I'm sure part of the reason for having their own license is simply NIH. Rather than trying to get an existing license approved they simply asked their lawyers to come up with a license they are comfortable with.
IANAL, but I can also imagine that they want their own license to insulate themselves from future evolution of other open source licenses. As the parent correctly notes, they could have picked GPLv2 without the "or later version" clause and that would have locked in their choice of license. But I am thinking of a more subtle effect - if a body of case law comes along that sets lots of precedents on how the GPL is interpreted, MS may find themselves in the position of distributing things under a license they don't like.
By using their own license the legal ecosystem is smaller, and their lawyers have a better chance to argue any cases that do come up.
To be price competitive, yes. To be competitive, no.
In the Minneapolis area there are still a number of boutique system builders. Walk in, pick out a motherboard, case, processor, memory, etc., and either take it home or watch them build it for you.
This is good on several levels. First, these people actually know what they are doing, and are capable of doing diagnostics and repairs. Second, a system you get from them is not bogged down with craplets and shovelware. Third is the whole immediate gratification factor.
I can buy a brand new system cheaper from Dell than I can from these guys, but it won't have exactly the components I want, and if I am reusing some components of an existing system the local guys will come out ahead. They don't even charge me anything to integrate my existing components for reuse.
There is still a market for the personalized service you get. Price is not everything, or Linux would have won by now.
I think it is too early to condemn the Vista adoption rate, for the simple fact that very few businesses are going to jump on a new release as soon as it comes out. Vista has only been in full release for 6 months at this point, and the places with the really big user bases are going to be very cautious in their rollout plans. At this point I wouldn't expect the GMs and GEs of the world to either roll it out company wide or even allow it to remain on new units that they bring in the door.
Give it another year and then I think you can legitimately say that Vista adoption is seriously lagging the growth of the market.
If we ignore the environmental burden caused by printing and delivering snail mail spam, I find it much less obnoxious than email spam for a number of reasons:
1. It comes in once a day, and I can sort it in a few seconds, as opposed to trickling in all day long and distracting me. 2. Since it has significant costs to send, it is almost never as blatantly stupid as most of the spam emails I get. 3. Since the post office does investigate mail fraud (at least in the US), most of the offers may be stupid, but they are usually legitimate.
And really, snail mail spam is an invasion of your privacy? Care to explain that one? If your privacy bar is set so high that a piece of mail dropping in a box counts as a significant imposition how do you handle walking down the street or using public transportation? Wouldn't someone actually being able to see you be far more of an invasion of your privacy?
From reading the various replies on this thread, it has become quite apparent that Comcast equipment/software/policies vary by region. Indeed, given how many customers they picked up in acquisitions, that is almost certainly the case. I was with Time Warner for years, and then they sold out to Comcast in my region. Comcast continued to operate the cable modem initially supplied by Roadrunner. It was installed a long time ago, but I don't recall any requirement for MAC authorization. They left a CD with an installer, but since Win 2K immediately detected the internet connection I never used it.
Recently I upgraded to their VOIP service as well, which required a new modem. I don't have a computer anywhere near the modem, and didn't get any requests to bypass my router or indeed run anything. He was on a walky-talky for a few minutes, then everything was done.
When I bought my last car they dealer conceded to selling it for a price lower than what was shown on the sticker.
How is MS offering a discount/incentive/license concession any different? Some MS sales rep had a potential sale of 500 seats, and had to sweeten the deal to get a sale. Purchasing people are always pushing for a better deal, and threatening to take their business elsewhere if they don't get it.
Actually, in most cases I would guess that it gets taken care of with an exchange of letters.
A sends B a letter saying "you infringe on patent xyz..."
B then: 1. Ignores the letter, essentially daring them to go to court over it. 2. Fires back with an explanation of how their implementation is non-infringing because... 3. Fires back with a listing of prior art that strongly implies the patent is non-enforceable.
In case 3, company A has a strong incentive to go away quietly and just go on to the next company in the list. Company B just wants the whole thing to go away and so never discloses to others the prior art they dug up.
Company A crosses B from the list of potential licensees and looks for someone that didn't do their research carefully enough.
I had something of the same reaction to your remark about a patent. This is a fairly good example of how hard the "obviousness" test of a patent can be to judge. When you hear about this, it is something of a "Doh!" moment, and you think how obvious it is. (I was immediately reminded of the chroma subsampling options in JPEG compression, which use different sampling rates for color and luminosity).
But the fact is that hundreds of millions of digital cameras have been made in an intensely competitive R&D environment, and only now is this innovation coming up. That argues that it is indeed non-obvious, or someone would have come up with it sooner.
Yet if this becomes popular and patented, many people are going to think it was obvious all along.
Methinks you have never worked for a big company where the bottom line is all that matters. If you can settle a lawsuit for $10K and defending it will cost $6K with a chance of getting hit with a judgment of $30K, they will settle virtually every time, often to the intense annoyance of individuals named in the suit.
They note that people were arrested, hence it is a criminal offense.
By your logic, until all murders and rapes were solved the police would never look at speeding, vandalism, theft, fraud, etc.
While I don't care for the RIAA, I don't see anything here that justifies outrage. The RIAA care about a particular crime being committed, spend some of their money on the investigation, and go to the cops and ask them to make the bust. Kind of like if my bike was stolen and I see it in somebody's garage on the other side of town. I tell the cops and they hopefully follow up on my tip.
In many ways, that doesn't really matter. Most organizations are sticking with XP for the moment, but the fact that they have Vista licenses to go with all the new boxes means that they can decide to pull the trigger on the update whenever they want.
If 1 year from now they decide to migrate from XP to Vista, and 1/3 of their PCs already have Vista licenses, the cost of the migration goes down, making it more likely.
Despite the groupthink here on/., I am not seeing anything that makes me believe corporations are plannning mass migrations to *nix/Mac OS. Barring such a migration, moving to Vista is more a question of when than of if.
If they have a problem with AutoCAD, and they can bring that kind of numbers to bear on the problem, don't you think that AutoDesk (or whomever makes AutoCAD) is going to be fairly solicitous of their needs?
The people that sell AutoCAD want to increase sales, so will tend to implement whatever large groups of people are asking for. If the group is large enough that their defection would be a significant hit to the market leader, the dollars to be made satisfying that market are also significant.
The harm and the foul is not just (or even primarily) to Adobe or Microsoft. By being willing to pirate (or infringe copyright if you must be pedantic) software, you are harming the market for alternatives to the Adobe and Microsoft products.
So if you aren't willing to pay for Photoshop, why aren't you looking at the competing products that cost much less? If the market leading app costs X, you would assume that a natural market exists for an app that costs X/10 (or in the case of FOSS, 0). Saying there is no harm because you would never pay X ignores that fact that you have harmed the vendor of the X/10 product.
1500 pixels square? As in an image of 1500 x 1500 pixels?
Lets do the math
1500 x 1500 pixels = 2.25 million pixels 4 bytes per pixel = 9MB per layer
30 layers = 270MB of image data.
That doesn't count memory consumed by the undo system, which can quickly get very large.
Plus the amount consumed by Photoshop itself.
Plus the fact that rather than composite 30 layers on the fly whenever a window is invalidated, there is undoubtedly some amount of paint caching going on, probably the equivalent of several more layers worth of data.
The 120MB that the file consumes when stored in some compressed format on disk is just the tip of the iceberg here.
And how many hundreds of thousands of lines of code are embedded in that Apache? Almost certainly it has at least some level of fly by wire technology, so a computer is moderating the flight control inputs. The radar/targeting/fire control system is going to have lots of software - what do you think draws the targeting pippers you see in war footage?
Even at the level of the engine there is likely to be an engine management chip analogous to the one in your car.
While I absolutely agree with your argument, the example you chose is not a particularly good one.
Movie studios have found that by spending lots of money on promotion of a movie they can almost guarantee themselves a good opening weekend. Nobody has seen the movie yet, just the splashy trailer and all the commercials. They go see it, it sucks, and by Monday everyone knows it is a steaming pile of feculent trash. The second week sales go to zero, and by the 3rd week the movie is gone.
If a movie is not given to reviewers in advance, the studio knows it is a dog and their only chance to get any money is opening weekend.
So in your example, I would say that any movie that brings in millions of dollars the second weekend is bringing in money because a significant number of people like it and are willing to pay money to see it.
I can well believe you with respect to the linearity of RAM usage. I am not claiming anything in particular about the behavior of Windows in situations where it has to treat RAM as a scarce resource.
What I am saying, is that so far as the O/S is concerned, there is no downside to using more RAM, up until the point where it runs out. If the system decides that it can benefit from having 1/2GB of disk cache, and no one else has asked for that RAM, why not let it have the cache? The RAM isn't being used for anything anyway.
My point was rather narrow - RAM that is not in use does nothing for you, and if by using it the system could prevent a page fault later, it may as well use it. Hence my contention that RAM usage in the idle state need not be a good predictor of O/S bloat or how memory constrained the system really is.
Once demand for RAM heats up and the system has to determine how to parcel it out to the running tasks is where things get interesting, and where a well designed memory manager will make a big difference.
It is kind of like screen real estate. You may not need your browser maximized to read/., but if your eyes aren't looking at anything else there is no downside to maximizing. It is only when you try to do that and watch a compile and watch a news feed that you have to decide how to apportion display space.
Wanting to have free memory makes little sense. RAM is a resource that the system should use whenever it gets a benefit from doing so. If by using more RAM it can increase working sets, disk caches, display caches, whatever, it should do so.
RAM that is not in use is a resource that you paid for that you are not getting any value from. The only time RAM should be free is if the O/S can't think of anything useful for it to do.
The hard part comes when the O/S has to figure out how to sensibly allocate RAM with the system under load. That is a hard problem, and I make no claims to know what it should do. But looking at RAM usage in the idle state is not a good indication of how the system will perform under load.
I fear that would likely still run in to antitrust issues, because it amounts to dictating the bundling arrangements.
I wonder if they could get away with simply demanding that a vanilla Vista CD be included with the box, so that the customer could do a reinstall if they wanted. It could even be in addition to the OEMs recovery CD, which would put all the craplets back on. But if you had to have the vanilla CD a reformat/reinstall would at least be feasible.
A backdoor is indeed possible, and given how robust the current encryption algorithms are the NSA would no doubt be happy to have a backdoor.
But don't forget the corollary - many of the systems running Vista are going to be systems that the NSA wants to be secure. There are lots of Windows systems in government and the military.
If the NSA inserts any kind of backdoor they also need to worry about how it could be exploited by a hostile power. Inserting a backdoor, no matter how secret, compromises the security of the system. Would they find it an acceptable tradeoff? I have no idea.
Semi-related anecdote: In WWII, the German Enigma cipher machine used a series of rotors as part of the encode/decode process. The rotors had different possible positions, indicated by letters. So part of the encoding process for a day was the rotor position for each of 3 rotors, such as ABC. A rotor position like AAC effectively turned a 3 rotor machine into a 2 rotor machine. So someone in the German command structure ordered that they not use any daily settings with a repeated position.
The codebreakers at Bletchley Park (this might be early enough that it was still the Polish equivalent of Bletchley) were able to use that simple fact as an attack vector, and it was one of the initial weaknesses they discovered in the Enigma.
Moral of the story: anything you do that reduces the randomness of a crypto system weakens it. A backdoor is a huge decrease in randomness.
I've never seen confirmation that MS apps make any significant use of non-documented OS APIs. The Office group writes much of their own code to be sure, but most of the big players do that.
It is easy enough to use a dependency checker and find all the symbols that a program imports from a DLL. If you cross reference the imports with the documentation in MSDN, it is easy to spot something that is not documented. Given all the axes to grind out there, I can't believe someone hasn't done this already, and I dont recall reading about all the incriminating evidence that was found.
A more plausible claim, and one much harder to prove or disprove, is that the Office team has access to Windows source code, so that rather than creating something from scratch they can just grab a copy of the menuing code and create their own version.
Slashdot Mirror
I'm sure part of the reason for having their own license is simply NIH. Rather than trying to get an existing license approved they simply asked their lawyers to come up with a license they are comfortable with.
IANAL, but I can also imagine that they want their own license to insulate themselves from future evolution of other open source licenses. As the parent correctly notes, they could have picked GPLv2 without the "or later version" clause and that would have locked in their choice of license. But I am thinking of a more subtle effect - if a body of case law comes along that sets lots of precedents on how the GPL is interpreted, MS may find themselves in the position of distributing things under a license they don't like.
By using their own license the legal ecosystem is smaller, and their lawyers have a better chance to argue any cases that do come up.
I've used Tran Micro on University Ave a few times, and have always been very satisfied.
I haven't tried them, but have heard similar good things about a General Nanosystems, just a block or two away.
To be price competitive, yes. To be competitive, no.
In the Minneapolis area there are still a number of boutique system builders. Walk in, pick out a motherboard, case, processor, memory, etc., and either take it home or watch them build it for you.
This is good on several levels. First, these people actually know what they are doing, and are capable of doing diagnostics and repairs. Second, a system you get from them is not bogged down with craplets and shovelware. Third is the whole immediate gratification factor.
I can buy a brand new system cheaper from Dell than I can from these guys, but it won't have exactly the components I want, and if I am reusing some components of an existing system the local guys will come out ahead. They don't even charge me anything to integrate my existing components for reuse.
There is still a market for the personalized service you get. Price is not everything, or Linux would have won by now.
I read the article this morning, and considered submitting it myself. For a tech site like Ars I thought the article was really very evenhanded.
/., but really, the article is worth reading.
I know this is
I think it is too early to condemn the Vista adoption rate, for the simple fact that very few businesses are going to jump on a new release as soon as it comes out. Vista has only been in full release for 6 months at this point, and the places with the really big user bases are going to be very cautious in their rollout plans. At this point I wouldn't expect the GMs and GEs of the world to either roll it out company wide or even allow it to remain on new units that they bring in the door.
Give it another year and then I think you can legitimately say that Vista adoption is seriously lagging the growth of the market.
If we ignore the environmental burden caused by printing and delivering snail mail spam, I find it much less obnoxious than email spam for a number of reasons:
1. It comes in once a day, and I can sort it in a few seconds, as opposed to trickling in all day long and distracting me.
2. Since it has significant costs to send, it is almost never as blatantly stupid as most of the spam emails I get.
3. Since the post office does investigate mail fraud (at least in the US), most of the offers may be stupid, but they are usually legitimate.
And really, snail mail spam is an invasion of your privacy? Care to explain that one? If your privacy bar is set so high that a piece of mail dropping in a box counts as a significant imposition how do you handle walking down the street or using public transportation? Wouldn't someone actually being able to see you be far more of an invasion of your privacy?
From reading the various replies on this thread, it has become quite apparent that Comcast equipment/software/policies vary by region. Indeed, given how many customers they picked up in acquisitions, that is almost certainly the case. I was with Time Warner for years, and then they sold out to Comcast in my region. Comcast continued to operate the cable modem initially supplied by Roadrunner. It was installed a long time ago, but I don't recall any requirement for MAC authorization. They left a CD with an installer, but since Win 2K immediately detected the internet connection I never used it.
Recently I upgraded to their VOIP service as well, which required a new modem. I don't have a computer anywhere near the modem, and didn't get any requests to bypass my router or indeed run anything. He was on a walky-talky for a few minutes, then everything was done.
When I bought my last car they dealer conceded to selling it for a price lower than what was shown on the sticker.
How is MS offering a discount/incentive/license concession any different? Some MS sales rep had a potential sale of 500 seats, and had to sweeten the deal to get a sale. Purchasing people are always pushing for a better deal, and threatening to take their business elsewhere if they don't get it.
Given that the developer willingly sold, calling it a hostile takeover seems a stretch.
Is Apple showing its hostility by attempting to drown him in dollar bills?
Actually, in most cases I would guess that it gets taken care of with an exchange of letters.
A sends B a letter saying "you infringe on patent xyz..."
B then:
1. Ignores the letter, essentially daring them to go to court over it.
2. Fires back with an explanation of how their implementation is non-infringing because...
3. Fires back with a listing of prior art that strongly implies the patent is non-enforceable.
In case 3, company A has a strong incentive to go away quietly and just go on to the next company in the list. Company B just wants the whole thing to go away and so never discloses to others the prior art they dug up.
Company A crosses B from the list of potential licensees and looks for someone that didn't do their research carefully enough.
That scenario plays out every day.
I had something of the same reaction to your remark about a patent. This is a fairly good example of how hard the "obviousness" test of a patent can be to judge. When you hear about this, it is something of a "Doh!" moment, and you think how obvious it is. (I was immediately reminded of the chroma subsampling options in JPEG compression, which use different sampling rates for color and luminosity).
But the fact is that hundreds of millions of digital cameras have been made in an intensely competitive R&D environment, and only now is this innovation coming up. That argues that it is indeed non-obvious, or someone would have come up with it sooner.
Yet if this becomes popular and patented, many people are going to think it was obvious all along.
Methinks you have never worked for a big company where the bottom line is all that matters. If you can settle a lawsuit for $10K and defending it will cost $6K with a chance of getting hit with a judgment of $30K, they will settle virtually every time, often to the intense annoyance of individuals named in the suit.
They note that people were arrested, hence it is a criminal offense.
By your logic, until all murders and rapes were solved the police would never look at speeding, vandalism, theft, fraud, etc.
While I don't care for the RIAA, I don't see anything here that justifies outrage. The RIAA care about a particular crime being committed, spend some of their money on the investigation, and go to the cops and ask them to make the bust. Kind of like if my bike was stolen and I see it in somebody's garage on the other side of town. I tell the cops and they hopefully follow up on my tip.
In many ways, that doesn't really matter. Most organizations are sticking with XP for the moment, but the fact that they have Vista licenses to go with all the new boxes means that they can decide to pull the trigger on the update whenever they want.
/., I am not seeing anything that makes me believe corporations are plannning mass migrations to *nix/Mac OS. Barring such a migration, moving to Vista is more a question of when than of if.
If 1 year from now they decide to migrate from XP to Vista, and 1/3 of their PCs already have Vista licenses, the cost of the migration goes down, making it more likely.
Despite the groupthink here on
If they have a problem with AutoCAD, and they can bring that kind of numbers to bear on the problem, don't you think that AutoDesk (or whomever makes AutoCAD) is going to be fairly solicitous of their needs?
The people that sell AutoCAD want to increase sales, so will tend to implement whatever large groups of people are asking for. If the group is large enough that their defection would be a significant hit to the market leader, the dollars to be made satisfying that market are also significant.
The harm and the foul is not just (or even primarily) to Adobe or Microsoft. By being willing to pirate (or infringe copyright if you must be pedantic) software, you are harming the market for alternatives to the Adobe and Microsoft products.
So if you aren't willing to pay for Photoshop, why aren't you looking at the competing products that cost much less? If the market leading app costs X, you would assume that a natural market exists for an app that costs X/10 (or in the case of FOSS, 0). Saying there is no harm because you would never pay X ignores that fact that you have harmed the vendor of the X/10 product.
1500 pixels square? As in an image of 1500 x 1500 pixels?
Lets do the math
1500 x 1500 pixels = 2.25 million pixels
4 bytes per pixel = 9MB per layer
30 layers = 270MB of image data.
That doesn't count memory consumed by the undo system, which can quickly get very large.
Plus the amount consumed by Photoshop itself.
Plus the fact that rather than composite 30 layers on the fly whenever a window is invalidated, there is undoubtedly some amount of paint caching going on, probably the equivalent of several more layers worth of data.
The 120MB that the file consumes when stored in some compressed format on disk is just the tip of the iceberg here.
And how many hundreds of thousands of lines of code are embedded in that Apache? Almost certainly it has at least some level of fly by wire technology, so a computer is moderating the flight control inputs. The radar/targeting/fire control system is going to have lots of software - what do you think draws the targeting pippers you see in war footage?
Even at the level of the engine there is likely to be an engine management chip analogous to the one in your car.
Does that software not trouble you at all?
While I absolutely agree with your argument, the example you chose is not a particularly good one.
Movie studios have found that by spending lots of money on promotion of a movie they can almost guarantee themselves a good opening weekend. Nobody has seen the movie yet, just the splashy trailer and all the commercials. They go see it, it sucks, and by Monday everyone knows it is a steaming pile of feculent trash. The second week sales go to zero, and by the 3rd week the movie is gone.
If a movie is not given to reviewers in advance, the studio knows it is a dog and their only chance to get any money is opening weekend.
So in your example, I would say that any movie that brings in millions of dollars the second weekend is bringing in money because a significant number of people like it and are willing to pay money to see it.
I can well believe you with respect to the linearity of RAM usage. I am not claiming anything in particular about the behavior of Windows in situations where it has to treat RAM as a scarce resource.
/., but if your eyes aren't looking at anything else there is no downside to maximizing. It is only when you try to do that and watch a compile and watch a news feed that you have to decide how to apportion display space.
What I am saying, is that so far as the O/S is concerned, there is no downside to using more RAM, up until the point where it runs out. If the system decides that it can benefit from having 1/2GB of disk cache, and no one else has asked for that RAM, why not let it have the cache? The RAM isn't being used for anything anyway.
My point was rather narrow - RAM that is not in use does nothing for you, and if by using it the system could prevent a page fault later, it may as well use it. Hence my contention that RAM usage in the idle state need not be a good predictor of O/S bloat or how memory constrained the system really is.
Once demand for RAM heats up and the system has to determine how to parcel it out to the running tasks is where things get interesting, and where a well designed memory manager will make a big difference.
It is kind of like screen real estate. You may not need your browser maximized to read
Wanting to have free memory makes little sense. RAM is a resource that the system should use whenever it gets a benefit from doing so. If by using more RAM it can increase working sets, disk caches, display caches, whatever, it should do so.
RAM that is not in use is a resource that you paid for that you are not getting any value from. The only time RAM should be free is if the O/S can't think of anything useful for it to do.
The hard part comes when the O/S has to figure out how to sensibly allocate RAM with the system under load. That is a hard problem, and I make no claims to know what it should do. But looking at RAM usage in the idle state is not a good indication of how the system will perform under load.
I fear that would likely still run in to antitrust issues, because it amounts to dictating the bundling arrangements.
I wonder if they could get away with simply demanding that a vanilla Vista CD be included with the box, so that the customer could do a reinstall if they wanted. It could even be in addition to the OEMs recovery CD, which would put all the craplets back on. But if you had to have the vanilla CD a reformat/reinstall would at least be feasible.
You know, decaf tinfoil works just as well as the real thing.
A backdoor is indeed possible, and given how robust the current encryption algorithms are the NSA would no doubt be happy to have a backdoor.
But don't forget the corollary - many of the systems running Vista are going to be systems that the NSA wants to be secure. There are lots of Windows systems in government and the military.
If the NSA inserts any kind of backdoor they also need to worry about how it could be exploited by a hostile power. Inserting a backdoor, no matter how secret, compromises the security of the system. Would they find it an acceptable tradeoff? I have no idea.
Semi-related anecdote: In WWII, the German Enigma cipher machine used a series of rotors as part of the encode/decode process. The rotors had different possible positions, indicated by letters. So part of the encoding process for a day was the rotor position for each of 3 rotors, such as ABC. A rotor position like AAC effectively turned a 3 rotor machine into a 2 rotor machine. So someone in the German command structure ordered that they not use any daily settings with a repeated position.
The codebreakers at Bletchley Park (this might be early enough that it was still the Polish equivalent of Bletchley) were able to use that simple fact as an attack vector, and it was one of the initial weaknesses they discovered in the Enigma.
Moral of the story: anything you do that reduces the randomness of a crypto system weakens it. A backdoor is a huge decrease in randomness.
Citations please?
I've never seen confirmation that MS apps make any significant use of non-documented OS APIs. The Office group writes much of their own code to be sure, but most of the big players do that.
It is easy enough to use a dependency checker and find all the symbols that a program imports from a DLL. If you cross reference the imports with the documentation in MSDN, it is easy to spot something that is not documented. Given all the axes to grind out there, I can't believe someone hasn't done this already, and I dont recall reading about all the incriminating evidence that was found.
A more plausible claim, and one much harder to prove or disprove, is that the Office team has access to Windows source code, so that rather than creating something from scratch they can just grab a copy of the menuing code and create their own version.