Here in the UK the banks actually did something useful for their customers some years ago - they set up one big network just for cash machines, called it "Link" and allow anyone to use a bank-owned Link cash machine free of charge.
Pretty much every cash machine has a "link" symbol on it.
The big issue at the moment is small cash machines appearing in pubs, clubs and shops which charge (generally about £1.50 per transaction). I don't have a problem with it, as I live in a city so I'm never that far away from a cash machine which doesn't charge. But there are areas (and they're apparently becoming larger) where there are no bank-owned cash machines for miles around, and the formerly bank-owned machines (often at petrol stations or leisure parks) are being sold to the companies that charge a fee.
And yes, I'm oversimplifing the issue, but it's increasingly clear that the current codebase is unsupportable. It's collapsing under its own weight. Can you imagine if glibc on Linux had to support every function call from 1990 forward? That's basically what MS is trying to do. It's a nightmare.
Glibc DOES support every function call from 1990 onward. The only difference (and it's a huge one) is that the number of function calls required for a standard C library in Unix is relatively small.
Microsoft is a victim of their own desire to do everything. Rather than let someone or something else provide APIs, they do so themselves. On the one hand, this guarantees that any given version of Windows can be more-or-less guaranteed to have the necessary libraries, thus ensuring application compatability and consistency.
Now this all works very nicely, until it becomes necessary to consider the issue of backward compatability. If you're not too bothered by people evaluating alternatives, that's fine. But if you are, then as soon as you break backward compatability you wind up forcing all of the people affected to sit down and think "Right. New version of Windows either means new business software and/or we stick with what we've got for as long as possible. That being the case, who out there can supply us with appropriate software?". The application drives the operating system, and if the applications either aren't OS specific or aren't available for your OS, the need for the customer to buy your OS above anyone elses evaporates.
IOW, Microsoft could do this. But they won't because it will make evaluating alternatives all the more interesting.
They did that years ago. In NT and all its children, DOS is a separate application provided as a compatability layer running on a kernel which has nothing to do with DOS.
At a basic level, it has the exact same capabilities and characteristics as any other application - ie. it runs in a protected memory space and applications may not access hardware directly.
As regards the WMF issue - the problem with that is intrinsic to a Von Neumann machine, insofar as data and code share the same memory address space. You can't tell whether you're supposed to execute the data stored in a pointer just by looking at it - you have to take it on trust that the pointer will always point at data or code as appropriate. In theory, any carefully crafted file (doesn't even need to be an image), thrown at the right bit of handling code, could cause havoc. Removing support may reduce the amount of code that needs auditing, but in doing so functionality is reduced - not the kind of thing that Microsoft tends to go for. And it still doesn't solve the fundamental issue - that code and data are intertwined in a typical memory map.
There is a way - not to make use of "Office versions that run on pre-Vista OS" illegal, but to make them impractical.
Change the file format again.
The only problem is that while that trick worked in the Windows '9x days, it's doubtful it will again. The install base not using Vista is so large that this is simply not going to achieve anything.
Linux isn't based on Unix. It borrows ideas heavily from it, but technically it's a different operating system because the kernel and userland tools have independent origins rather than being forked off either BSD or SysV.
This is a bit like saying that the bird which just waddled past, with duck-like feathers and a yellow beak, making a "quack quack" noise wasn't a duck.
Eventually I would like to see VMs in production too. Oh, and we use the enterprise version. so there
As would I. I reckon I could run my entire infrastructure (currently about a dozen boxes) on two reasonably-basic servers running VMWare Enterprise and clustered for redundancy.
Have you had a chance to use VMware Enterprise in that kind of configuration? Is it any good?
Seriously though, you know its true that the bulk of business use it typing letters, contracts, whatever; a little email...
And then you keep on getting sent attachments you can't open.
It sounds silly, that the only reason for upgrading is "So-and-so keeps sending me attachments I can't open". Why can't so-and-so use the "Export" function?
But sooner or later, "So-and-so" becomes "lots of people", and the amount of hassle involved in getting everyone to use a simple enough format outweighs the cost of just buying a faster computer.
They can, provided they can get the House of Commons to agree to it in the first place. If it comes down to it, they could use the Parliament Act to get past the Lords.
Ordinarily, this would sound completely preposterous - why would the MPs in the House of Commons agree to such legislation? - but you've got to bear in mind that the party with the most MPs gets to choose Prime Minister, who then appoints his Cabinet. Therefore, the Labour-controlled Cabinet are asking a group of MPs who are >50% members of the Labour party to vote on it.
Historically, most Labour MPs have been good at towing the party line. It's therefore well within the bounds of probability that it could pass.
Not that I expect it to achieve much, as he's Labour.
The law itself is actually quite short and, once you get your head around the legalese, just about readable.
It's here if you want to have a read. Do so, there is nothing more annoying than debating legislation with people who claim to know it but haven't actually read it.
or introduce prison sentences longer than 2 years.
Oh, so they can only send me to prison for 2 years? Well, that's good to know. I'm sure the fact that it's two years (rather than, say, five) will make a lot of difference to my employability when I come out.
The fact that John repealled Magna Carta a few weeks later gets glossed over by most commentaters.
I wish I had known that when I was in school and the history teacher took an assembly telling us all about how wonderful Magna Carta was and how it was the foundation of modern democracy. "Excuse me, Sir, but what about when King John repealed it a few weeks later?"
Of course, but if they want to watch the videos in HD format, they will have to buy a separate player or another computer with Windows Vista.
If they're in the movie industry, creating these movies, I sincerely doubt they will find DRM to be a problem. The DRM will almost certainly be added later, at the mastering stage.
the patent officials are probably not knowledgable enough in the field of the patent to make a judgement call as to whether or not the material really represents patentable, original ideas.
Dammit, that's their job! If they're not knowledgeable enough thenwhy are they doing it?
Besides which, I keep on seeing people pulling prior art out in the form of existing patents. Do they not use their own database?
The court assume the patent office has done its job.
This is something I keep hearing, yet have trouble believing.
The US patent office has clearly given up on examining patents (except in possibly the most glaringly obvious of caases), figuring that the lawyers can battle it out in court.
The courts have decided "well, the patent was awarded so there must be some element of patentability to it".
Is it really the case that these two departments, with the hundreds, if not thousands of people they keep in employment, has yet to notice this? That nobody in the courts has actually telephoned the patent office and said "Look, I know this might be a stupid question, but are you guys actually bothering to read patent applications before you grant them?".
Neither one of the Fujitsu products, RAFT and GlobalStore, is among the products approved by the major credit card companies.....
Seems like something went wrong
I'll tell you what went wrong - the credit card companies were accepting transactions put through by non-approved software. I'll get modded to hell and back for this, but this strikes me as exactly the kind of place where Trusted Computing could be useful - as a means to guarantee that the software being used to conduct financial transactions is approved, and hasn't been tampered with.
The credit card would be a little usb stick that stored a processor and a key. When you bought something, you'd stick the key into a usb port, the stick would show you the transaction amount, you'd push a button allowing the processor to compute a hash value which would be trasmitted to the bank for verification.
I'd disagree. What has happened is that with every version of Windows, the common attack vector was more secure. Result: The typical attack vector changed.
WinNT : Attack vector was generally through IIS. At the time, there weren't vast swathes of botnets and NT4 was never very popular outside of businesses.
Win2K : Attack vector through any number of services, email being one of the most common (Taking advantage of lousy security in commonly-used software on top of the OS, such as Outlook/Outlook express). Attacks generally worm-like in nature and sought to do little more than replicate.
WinXP pre-SP2 and pre-Firewalls becoming ubiquitous : Attack through any number of services, with nothing particularly standing out as being common. Attacks either automated or kiddie-driven; with a definite aim - frequently to build a botnet.
WinXP post-SP2 and post-Firewalls in general : Attack through the user and client software - either attracting them with eye-candy or some dubious website scripting.
Beta went digital years ago. Same form factor for the tapes, but the data they've held has been digitally captured, stored and processed for years. Punch "digibeta" into Google and see what comes back.
It's not just the US here, though. History has taught us that any leader can lie through their teeth to their own ends - I'm perhaps skirting a bit close to Godwin's Law here, but Neville Chamberlain famously came back from Germany with a promise signed by a certain chap with a distinctive moustache not to invade Poland.
These planes cost several million dollars each - it's not like a PC where you expect to throw it out after 3-5 years. And it's all very well saying "They probably can't shut them down remotely" - I'm not an aviation expert so I'm not really qualified to comment. Even so, I am at a loss to understand why we should trust the lives of our air force and their ability to protect the country in what is essentially a black box we don't know the workings of. Providing a copy of the source code and a development kit (if such a thing even exists for an airplane - I suspect if it does it won't bear much resemblance to, say, Linux) so we can audit it and continue development ourselves if necessary isn't an issue of "do we trust each other" - it strikes me as perfectly reasonable.
Slashdot Mirror
Here in the UK the banks actually did something useful for their customers some years ago - they set up one big network just for cash machines, called it "Link" and allow anyone to use a bank-owned Link cash machine free of charge.
Pretty much every cash machine has a "link" symbol on it.
The big issue at the moment is small cash machines appearing in pubs, clubs and shops which charge (generally about £1.50 per transaction). I don't have a problem with it, as I live in a city so I'm never that far away from a cash machine which doesn't charge. But there are areas (and they're apparently becoming larger) where there are no bank-owned cash machines for miles around, and the formerly bank-owned machines (often at petrol stations or leisure parks) are being sold to the companies that charge a fee.
And yes, I'm oversimplifing the issue, but it's increasingly clear that the current codebase is unsupportable. It's collapsing under its own weight. Can you imagine if glibc on Linux had to support every function call from 1990 forward? That's basically what MS is trying to do. It's a nightmare.
Glibc DOES support every function call from 1990 onward. The only difference (and it's a huge one) is that the number of function calls required for a standard C library in Unix is relatively small.
Microsoft is a victim of their own desire to do everything. Rather than let someone or something else provide APIs, they do so themselves. On the one hand, this guarantees that any given version of Windows can be more-or-less guaranteed to have the necessary libraries, thus ensuring application compatability and consistency.
Now this all works very nicely, until it becomes necessary to consider the issue of backward compatability. If you're not too bothered by people evaluating alternatives, that's fine. But if you are, then as soon as you break backward compatability you wind up forcing all of the people affected to sit down and think "Right. New version of Windows either means new business software and/or we stick with what we've got for as long as possible. That being the case, who out there can supply us with appropriate software?". The application drives the operating system, and if the applications either aren't OS specific or aren't available for your OS, the need for the customer to buy your OS above anyone elses evaporates.
IOW, Microsoft could do this. But they won't because it will make evaluating alternatives all the more interesting.
They did that years ago. In NT and all its children, DOS is a separate application provided as a compatability layer running on a kernel which has nothing to do with DOS.
At a basic level, it has the exact same capabilities and characteristics as any other application - ie. it runs in a protected memory space and applications may not access hardware directly.
As regards the WMF issue - the problem with that is intrinsic to a Von Neumann machine, insofar as data and code share the same memory address space. You can't tell whether you're supposed to execute the data stored in a pointer just by looking at it - you have to take it on trust that the pointer will always point at data or code as appropriate. In theory, any carefully crafted file (doesn't even need to be an image), thrown at the right bit of handling code, could cause havoc. Removing support may reduce the amount of code that needs auditing, but in doing so functionality is reduced - not the kind of thing that Microsoft tends to go for. And it still doesn't solve the fundamental issue - that code and data are intertwined in a typical memory map.
There is a way - not to make use of "Office versions that run on pre-Vista OS" illegal, but to make them impractical.
Change the file format again.
The only problem is that while that trick worked in the Windows '9x days, it's doubtful it will again. The install base not using Vista is so large that this is simply not going to achieve anything.
Linux isn't based on Unix. It borrows ideas heavily from it, but technically it's a different operating system because the kernel and userland tools have independent origins rather than being forked off either BSD or SysV.
This is a bit like saying that the bird which just waddled past, with duck-like feathers and a yellow beak, making a "quack quack" noise wasn't a duck.
Eventually I would like to see VMs in production too. Oh, and we use the enterprise version. so there
As would I. I reckon I could run my entire infrastructure (currently about a dozen boxes) on two reasonably-basic servers running VMWare Enterprise and clustered for redundancy.
Have you had a chance to use VMware Enterprise in that kind of configuration? Is it any good?
Seriously though, you know its true that the bulk of business use it typing letters, contracts, whatever; a little email...
And then you keep on getting sent attachments you can't open.
It sounds silly, that the only reason for upgrading is "So-and-so keeps sending me attachments I can't open". Why can't so-and-so use the "Export" function?
But sooner or later, "So-and-so" becomes "lots of people", and the amount of hassle involved in getting everyone to use a simple enough format outweighs the cost of just buying a faster computer.
Then you can have a database server, and a whole bunch of other virtual machines. All logically seprate... but on one peice of hardware.
That way, when that one piece of hardware fails, you lose the entire damn business rather than just one or two services. Wonderful, isn't it?
(Yes, I'm aware of the enterprise versions of VMware which support clustering)
Until a timer expires, a network packet arrives, a key is pressed and Windows sends a message more-or-less simultaneously.
Changing one line: $5
Knowing which line to change: $1000
They can, provided they can get the House of Commons to agree to it in the first place. If it comes down to it, they could use the Parliament Act to get past the Lords.
Ordinarily, this would sound completely preposterous - why would the MPs in the House of Commons agree to such legislation? - but you've got to bear in mind that the party with the most MPs gets to choose Prime Minister, who then appoints his Cabinet. Therefore, the Labour-controlled Cabinet are asking a group of MPs who are >50% members of the Labour party to vote on it.
Historically, most Labour MPs have been good at towing the party line. It's therefore well within the bounds of probability that it could pass.
I've just done so.
Not that I expect it to achieve much, as he's Labour.
The law itself is actually quite short and, once you get your head around the legalese, just about readable.
It's here if you want to have a read. Do so, there is nothing more annoying than debating legislation with people who claim to know it but haven't actually read it.
or introduce prison sentences longer than 2 years.
Oh, so they can only send me to prison for 2 years? Well, that's good to know. I'm sure the fact that it's two years (rather than, say, five) will make a lot of difference to my employability when I come out.
</sarcasm>
The fact that John repealled Magna Carta a few weeks later gets glossed over by most commentaters.
I wish I had known that when I was in school and the history teacher took an assembly telling us all about how wonderful Magna Carta was and how it was the foundation of modern democracy. "Excuse me, Sir, but what about when King John repealed it a few weeks later?"
However, they won't pass this information up the chain to the executives, who will instead blame poor viewing figures on piracy.
Of course, but if they want to watch the videos in HD format, they will have to buy a separate player or another computer with Windows Vista.
If they're in the movie industry, creating these movies, I sincerely doubt they will find DRM to be a problem. The DRM will almost certainly be added later, at the mastering stage.
Someone who still posts embarrassing pictures after you told them not to really isn't a very good "friend".
True, but this doesn't stop it from being a potential problem if an employer sees the picture.
Now is there any chance you could actually post a link to the article?
the patent officials are probably not knowledgable enough in the field of the patent to make a judgement call as to whether or not the material really represents patentable, original ideas.
Dammit, that's their job! If they're not knowledgeable enough thenwhy are they doing it?
Besides which, I keep on seeing people pulling prior art out in the form of existing patents. Do they not use their own database?
The court assume the patent office has done its job.
This is something I keep hearing, yet have trouble believing.
The US patent office has clearly given up on examining patents (except in possibly the most glaringly obvious of caases), figuring that the lawyers can battle it out in court.
The courts have decided "well, the patent was awarded so there must be some element of patentability to it".
Is it really the case that these two departments, with the hundreds, if not thousands of people they keep in employment, has yet to notice this? That nobody in the courts has actually telephoned the patent office and said "Look, I know this might be a stupid question, but are you guys actually bothering to read patent applications before you grant them?".
Neither one of the Fujitsu products, RAFT and GlobalStore, is among the products approved by the major credit card companies.....
Seems like something went wrong
I'll tell you what went wrong - the credit card companies were accepting transactions put through by non-approved software. I'll get modded to hell and back for this, but this strikes me as exactly the kind of place where Trusted Computing could be useful - as a means to guarantee that the software being used to conduct financial transactions is approved, and hasn't been tampered with.
The credit card would be a little usb stick that stored a processor and a key. When you bought something, you'd stick the key into a usb port, the stick would show you the transaction amount, you'd push a button allowing the processor to compute a hash value which would be trasmitted to the bank for verification.
Do you mean a bit like Chip & PIN?
I'd disagree. What has happened is that with every version of Windows, the common attack vector was more secure. Result: The typical attack vector changed.
WinNT : Attack vector was generally through IIS. At the time, there weren't vast swathes of botnets and NT4 was never very popular outside of businesses.
Win2K : Attack vector through any number of services, email being one of the most common (Taking advantage of lousy security in commonly-used software on top of the OS, such as Outlook/Outlook express). Attacks generally worm-like in nature and sought to do little more than replicate.
WinXP pre-SP2 and pre-Firewalls becoming ubiquitous : Attack through any number of services, with nothing particularly standing out as being common. Attacks either automated or kiddie-driven; with a definite aim - frequently to build a botnet.
WinXP post-SP2 and post-Firewalls in general : Attack through the user and client software - either attracting them with eye-candy or some dubious website scripting.
All that remains is "what comes next".
Beta went digital years ago. Same form factor for the tapes, but the data they've held has been digitally captured, stored and processed for years. Punch "digibeta" into Google and see what comes back.
It's not just the US here, though. History has taught us that any leader can lie through their teeth to their own ends - I'm perhaps skirting a bit close to Godwin's Law here, but Neville Chamberlain famously came back from Germany with a promise signed by a certain chap with a distinctive moustache not to invade Poland.
These planes cost several million dollars each - it's not like a PC where you expect to throw it out after 3-5 years. And it's all very well saying "They probably can't shut them down remotely" - I'm not an aviation expert so I'm not really qualified to comment. Even so, I am at a loss to understand why we should trust the lives of our air force and their ability to protect the country in what is essentially a black box we don't know the workings of. Providing a copy of the source code and a development kit (if such a thing even exists for an airplane - I suspect if it does it won't bear much resemblance to, say, Linux) so we can audit it and continue development ourselves if necessary isn't an issue of "do we trust each other" - it strikes me as perfectly reasonable.