BK isn't actually SCCS based, but it looks like SCCS to external programs which allows make, emacs, patch etc to work seamlessly. It certainly resembles SCCS in the way it manages deltas between versions however.
The Changeset idea is very nice and you can do a lot of cool stuff. eg if someone integrates a bugfix and your manager decides at the last minute that it shouldn't go in, you can generate an anti-changeset which reverses it. Then you can generate an anti-anti-changeset to reintegrate it. I'd like to see someone trying that on Perforce on a change with 100 files inside.
Perforce is firmly in the low-cost bracket when it comes to SCM systems. It's better than VSS or CVS, but it doesn't touch high end systems such as Clearcase (UCM) or Bitkeeper.
Bitkeeper simply beats the pants out of Perforce if you ever have to worry about doing a lot of complex merging or maintaining branches in several sites (in different parts of the world; try having a team in India with an unreliable internet connection) and/or you have a lot of engineers working on a laptop away from the office using wireless or GPRS to dial in. Instead of checking out the raw tree as of a certain point in time, BK is distributed - it clones the entire revision history, and you essentially use it to generate patches which are then resolved and reintegrated when you deliver your work. Any machine, including a laptop away from the office, can become a BK server and allow other people to resync and deliver work to that point. The other nice thing is it's very very stable. I've never had it crash or corrupt the database (and if the database ever did get corrupted it would be straightforward to get a replicated clone from another user and replace it - allowing life to go on).
Perforce has that weird thing with caches and proxies that try to keep the load off the main server. It sounds like there are a lot of points of failure.
It's false to imply that in an open source scenario the situation with Microsoft wouldn't happen. Is there anyone out there in the open source community that would be prepared to support someone working on Linux 2.0 (which is about the same age as VB6 AFAIK), or do you suspect that people are more likely to suggest that the user upgrades to 2.4 or 2.6 ?
The response in the open source world tends to be "your code is out of date, get the latest version which fixes all these bugs" forgetting that many corporate institutions are reluctant to even consider this. That's basically the same as Microsoft's response.
Solaris on SPARC is a rock solid heavy-lifting production environment that doesn't break. End of story. 6 year old Ultra 60s can take the latest Solaris OS and run to provide a perfectly usable environment for coding and development work. Linux's SMP support doesn't even touch the kind of scalability that you'll get on a big Sun box.
Sure, I wouldn't want to run Solaris at home. It's not for doing leisurely stuff or for playing games, it probably won't support your video card, never mind talk to your PDA, and a lot of your USB devices won't work and so on. But that's not what it's for.
The electronic vote count received shortly after the polls close isn't even the final count. I seem to recall with Ohio that the final result went in just a few days before the electoral college vote, which was - correct me if I'm wrong - about a month after the poll took place.
I can only speak for the UK, where the whole business is done and dusted by the morning after the poll.
In a secret ballot, you cannot prove that the computer has not introduced an error. No mechanism of verification exists. If an electronic vote counting machine records the vote cast differently from the vote that was actually enters, how can you tell ? The only way you can do it is record the voter's ID against the vote cast and at a later stage ask the voter to reconfirm their vote in the event of a challenge. Of course if you do that your secret ballot is gone.
I know that computers are less error prone and so do you and most other people here, the trouble is that Joe Public doesn't. Insisting that you are right and he is wrong is Big Brother talking, and suddently your democracy's gone.
The trouble is that with the "tinfoil hat" remark any politician or government could easily dismiss any allegations of misconduct or fiddling in the voting system as being tinfoil hat paranoia and throw them out. And people like you would be around to agree with them. The trouble is that with mechanized vote counting, someone can cry "foul" and you've basically got no way to show the public that the system is incapable of error. With a paper ballot you can recount and recount with the candidates watching, and the scope for an error is removed.
Think about it. Your vote gets sucked into a machine (electronic or mechanical). How can you be confident that negligence or (less likely but possible) deliberate interference or sabotage have not taken place and that the vote you have entered is counted properly ?
Whereas if your candidate and all the other candidates can be there watching the vote get counted, they can observe and act against any negligence or sabotage. You simply cannot do that with any kind of automated vote counting, period.
The public interest is for a hand-counted vote, observed by all the candidates and other independent members of the public, which in other countries is typically completed well within 24 hours of the polls closing.
Any kind of mechanised vote counting whatsoever serves to hide the vote counting process from the electorate. Receipts are a red herring; they are the only way to verify the electronic count and, as a result, render the electronic count completely redundant.
I love the part about "real blood". It would be particularly interesting if a piece of the bloodstain could be extracted, as then we'd know the makeup of Christ's DNA. Future technology could permit us to clone as many Christs as we liked, who would in turn go around the place healing people, thus dramatically revolutionizing contemporary medicine.
Of course, the DNA would likely reveal that the person buried in the shroud had two parents of Middle Eastern origin. Then the shroud's fans would have to either deny the shroud's authenticity, or deny that Christ was the son of God. There's a thought - what would God's DNA look like and how much would people pay if it could be cloned ? Everyone would want their child genetically engineered with God genes.
In the US we take these as a given. I've encountered people where I work that are from other countries that truly do not comprehend the value of freedom and democracy
Actually in the US you don't take these as given, and I've encountered plenty of Americans who do not comprehend the value of freedom and democracy. These are the Americans who voted for Bush, who can't see the problems with the Patriot Act or the war on Iraq, and who really believe that the Department of Homeland Security and it's powers are there to make them safer by taking away their freedoms and regulating their liberty. These are people who ignored the founding fathers who made their famous comments about the perils of trading freedom for security. Don't get me wrong, I think the USA is a great country with many great people, But I'm afraid that a majority of those who voted there are under the misapprehension that what their government is doing and the way their country is run constitutes a free society.
I have nothing against stickers on textbooks saying "evolution is theory not a fact", provided we can also have a sticker on the front of each bible saying "God is a belief, not a fact". Do you think the creationists would go for it ?
He has achieved some incredible things, but I wonder how consistent his ideology is. He regards using non-free software as unethical, but before Linux the GNU operating system was used with various non-free kernels.
How do you think the first versions of GCC got compiled ? Chances are with a non-free C compiler.
I'm an atheist so I don't believe in your "saviour" mumbo jumbo or made-up deities. However, I can't see how Christ would have approved of Christmas. It's all about gluttony and apart from anything else it brings out the worst taste in everyone.
It's not a question though, of "once a free kernel is available". It's a question of "once a free kernel is available and good enough. In terms of GNU history, a little time had to pass before this was true. Meantime, SunOS/Solaris and other alternatives had to fill that gap.
There is nothing surprising in reading that RMS supports the use of GNU tools on non-free kernels. Outside of the kernel, much of the original tools were developed on platforms such as Solaris. This was pretty necessary at the time because Linux wasn't yet mature and the Hurd was, well, pretty much where the Hurd is now.
Enterprises do not want to run around trying to find stable kernels and stable machines to run them on. Solaris on SPARC provides them with an OS which is specifically tuned to that hardware. That is an option you simply do not have with Linux, which is not tuned to any hardware specifically. It becomes more of a problem when you get into fairly big multiway SMP boxes with more than four processors. Such machines are not readily available under x86, and if they were it's hard to say exactly how well Linux would run on them.
So it's not really a valid response to say "the comparison isn't fair when you use Sun's hardware". Larger businesses just want the most stable platform they can find with the appropriate backup during the rare occasions when it fails. This is the niche that Linux - excellent as it is as a general-purpose server OS - is unlikely to fill.
To your other point, open source and full documentation is a great idea, but someone still has to compile the code and load it on to the machine. At the end of the day, it still comes down to a small group of people, in whom you have to place your trust to ensure the election is conducted fairly. To turn it around a little, if it transpired that one of those people was corrupt, how would you know ?
This reminds me of Ken Thompson's argument about the Trojan Horse. Sure you've got the source right there, but how can you say for sure that this is the actual code the machine will be executing ? Thompson points out that you can't trust code which you did not entirely create yourself.
This is why I think open source in voting machines is still a red herring. You could say that I'm being completely paranoid - sure, and in 99% of other cases involving computers you'd be right. But the trouble is that in a democratic system, people in opposition will (should!) sling all the mud they can to try to undermine the election of their opponent. Electronic voting systems can't be defended from this type of attack, and as a result people start to question it, thereby losing confidence in the democratic system. A hand count where the opposing candidate was free to attend and "put up or shut up" if you like about the conduct of the count, is disarmed from undermining the vote in this way.
In theory voting machines are approximately as trustworthy as a hand count. In practice, voting machine software has bugs, can be interfered with by partisan individuals, etc.
The advantage of a manual count is that it can be open to the public, so the candidates, lawyers and anyone else can watch each vote being counted and monitor it so that they can satisfy themselves that the wheels of democracy are turning as they should be. You don't have the election officials and the makers of the software or the machines going into a corner to figure out whether the count is wrong or not. This level of verification isn't possible with any kind of automated count. At a manual count, you can also watch the election officials as they filter the ballots for ballot papers which haven't been marked properly (hanging chads anyone?). You can't do that with an automated count.
A spoiled ballot in the UK is one which has no intelligible mark anywhere for a candidate. Some argue that the ability to spoil a ballot (ie write "all the candidates are wankers" on the ballot paper) is a democratic right. I thought it meant the same thing in the US, but I think your point is that the machine is able to prevent people from accidentally making the wrong mark on the ballot. With manual counts in the UK this problem doesn't exist, as if you make any kind of mark on the ballot paper which can reasonably be associated with one of the candidates, it will be counted as a vote (and the election administrator will seek the assent of the candidates present at the count when making this decision).
This is perhaps the better of the systems involved, but you still have to trust the software that's doing the optical scanning. You also have to trust the guys who have calibrated the scanning machine.
Easier not to bother, and just count 'em by hand. It takes a few more hours, but it's cheaper and better.
But that's silly (as is the article in general). You can't make extrapolations about what you think the vote might be against what it actually was. Exit polls, telephone polls and other kinds of poll are sometimes right but frequently wrong. Furthermore, the argument goes completely out of the window in a close vote (like the recent one) where the polls simply can't tell you accurately about the 100 votes that might have swung the election.
How would it work ? You could define a rule "if the real vote is X% away from the exit polls, we do a recount". Then of course, Mr. Corrupt Election Interferer (whoever it is or could be) knows that when he fiddles the vote, all he needs to do is insure his fiddling slips the result in under the radar, which he can easily do if he has access to the running totals on the exit polls. That's assuming, of course, that the exit polls themselves haven't been fiddled (if you give them an official role in a count, then why not?).
The point is that at the end of the day the computer spits out a number. How can you look at the number and say "I have confidence that the computer is right" ? What I'm saying is that logically, because the result cannot immediately be verified against a trustable source, you have to assume that the computer's result is wrong and will need recounting against paper receipts in all cases. Therefore, there is no point in having the electronic count in the first place.
To build an electronic voting system that can come close to being considered reliable, you have to wrap it in all kinds of checks and balances, set all kinds of standards and make sure they're enforced, do all sorts of monitoring of the vote and checking of the result. And even then you've still got a system which isn't transparent to Joe Public who can barely program a video recorder.
Is it really the best way to spend taxpayer dollars ?
"faster and more accurate". Define "more accurate". How do you measure the accuracy of an electronic vote ?
The "it takes forever" point is specious in the extreme, mainly because of the underlying implication that it is more important to get a quick result than it is to get a correct one. Either way, in the UK the count is complete within 12 hours of the polls closing, so we usually know who's going to be Prime Minister the next morning. We've got a quarter of the number of votes you guys have. Meanwhile, in Ohio they're still counting votes - two weeks after the poll. What's that all about ?
And you're still ignoring the point. On what criteria can the losing candidate question the machines ? There are no metrics available which can determine whether the vote is valid or not. It's not like the hanging chads situation where people identified a problem that would cause the votes to be counted wrongly - the e-voting machine just spits out a number, how can you tell if it is questionable or not ?
Your point "if there's any question about the electronic tally" implies that there are some way to say whether the electronic vote is sound or unsound.
That is of course wrong; there is no way to say whether the electronic vote is questionable or not. What you then end up with is a situation where the losing candidate can simply question the vote, and there's no way to defend that.
This means that there is no point in having an electronic vote to start with. Trust me guys - pen and paper is the way to do this. It's always verifiable, the count can be observed by the candidates and by independent observers. There's no basis for bugs, security breaches or hacking, or any of that other nonsense.
Slashdot Mirror
BK isn't actually SCCS based, but it looks like SCCS to external programs which allows make, emacs, patch etc to work seamlessly. It certainly resembles SCCS in the way it manages deltas between versions however.
The Changeset idea is very nice and you can do a lot of cool stuff. eg if someone integrates a bugfix and your manager decides at the last minute that it shouldn't go in, you can generate an anti-changeset which reverses it. Then you can generate an anti-anti-changeset to reintegrate it. I'd like to see someone trying that on Perforce on a change with 100 files inside.
Perforce is firmly in the low-cost bracket when it comes to SCM systems. It's better than VSS or CVS, but it doesn't touch high end systems such as Clearcase (UCM) or Bitkeeper.
Bitkeeper simply beats the pants out of Perforce if you ever have to worry about doing a lot of complex merging or maintaining branches in several sites (in different parts of the world; try having a team in India with an unreliable internet connection) and/or you have a lot of engineers working on a laptop away from the office using wireless or GPRS to dial in. Instead of checking out the raw tree as of a certain point in time, BK is distributed - it clones the entire revision history, and you essentially use it to generate patches which are then resolved and reintegrated when you deliver your work. Any machine, including a laptop away from the office, can become a BK server and allow other people to resync and deliver work to that point. The other nice thing is it's very very stable. I've never had it crash or corrupt the database (and if the database ever did get corrupted it would be straightforward to get a replicated clone from another user and replace it - allowing life to go on).
Perforce has that weird thing with caches and proxies that try to keep the load off the main server. It sounds like there are a lot of points of failure.
It's false to imply that in an open source scenario the situation with Microsoft wouldn't happen. Is there anyone out there in the open source community that would be prepared to support someone working on Linux 2.0 (which is about the same age as VB6 AFAIK), or do you suspect that people are more likely to suggest that the user upgrades to 2.4 or 2.6 ?
The response in the open source world tends to be "your code is out of date, get the latest version which fixes all these bugs" forgetting that many corporate institutions are reluctant to even consider this. That's basically the same as Microsoft's response.
Solaris on SPARC is a rock solid heavy-lifting production environment that doesn't break. End of story. 6 year old Ultra 60s can take the latest Solaris OS and run to provide a perfectly usable environment for coding and development work. Linux's SMP support doesn't even touch the kind of scalability that you'll get on a big Sun box.
Sure, I wouldn't want to run Solaris at home. It's not for doing leisurely stuff or for playing games, it probably won't support your video card, never mind talk to your PDA, and a lot of your USB devices won't work and so on. But that's not what it's for.
The electronic vote count received shortly after the polls close isn't even the final count. I seem to recall with Ohio that the final result went in just a few days before the electoral college vote, which was - correct me if I'm wrong - about a month after the poll took place.
I can only speak for the UK, where the whole business is done and dusted by the morning after the poll.
In a secret ballot, you cannot prove that the computer has not introduced an error. No mechanism of verification exists. If an electronic vote counting machine records the vote cast differently from the vote that was actually enters, how can you tell ? The only way you can do it is record the voter's ID against the vote cast and at a later stage ask the voter to reconfirm their vote in the event of a challenge. Of course if you do that your secret ballot is gone.
I know that computers are less error prone and so do you and most other people here, the trouble is that Joe Public doesn't. Insisting that you are right and he is wrong is Big Brother talking, and suddently your democracy's gone.
The trouble is that with the "tinfoil hat" remark any politician or government could easily dismiss any allegations of misconduct or fiddling in the voting system as being tinfoil hat paranoia and throw them out. And people like you would be around to agree with them. The trouble is that with mechanized vote counting, someone can cry "foul" and you've basically got no way to show the public that the system is incapable of error. With a paper ballot you can recount and recount with the candidates watching, and the scope for an error is removed.
Think about it. Your vote gets sucked into a machine (electronic or mechanical). How can you be confident that negligence or (less likely but possible) deliberate interference or sabotage have not taken place and that the vote you have entered is counted properly ?
Whereas if your candidate and all the other candidates can be there watching the vote get counted, they can observe and act against any negligence or sabotage. You simply cannot do that with any kind of automated vote counting, period.
The public interest is for a hand-counted vote, observed by all the candidates and other independent members of the public, which in other countries is typically completed well within 24 hours of the polls closing.
Any kind of mechanised vote counting whatsoever serves to hide the vote counting process from the electorate. Receipts are a red herring; they are the only way to verify the electronic count and, as a result, render the electronic count completely redundant.
I love the part about "real blood". It would be particularly interesting if a piece of the bloodstain could be extracted, as then we'd know the makeup of Christ's DNA. Future technology could permit us to clone as many Christs as we liked, who would in turn go around the place healing people, thus dramatically revolutionizing contemporary medicine.
Of course, the DNA would likely reveal that the person buried in the shroud had two parents of Middle Eastern origin. Then the shroud's fans would have to either deny the shroud's authenticity, or deny that Christ was the son of God. There's a thought - what would God's DNA look like and how much would people pay if it could be cloned ? Everyone would want their child genetically engineered with God genes.
In the US we take these as a given. I've encountered people where I work that are from other countries that truly do not comprehend the value of freedom and democracy
Actually in the US you don't take these as given, and I've encountered plenty of Americans who do not comprehend the value of freedom and democracy. These are the Americans who voted for Bush, who can't see the problems with the Patriot Act or the war on Iraq, and who really believe that the Department of Homeland Security and it's powers are there to make them safer by taking away their freedoms and regulating their liberty. These are people who ignored the founding fathers who made their famous comments about the perils of trading freedom for security. Don't get me wrong, I think the USA is a great country with many great people, But I'm afraid that a majority of those who voted there are under the misapprehension that what their government is doing and the way their country is run constitutes a free society.
I have nothing against stickers on textbooks saying "evolution is theory not a fact", provided we can also have a sticker on the front of each bible saying "God is a belief, not a fact". Do you think the creationists would go for it ?
He has achieved some incredible things, but I wonder how consistent his ideology is. He regards using non-free software as unethical, but before Linux the GNU operating system was used with various non-free kernels.
How do you think the first versions of GCC got compiled ? Chances are with a non-free C compiler.
"Probability of 2.7%"
...
There you go, I always thought probability was between zero and one (inform yourself)
Linux is not tuned to an entire machine or hardware configuration. Sure it's tuned to an architecture, I didn't deny that.
The 512 way machine you're talking about is not an SMP box is it ? Where can I buy one ?
I'm an atheist so I don't believe in your "saviour" mumbo jumbo or made-up deities. However, I can't see how Christ would have approved of Christmas. It's all about gluttony and apart from anything else it brings out the worst taste in everyone.
It's not a question though, of "once a free kernel is available". It's a question of "once a free kernel is available and good enough. In terms of GNU history, a little time had to pass before this was true. Meantime, SunOS/Solaris and other alternatives had to fill that gap.
There is nothing surprising in reading that RMS supports the use of GNU tools on non-free kernels. Outside of the kernel, much of the original tools were developed on platforms such as Solaris. This was pretty necessary at the time because Linux wasn't yet mature and the Hurd was, well, pretty much where the Hurd is now.
If RMS criticized this idea he'd be a hypocrite.
You're missing the point.
Enterprises do not want to run around trying to find stable kernels and stable machines to run them on. Solaris on SPARC provides them with an OS which is specifically tuned to that hardware. That is an option you simply do not have with Linux, which is not tuned to any hardware specifically. It becomes more of a problem when you get into fairly big multiway SMP boxes with more than four processors. Such machines are not readily available under x86, and if they were it's hard to say exactly how well Linux would run on them.
So it's not really a valid response to say "the comparison isn't fair when you use Sun's hardware". Larger businesses just want the most stable platform they can find with the appropriate backup during the rare occasions when it fails. This is the niche that Linux - excellent as it is as a general-purpose server OS - is unlikely to fill.
You have your own indestructible polyalloy cyborg from the future ? I like it.
To your other point, open source and full documentation is a great idea, but someone still has to compile the code and load it on to the machine. At the end of the day, it still comes down to a small group of people, in whom you have to place your trust to ensure the election is conducted fairly. To turn it around a little, if it transpired that one of those people was corrupt, how would you know ?
This reminds me of Ken Thompson's argument about the Trojan Horse. Sure you've got the source right there, but how can you say for sure that this is the actual code the machine will be executing ? Thompson points out that you can't trust code which you did not entirely create yourself.
This is why I think open source in voting machines is still a red herring. You could say that I'm being completely paranoid - sure, and in 99% of other cases involving computers you'd be right. But the trouble is that in a democratic system, people in opposition will (should!) sling all the mud they can to try to undermine the election of their opponent. Electronic voting systems can't be defended from this type of attack, and as a result people start to question it, thereby losing confidence in the democratic system. A hand count where the opposing candidate was free to attend and "put up or shut up" if you like about the conduct of the count, is disarmed from undermining the vote in this way.
In theory voting machines are approximately as trustworthy as a hand count. In practice, voting machine software has bugs, can be interfered with by partisan individuals, etc.
The advantage of a manual count is that it can be open to the public, so the candidates, lawyers and anyone else can watch each vote being counted and monitor it so that they can satisfy themselves that the wheels of democracy are turning as they should be. You don't have the election officials and the makers of the software or the machines going into a corner to figure out whether the count is wrong or not. This level of verification isn't possible with any kind of automated count. At a manual count, you can also watch the election officials as they filter the ballots for ballot papers which haven't been marked properly (hanging chads anyone?). You can't do that with an automated count.
A spoiled ballot in the UK is one which has no intelligible mark anywhere for a candidate. Some argue that the ability to spoil a ballot (ie write "all the candidates are wankers" on the ballot paper) is a democratic right. I thought it meant the same thing in the US, but I think your point is that the machine is able to prevent people from accidentally making the wrong mark on the ballot. With manual counts in the UK this problem doesn't exist, as if you make any kind of mark on the ballot paper which can reasonably be associated with one of the candidates, it will be counted as a vote (and the election administrator will seek the assent of the candidates present at the count when making this decision).
This is perhaps the better of the systems involved, but you still have to trust the software that's doing the optical scanning. You also have to trust the guys who have calibrated the scanning machine.
Easier not to bother, and just count 'em by hand. It takes a few more hours, but it's cheaper and better.
But that's silly (as is the article in general). You can't make extrapolations about what you think the vote might be against what it actually was. Exit polls, telephone polls and other kinds of poll are sometimes right but frequently wrong. Furthermore, the argument goes completely out of the window in a close vote (like the recent one) where the polls simply can't tell you accurately about the 100 votes that might have swung the election.
How would it work ? You could define a rule "if the real vote is X% away from the exit polls, we do a recount". Then of course, Mr. Corrupt Election Interferer (whoever it is or could be) knows that when he fiddles the vote, all he needs to do is insure his fiddling slips the result in under the radar, which he can easily do if he has access to the running totals on the exit polls. That's assuming, of course, that the exit polls themselves haven't been fiddled (if you give them an official role in a count, then why not?).
The point is that at the end of the day the computer spits out a number. How can you look at the number and say "I have confidence that the computer is right" ? What I'm saying is that logically, because the result cannot immediately be verified against a trustable source, you have to assume that the computer's result is wrong and will need recounting against paper receipts in all cases. Therefore, there is no point in having the electronic count in the first place.
To build an electronic voting system that can come close to being considered reliable, you have to wrap it in all kinds of checks and balances, set all kinds of standards and make sure they're enforced, do all sorts of monitoring of the vote and checking of the result. And even then you've still got a system which isn't transparent to Joe Public who can barely program a video recorder.
Is it really the best way to spend taxpayer dollars ?
"faster and more accurate". Define "more accurate". How do you measure the accuracy of an electronic vote ?
The "it takes forever" point is specious in the extreme, mainly because of the underlying implication that it is more important to get a quick result than it is to get a correct one. Either way, in the UK the count is complete within 12 hours of the polls closing, so we usually know who's going to be Prime Minister the next morning. We've got a quarter of the number of votes you guys have. Meanwhile, in Ohio they're still counting votes - two weeks after the poll. What's that all about ?
And you're still ignoring the point. On what criteria can the losing candidate question the machines ? There are no metrics available which can determine whether the vote is valid or not. It's not like the hanging chads situation where people identified a problem that would cause the votes to be counted wrongly - the e-voting machine just spits out a number, how can you tell if it is questionable or not ?
Your point "if there's any question about the electronic tally" implies that there are some way to say whether the electronic vote is sound or unsound.
That is of course wrong; there is no way to say whether the electronic vote is questionable or not. What you then end up with is a situation where the losing candidate can simply question the vote, and there's no way to defend that.
This means that there is no point in having an electronic vote to start with. Trust me guys - pen and paper is the way to do this. It's always verifiable, the count can be observed by the candidates and by independent observers. There's no basis for bugs, security breaches or hacking, or any of that other nonsense.