, it's only geeks like us who tend to see the world in 1 and 0s. Most people do not really care, they just want to make money to put food on the table. And IPv6, as far fetched as it sounds, is very closely related to food at this point.
Slashdot Mirror
Sounds like a good idea. Cache is a very good concept, that has proved itself across and in different systems.
Basically in an environment where size matters less than speed, cache brings benefits.
i.e. IMO, cache servers would do Internet good as a whole, provided they are:
1. Enviromentally competitive - draw little power, so they do not contribute to a whole lot of watts Internet contributes to already. I am talking about inexpensive systems with lots of cheap hard drive space. A lot of cache server systems to 'speed up' Internet means a lot of power draw, so obviously a factor.
2. Legal P2P content? Oh, lets play police again shall we? 90% of p2p traffic is illegal, so how caching 10% of the part that is legal is a solution? Besides, is encrypted traffic illegal? Speaking in Apache terms, is it an 'allow-deny' or a 'deny-allow' rule?
Perhaps use routers with a hard drive. That way it is a polymorph system, that acts as a good old TCP/IP router, but may retrieve content from cache, if the connection allows it. Replacing old routers with such upgraded systems can be done over time.
A cheap harddrive is what, a 5W? It only needs to be faster than cache-miss data retrieval equivalent, right?
I think Mr. Gates should finally step off his self-raised piedestal of fame and grandeure.
Somehow it is so human and unfortunately, in my opinion, completely wrong, to assume a company is success because of one single reason, or a small set of key reasons, that a chairman consciously can express. There is a whole lot of history behind Microsoft, and some things that happened in the past, at the right time, and the right place, helped it to become what it is. No more no less. Granted Microsoft started a whole market of consumer computerism, but it quickly monopolized it like a big predator that just killed a big meaty prey. A wise person would not make hasty conclusions, and instead, tell Microsoft for what it is - an opportunist, a scavenger, a conman, yet with a charm and blessing of efficiency, modest competency and brains. Last being only the merit of smart geeks in search of a job. Many people would do anything for food. Why not sit in the office at Redmond and type away chunks of IE code? It's not peer reviewed anyway, and everyone around cheers you up telling you how good programmer you are and how you contribute to the better IT of the planet. Talk about sandbox-career.
The fact is, today, Microsoft is by no means the most innovative, successful, inspiring, and competent company. Multitude of recent facts related to them stand as proof. Things like OOXML vote rigging, Vista- disaster (and for people who actually took time to read Microsoft blogs years prior to the actual release, a disaster even worse), all their semi-tech talk about features, that end up vaporware. It has, and is a big circus. The sad fact is anyone can succeed, given enough programmers, and a aggressive strategy. It's business. And the rules of engagement in business allow and encourage such strategies. It does not mean products are any good. Given the absence of choice, they might be. But look at IE for instance. Self-crowned jewel of Internet Surfing for Average Joe, until some people decided to show the world how bad it really was - inventing Firefox.
Consumerism appreciates Microsoft. Awareness and modern market does not. A market where software is peer-reviewed, criticized, praised, compared to, dissected as source code, and what not. In the midst of it all, Microsoft tries to play it like it is still 80's. Where you buy a floppy disk worth 100$ of software, bring it home and plug it into your MS-DOS, typing away obscure stuff at command line, on which you read in a equally expensive "Learn Computers" book.
Sadly, now a whole lot of young aspiring managers heed to the Message of Gates, an aging billionaire with supposed infinite knowledge of IT. I doubt he is one. He is just one hell of a CEO.
True innovation, inspiration and science are whole another field. But then again, let us not confuse science with good old marketing, which Microsoft excels at. In fact it should have been called Marketsoft. What does Micro- stand for again??
And of course, i have no idea how to market stuff. But a better product needs less marketing. And I am talking about science.
REALLY good point! The only downside is human factor, of course, - will enough people be willing to spend precious minutes of their life to do take this kind of procedure?
How about this one: http://images.google.com/images?q=cat
Telling apart animals is CONSIDERABLY different than telling apart symbols, in my opinion. Given sufficiently low resolution, sufficiently bad lightning and focus, even you will have a problem telling apart a cat from a dog. Just my opinion.
What if the service retrieves images by google search for "animal photo" and does post-processing on them, within boundaries of relatively easy human recognition. Google indexes images from real websites, where real people put and label photos, so this would make for a highly unpredictable set of images, given they are thus altered, even if a bot does index these same images too, since storing them is of no use. The key factor then becomes to comparing images from bot's index to service image, and given a sheer amount of images, i don't think the bot can be successful, even if it can compute a "similiarity percentage". Additionally, computers have trouble comparing visual data of low resolution. On top of that, since google presents images of different resolution, the bot would have to resize them. All this would be a quite a complex system for people behind spam. When spammers are outlawed enough, this throat should get really thin IMHO.
All your fragmented heap are belong to us now.
Oh, they suck BIG time!
You are comparing apples to oranges. Spaniards were not to the Aztec, what emerging music distribution is to old fashioned music label corporations. Please do not make us weep for the poor record label suits who just want to put bread on their table ;) Spaniards were aggressive and cared little for the native population compared to how much they cared for their gold which contributed greatly in their eyes to saving and promoting their empire (which was at war most of the time, hence needed all resources it could get its hands on). Eventually their worst side took over, and Aztec people were no more. Independent modern idealist musicians just want to make their little wages, drink beer and make music, while it is in vital interests of music publishers to find and exploit such people, offering little value to the deal themselves, apart from promised distribution, which many musicians do not even care about, yet having little choice of alternative, sign under. I see how compelling it is to draw parallels, but its fantasyland.
As to vanishing jobs, how about getting a real job. Desire [to make money doing abstract business] is not an occupation. Everyone can be a businessman given the right head, but business is so abstract of a concept, same people can put their talent to use in businesses of very different flavours and of different objectives.
Good blacksmiths, although fewer left, were never so valued as at this time. Handcraft is always in demand, only it's market changes.
As to other jobs, like chimney sweeping, well, it is the way life goes. You are thinking in terms of markets and jobs, but unfortunately, priorities and demands shift, and there is no reason to keep chimneysweepers around and paying them only because they can apparently do no other thing.
So, as you said, to the dustbin of history with all of it. Impermanence prevails.
Not sure what you mean by "information wants to be free" but RatioFaker appears to be ad-ware supported. Which means it infests the host with unwanted software that has nothing to do with it's own operation or goals. I would not touch anything like that with a 10-feet pole.
OK, thanks for the explanation.
Why not just write 0s or 1s all over writeable area? I mean each and every sector on each track on each platter. Why all the grinding and shredding? Unless it is somehow possible to recover WIPED data, it should not be neccessary...
People who, for one reason or another like or know enough to do a driver update without smashing their machine to pieces, will always prefer PCs because PCs were and will stay to be the bleeding edge of hardware that drives all these games today. It is perhaps appropriate to call the whole PC gaming a sort of testing grounds for the future of gaming, and every 5 years or so, some manufacturer or another (MS, Sony, Nintendo at this time) decide to cement the testing grounds into a stable, non-volatile gaming platform that one can owe for more than a year and play games at without thinking about at least, graphic driver update. Nevertheless, the testing grounds that PCs are will remain, because there is a purpose to it. Another advantage is that since it is all testing, it is all bleeding edge, and most hardcore gamers breathe bleeding edge. Ever seen a 15 year old who knows everything about NVidia's roadmap for two years ahead, yet has hardly ever been intim with a female? I have.
You just can't expect computers to die as a gaming platform, because no matter how nice it is to have a non-changing console development platform that you don't have to update drivers for, and with which you can just have fun developing games, without worrying about drivers and funky crashes, version conflicts etc, it is still not an option to expect the gaming hardware market (which as most historians of the field know kick started and fueled 3d mathematics and algorithms, plus GPU design since abouit 1995 with the advent of 3dfx Voodoo, Riva and Rage chips) to freeze every 5 years, so that little kids can play their shiny little white PSx that site under their TV.
It is simply two parallel markets, and the only thing they share is the game industry.
You on the other hand, are truly gifted in the imagination department :-)
;)
Because, in reality, you don't accidentally check in debug code in any builds, unless (1) you started programming yesterday or have lowest IQ of any programmer on Earth. Things like C/C++ preprocessor that help you with "#define DEBUG" (Most compilers, builders, and IDEs do it for you when you create building profiles, which most programmers do, if not - see (1)) and things like "#ifdef DEBUG" make it REALLY hard to check out debug builds. I understand though you were being sarcastic weren't you? It truly requires imagination to believe that excuse you wrote...
I agree. In any case, if you debug, it probably is faster for later checking to just write these usernames and passwords to a file on your local filesystem. Debugging where debug information goes to an email account is something entirely different an agenda.
The guy is a fraud. And whoever put up the message on the site knows it and pretends innocense. Weird, all this..
Has this guy John Terry never heard of compiling profiles, and a possibility of doing a '#define DEBUG' and testing if this is a debug scenario build? In C/C++ these facilities have been around for decades, and if MS thinks .NET applications are the future, I hope they have similiar functionality for .NET languages and or developing environments as well. Kind of childish of John Terry promising all this wonderful functionality yet lacking basic developer skills.
:)
For me it rather looks like he had a debug and release profile, and it was malicious intent. You have to really be an amateur to not use #ifdef DEBUG for cases like this, and if he knew how to API GMail service, he appears to be smart enough to not accidentally have forgotten a debug code in the source. As for the solution, and to all who argue whether we should likewise trust the likes of Mozilla's Thunderbird, Firefox etc, perhaps a logical step forward in innovation and an end to all this plain text password juggling would be a single OS-wide facility that will receive, and transport/delagate password and other sensitive information, as server that will expose it's API to client (software that accepts user password input). This way you won't have to speculate where does your password go after you type it in, a some form of manifest could lock software to letting only this OS-wide facility manage passwords on behalf of a user application that is of convenience to a user, and display a warning if the software violates it's security manifest by doing anything itself with decoded or plain-test input password.
The downside is that the facility will have to hold off attacks on itself, attacks that will try to compromise is, it being a trusted module, but we do trust a Linux kernel, why can't we trust a password delegation facility?
Granted both Linux and Windows have authentication facilities like this, but software DOES NOT have to use it, and user's do not know when their favourite software uses on or not. In Linux one has to review the source code (and do so for every MD5 signature change, to be sure nothing is compromised yet again), but users can't read source code, so the problem boils down to human availability, or to be more precise, availability of a developer per software version, who can assure you the software you use uses a generally trusted facility.
To sum up, we are still in stone age of information systems. It's amazing how much trust users learn to have to a piece of software they have never seen, how naive they must be to think anything they download and install from Internet is safe for their persona in one way or another. Is it perhaps that they trust their OS that much, they think it will protect them? It doesnt. And there is no reason to blame the users, the communication user-developer is really almost as bad as it was when computers were the size of a living room...Except everyone knows now what a 'blog' is, what is MySpace and Facebook, and what it means to be 'online'
Stop talking nonsense, this issue happened to my brother yesterday, and I had to coach him by the telephone to do a System Restore. Its a goddamn mess, this Vista thing. Don't you tell me about a REAL thing. It was real for me. How is that for a second hand account? Of course, we are all retarded here, and you are one smart ass that has got it all figured out - Vista is great, Microsoft is misunderstood, and Mozilla Firefox developers hate to admit bugs. Riiiight.
I think you misunderstood me. The connection problem and all its implications can clearly be solved on the OS' part by falling back to IPv4. That would make a truly good OS. I said "for users" because indeed to them it's just an error, and I consider it blaming on the users if a lazy piece of software spits out something like "IPv6 host not found.", which makes no sense to at least half of the users. And all that stands between such error messages and experiences that just function is something about 15 to 50 lines of C code. Granted, "missing functionality" as rather ambiguous expression, as it is something very debated by those implementing software, but i belive this case calls for exactly that expression. I mean company like MS, with all those funds, and the money it spends on most incredible useless things, you GOT to agree they could have done it - an experience that just works. Only this at least, this little thing. Don't take me out of context here, but the answer to your question in this very case is YES.
I think if both protocol stacks are present, with IPv6 doing the job first, they really have no choice. It is either fallback or connection error. And connection error is for users a missing functionality.
There is not timeout, as nothing needs to time out, simply a missing address record error, so no need to really _wait_ for anything.
Besides, if the users do not know what a protocol is, much less how to disable IPv6, they have even less options. Which is why a consumer friendly company would program a fallback and not discuss it too much. Users do not care, remember?
Just my opinion..
I think you disregard the fact that Microsoft, unlike whatever it's PR machine and spokespeople say, does NOT care much for it's customers, as unlogical as it sounds. The company is too big and has grown to include too many influential facets in commercial investors of all important sorts, lobbyists, and just people wanting to make a fortune off the software giant. At this point, Microsoft is ruled and directed by those who have power, consumers come second. Vista is a good display of that. In the beginning a lot of features were promised for the consumer, that indeed made sense, like WinFS. This was done to attact attention of users, and keep their interest while Vista was under development all those years. It was like a concept car at a car show. It felt real, but not really available, as the manufacturer was still in the process of making it. When the promised release date, after promising too much and too long, started to hang above our heads, Microsoft had to unveil the real Vista - that which left out the more important features, left in all that is eye candy, and all the features that were lobbyied in by various power players in the industry. Money had been invested, hands were shaken at meetings, and all this time the people who Vista depended on for survival and prosperity - users - were watching a show where the star turned out to be quite a different person they thought it to be. IPv6, regardless of it's non commercial nature as a protocol and idea (survival of the Internet as we know it), relates to a whole lot of industry sectors, with considerable sums of money invested by Cisco for example. Routers have to be manufactures, and those who manufacture them deliver the Internet backbone, and all the money they sit on - it has to flow. 96 or so percent of OS market means that money, properly invested in Vista, is money VERY WELL spent, and router manufacturers wanting to sell new hardware know that. If Microsoft wanted (with 1500 or so good software engineers, albeit at a wrong job IMO, i am more than sure it could) with about 15-50 lines of code make IPv6 connection timeout fallback to IPv4 lookup, it would. But it does not really want to does it? After all, if all hosts that answer to Vista connection requests of those 96 percent of users, can do away with IPv4, when will IPv6 come ? According to ISP's and Internet hardware manufacturers, not soon enough. Unfortunately, it's only geeks like us who tend to see the world in 1 and 0s. Most people do not really care, they just want to make money to put food on the table. And IPv6, as far fetched as it sounds, is very closely related to food at this point.