The difference between contract and license:
A license allows one to do something that otherwise would be illegal, such as drive, perform brain surgery, distribute someone else's copyrighted work. In both contracts and licenses, considerations may change hands - such as money. But a contract can NEVER make an illegal act legal unless license rights are part of the considerations changing hands. (See SCO v. World)
Contracts also must be agreed to by all parties. GPL is a license because I can release my works under the GPL with no else's consent. Software released to users under the GPL do not require "click thru" agreements because they are pure license, not contracts. Microsoft's "End User License Agreement" is a contract, because it requires the user's agreement. That's why such horrific "EULA"s have a shred of legitimacy. Breach of the GPL (or any copyright) is called infringement, not violation. Copyright infringers can NOT be sued for breach of contract because they haven't signed a contract. If they haven't complied with the license terms, then they can not legally distribute dirative works. Period. No other legal considerations involved.
I used this to explain GPL vs proprietary to a non-techy friend:
Suppose I owned a good fishing spot coupla acres big. Friend Joe asks if he can fish from there, I say "sure, but share". He's ok with that. He comes back, says it's great, but his wife doesn't like to get her feet wet...would I build a dock? No, but he can, with this provision: He must allow anyone to use the dock, if I let them use the land. I promise that no one will get exclusive use of the land. Joe's ok with that.
Fred loves it too, but is tired of dragging his BBQ equip out there twice a year, and asks to build a brick grill. Same rules: All who share the land share the dock and the grill.
Time goes on, and we all get the use of each other's efforts in making a good fishing spot into a great getaway.
Each thing added to the land belongs to the person who added it. If anyone doesn't like the rules I run my land by, they are free to go somewhere else. They just can't use my fishing spot if they won't share. The diff with software under the GPL: Everyone can simultaneously use a virtual BBQ. Physical laws mean the BBQ at the fishing hole has be shared by timeslice, and that I can't share with EVERYONE, only a few dozen or so buddies.
"Unclassified" doesn't mean valueless or null, etc. "Classified" data is that which has an impact (of varying degrees) on national security. "Unclassified" is all data that isn't in one of the "Classified" categories.
By itself or in normal amounts or normal handling, Unclassified has no impact on nat'l security. Nat'l security has nothing to do with what *the*company* considers important.
Examples: almost all contractor's business info that doesn't overlap class. stuff, source code to company tools, blueprints for the company's buildings, etc.
"Unclassified" includes personal info and company trade secrets, etc. Ever hear of "identity theft"?
So yes, "unclassified" almost surely can be "stolen", in any country.
of the Dylan song from the 1960's "The Mighty Quinn". Ev'rybody's 'neath the trees,
Feeding pigeons on a limb
But when Quinn the Eskimo gets here,
All the pigeons gonna run to him.
A cat's meow and a cow's moo, I can recite 'em all,
Just tell me where it hurts yuh, honey,
And I'll tell you who to call. ...
It's about a pusher and his clients.
free speech in any form, especially on the internet, will just route around censorship one way or the other.
When a website uses my bandwidth to transfer data I don't want, they're stealing from me.
It's immoral to call it "stealing" when I'm protecting a resource I pay for. That's good economic sense, not "stealing".
I'm going to file for a patent on polygonal wheels. By making the polygon's sides nonlinearly bulge outward, following the formula (pi)*r**2, a smoother rolling experience may be enjoyed. This enhancement may be done to polygonal wheels that have any number of sides.
? Such as Unix's ~./* ? This feature has been a part of EVERY shared computer system I've ever worked on, and I started programming in 1976. It was considered old hat, even then. I saw nothing in TFA that distinguishes this from normal home dir usage. Did I miss something?
As I quoted, I was replying to the "clearly stated" part of your comment, which is ontopic as a reply to your post. MS's claims of patent infringement, without supplying patent numbers, is one example of MS's "intellectual property" FUD, which makes your comment also ontopic. And I did RTFP.(PS: check sig, I think you have me confused with a grandpa poster)
Using either term "IP" or "235 patents" to accuse someone of infringement should be followed immediately with the specifics, in order to clarify (prove) or work towards a resolution.
It's like using the term "titled property", which would include cars, real estate, etc. If someone is accused of burning "titled property", are they accused of burning a car or a house, or something else altogether?
If talking about whether a company is a generally good investment, "Intellectual property", "235 patents being infringed", "titled property", etc, are probably all useful terms. When accusing a specific group hardworking people (Linux developers, in this case) of infringing copyrights, patents, etc, you'd better be specific or expect to treated harshly.
clearly stated that their patents are infringed
OK. Tell me ONE patent number identified by Microsoft as being infringed by Linux. Just one. They claim hundreds, just tell me ONE of them.
Only then can you claim Microsoft "clearly" stated their patents are infringed. Without the patent numbers, the statements are vague, they're FUD and they're BS.
The atom bombs dropped on Japan fit into B29 bombers (one each). That limits their upper weight to probably a couple tons. The Saturn V can accelerate 100 tons much of the way to the 60 miles high and 17,000 mph needed for orbit. If it was going to be used to pop Israel from Iran, (just to pick a coupla countries at random), it could probably move 300-400 tons worth - no need to kick it that high or fast. Think approx 150 Hiroshima-sized bombs on Israel - the ground zeroes would overlap.
"Overkill" seems an inadequate word. Keep in mind also that it took a huge chunk of the US GDP (not federal budget - gross domestic product) to achieve the moon shots the Saturns were built for. No "terrorist" country is gonna be able to build anything that big for the next 1000 years. The Saturns are dinosaurs - impressive, huge, and history.
That's what I was tapping about - the thought that the money "comes from somewhere else" when the burden is shifted from US taxpayers to airlines, who pass it on to passengers, who are...taxpayers (tho not all are US). This example is unusual in that airlines do so much business outside the US, but I think you understand: No matter where the money "comes from", ultimately, it's coming from individuals.
There's still some contention about where the funding will come from.
Actually, there's absolutely no doubt where the money will come from - we the people. The contention is whose hands it will go thru first before the system is complete. The "who pays for it" question is a distraction in many, many public projects, such as "who pays for a cleaner environment?" "Who pays for (existing | preventing) illegal immigration?", etc.
In a way, it can be said that governments and companies have no money at all, except that which they receive from individuals. For example, car makers objected massively to adding airbags, and one excuse they pulled up was cost. But who pays for every part of the car when it's bought? Car makers? uh, no. Every added cost to everything is always passed on to the people who buy products and use services. It must be, or the companies providing products and services would eventually go out of business. The "who pays for it" debate is always part of the push and shove of hogs eating out of the gov't trough. Sadly, most people don't get this at all.
On my Apple ][, I changed every keyword in the BASIC interpreter, and added a few more. I read in Byte magazine how to do it, and was learning 6502 Asm at the time. Doesn't really count as a 'virus', but if you formatted a disk from one of mine, you had to learn a new 'dialect' of Basic.
Observations: > the scan s/w itself is not FOSS. > OpenLogic, who issued the press release (it's NOT a news story) seems to be focussed on managing FOSS dists in the enterprise. > According to their website, OpenLogic is a Microsoft "technology partner" ("indentured servant"). ditto with Novell. But they're also partners with redhat. Tainted, but there's hope.
They do seem to be paying for and building up a developer community, but I didn't see a list of what projects they're working on or who's helping them. They may be too new to have anything going yet.
I don't see how this can work in a shop with developers compiling their own, although it would be easy to scan for pre-compiled binaries.
Given the EULAs that are being added to non-FOSS packages lately, I'd worry lots more about them than FOSS. This goes double for the properly purchased and licensed proprietary S/W. It seems like licensing some vendors' software is just a way to put yourself on their litigation radar. Screw that - I want FOSS just for the legal simplicity.
"Ideal world" nothing. True industry standards are a reality in many industries including the software industry (ASCII, TCP/IP, FORTRAN(1), C, etc). "Ambiguities" should NEVER be written into a standard. As TFA says, ODF working groups couldn't finish the formulas for the standard, so that part was omitted from the standard for the time being.
The latest NIST work [...] confirms the institute's 1998 results using the same method while reducing the measurement uncertainty by about 40 percent, thanks mainly to improvements in the hardware used in the experiments.
The spec is "hard" because it is constantly refined to real-world acheivable precision, adding a few more decimal places every few years. This sometimes requires re-defining it as some real-world item that can be exactly reproducedanywhere. Precise laboratory definition provides more than a standard measure of weight, it also provides a standard measure of the quality of the laboratories that work to the standard. The whole purpose of any standard is (should be) to remove ambiguity.
Real standards are NOT hostage to the whim of a single company, but instead are guided by the whole industry. By referencing Office, MS can change the "standard" without going thru a standards process or industry body. And MS can do it without notice, since many of MS' licenses allow unnotified software updates.
Saying that the standard will "reference MS office" is no different than saying that MS Office IS the standard.
1."Consistently separating words by spaces became a general custom about the tenth century A.D., and lasted until about 1957, when FORTRAN abandoned the practice." --Sun FORTRAN Reference Manual
Absolutely not. A standards specification should stand on its own, or reference other standards. "MS office" is an implementation, not a standard. It can't be used to define a standard any more than the wheelbase of your car can define what a roadway should be. Further, if ooxml is as "free" as MS would have politicians believe, then referring back to a proprietary product destroys that "freedom". (It's really not free, anyway, but just for the sake of discussion...)
To successfully rig an election...
I was addressing the part that went "I don't think that Microsoft would take the chance of...", not what it would take to do so. I agree, it would be complex. With programmable voting machines, it would first require technical hooks into the machinary. And that makes the morality of MS a relevant issue.
It's all just a dream, anyway. I doubt elected officials would care or think the risk great enough to do anything this drastic.
And as for guarantees with Access databases, allow me;) to quote Microsoft: To the maximum extent permitted by law, Microsoft... disclaim[s] all... warranties... including... fitness for a particular purpose (From the Access 2000 EULA) Not just a little bit of uncertainty in their product, but the maximum uncertainty allowed by law.
Some guys (I can't find cite, sorry) came up with a really cool verification scheme. Machine prints two copies of the voter's ballot, one for the voter, and one that gets stored and counted. The voter's ballot is transparent except for the printing. The ballot paper is printed with a dot-matrix code. Normally, dot-matrix text is (for example) 9 * 16 dots. In this dot-matrix, however, each dot is subdivided into nine or more smaller dots. The 9*16 matrix is edge to edge. Each character printed only gets a few of the smaller dots, chosen by encrypted method. If you eyeball the ballot, all you see is white noise.
The characters on the two ballots are a one's complement of each other. The two ballots must be laid one over the other to correctly form the letters of the voter's choices. The voter could then verify, over the internet, that what the central counting system received matches the ballot the voter placed. The central system's website would put up a lifesize graphic of the ballot received and counted. The voter would place his/her transparent ballot directly on the monitor, overlaying the image. Together, the dots would line up and the voter's choice is clearly spelled out. The voter's choice is kept secret; the encrypted ballot is hard to fake, impossible to replace (if the voter verifies); the overall system is verifiable at every level. And the actual vote is still paper.
Agreed. At most, tabulating machines to count'em up, simple pulse coded numerical uploads to a central tabulator. But those machines don't even havta be programmable.
Sorry for the flame-baitish subject, but it had to be said...
I don't think that Microsoft would take the chance of attempting to rig an election
This is the company that was caught falsifying evidence right in front of a federal judge. Would they rig an election behind the scenes? Without a second thought.
The problem isn't so much that Microsoft won't open their code
Yes, this is a problem. Microsoft's customers are its business partners, not the people who buy computers. MS has been putting 'hooks' in their code for years to allow marketing access to the millions of MS users worldwide. There's a very reasonable probability that MS has similar hooks for gov't access. Gov't access to the voting booth and the ballots is a very, very dangerous thing. And yes, non-gov't black hats are also a worry.
As a unique hardware based design (code in PROM) it'd be very difficult to compromise
This is called 'security thru obscurity' and no, it doesn't work, even a little bit. All that's required is for one copy of the machine to be made publicly available. This has been proven in real world experiences time and again.
Diebold ATM machines seem to be very, very secure
HOW do you know? Please offer citations, including links to skilled, unbiased research. Diebold fought (and still fights) so hard to prevent its voting machines from being researched that I'd be surprised to find any trustworthy research on their ATMs.
...vendors allow their source code to be viewed,... only a few special people will get to look at it
The obvious answer here of course is open source code. Don't wait on Diebold or someone else to write it; we the people need to. I have no people-organizing skills of my own, but I am a good programmer. I'd be willing to put many, many free hours into this, if there were a project out there doing it.
The point of Open Source is that the service is what is being sold, not the software. It's just like getting work done at a machine shop or auto repair shop - the parts are less significant than the service. And Dell's service in this case sounded like it sucked.
Slashdot Mirror
The difference between contract and license: A license allows one to do something that otherwise would be illegal, such as drive, perform brain surgery, distribute someone else's copyrighted work. In both contracts and licenses, considerations may change hands - such as money. But a contract can NEVER make an illegal act legal unless license rights are part of the considerations changing hands. (See SCO v. World)
Contracts also must be agreed to by all parties. GPL is a license because I can release my works under the GPL with no else's consent. Software released to users under the GPL do not require "click thru" agreements because they are pure license, not contracts.
Microsoft's "End User License Agreement" is a contract, because it requires the user's agreement. That's why such horrific "EULA"s have a shred of legitimacy.
Breach of the GPL (or any copyright) is called infringement, not violation.
Copyright infringers can NOT be sued for breach of contract because they haven't signed a contract. If they haven't complied with the license terms, then they can not legally distribute dirative works. Period. No other legal considerations involved.
I used this to explain GPL vs proprietary to a non-techy friend: Suppose I owned a good fishing spot coupla acres big. Friend Joe asks if he can fish from there, I say "sure, but share". He's ok with that. He comes back, says it's great, but his wife doesn't like to get her feet wet...would I build a dock? No, but he can, with this provision: He must allow anyone to use the dock, if I let them use the land. I promise that no one will get exclusive use of the land. Joe's ok with that.
Fred loves it too, but is tired of dragging his BBQ equip out there twice a year, and asks to build a brick grill. Same rules: All who share the land share the dock and the grill.
Time goes on, and we all get the use of each other's efforts in making a good fishing spot into a great getaway.
Each thing added to the land belongs to the person who added it. If anyone doesn't like the rules I run my land by, they are free to go somewhere else. They just can't use my fishing spot if they won't share.
The diff with software under the GPL: Everyone can simultaneously use a virtual BBQ. Physical laws mean the BBQ at the fishing hole has be shared by timeslice, and that I can't share with EVERYONE, only a few dozen or so buddies.
"Unclassified" doesn't mean valueless or null, etc. "Classified" data is that which has an impact (of varying degrees) on national security. "Unclassified" is all data that isn't in one of the "Classified" categories.
By itself or in normal amounts or normal handling, Unclassified has no impact on nat'l security. Nat'l security has nothing to do with what *the*company* considers important. Examples: almost all contractor's business info that doesn't overlap class. stuff, source code to company tools, blueprints for the company's buildings, etc.
"Unclassified" includes personal info and company trade secrets, etc. Ever hear of "identity theft"?
So yes, "unclassified" almost surely can be "stolen", in any country.
of the Dylan song from the 1960's "The Mighty Quinn".
...
Ev'rybody's 'neath the trees,
Feeding pigeons on a limb
But when Quinn the Eskimo gets here,
All the pigeons gonna run to him.
A cat's meow and a cow's moo, I can recite 'em all,
Just tell me where it hurts yuh, honey,
And I'll tell you who to call.
It's about a pusher and his clients.
free speech in any form, especially on the internet, will just route around censorship one way or the other.
When a website uses my bandwidth to transfer data I don't want, they're stealing from me.
It's immoral to call it "stealing" when I'm protecting a resource I pay for. That's good economic sense, not "stealing".
I'm going to file for a patent on polygonal wheels. By making the polygon's sides nonlinearly bulge outward, following the formula (pi)*r**2, a smoother rolling experience may be enjoyed.
This enhancement may be done to polygonal wheels that have any number of sides.
? Such as Unix's ~./* ? This feature has been a part of EVERY shared computer system I've ever worked on, and I started programming in 1976. It was considered old hat, even then.
I saw nothing in TFA that distinguishes this from normal home dir usage. Did I miss something?
As I quoted, I was replying to the "clearly stated" part of your comment, which is ontopic as a reply to your post. MS's claims of patent infringement, without supplying patent numbers, is one example of MS's "intellectual property" FUD, which makes your comment also ontopic. And I did RTFP.(PS: check sig, I think you have me confused with a grandpa poster)
Using either term "IP" or "235 patents" to accuse someone of infringement should be followed immediately with the specifics, in order to clarify (prove) or work towards a resolution.
It's like using the term "titled property", which would include cars, real estate, etc. If someone is accused of burning "titled property", are they accused of burning a car or a house, or something else altogether?
If talking about whether a company is a generally good investment, "Intellectual property", "235 patents being infringed", "titled property", etc, are probably all useful terms. When accusing a specific group hardworking people (Linux developers, in this case) of infringing copyrights, patents, etc, you'd better be specific or expect to treated harshly.
clearly stated that their patents are infringed
OK. Tell me ONE patent number identified by Microsoft as being infringed by Linux. Just one. They claim hundreds, just tell me ONE of them.
Only then can you claim Microsoft "clearly" stated their patents are infringed.
Without the patent numbers, the statements are vague, they're FUD and they're BS.
Even checking "Read Only" would irreversibly damage the contents
Maybe this research is funded by Microsoft.
The only conceivable risk is if blueprints for the factories for building the technology is leaked. Even then, half-century-old rocket tech? nah.
The atom bombs dropped on Japan fit into B29 bombers (one each). That limits their upper weight to probably a couple tons. The Saturn V can accelerate 100 tons much of the way to the 60 miles high and 17,000 mph needed for orbit. If it was going to be used to pop Israel from Iran, (just to pick a coupla countries at random), it could probably move 300-400 tons worth - no need to kick it that high or fast. Think approx 150 Hiroshima-sized bombs on Israel - the ground zeroes would overlap.
"Overkill" seems an inadequate word. Keep in mind also that it took a huge chunk of the US GDP (not federal budget - gross domestic product) to achieve the moon shots the Saturns were built for.
No "terrorist" country is gonna be able to build anything that big for the next 1000 years.
The Saturns are dinosaurs - impressive, huge, and history.
I wonder they're going to remove the working examples of Saturn V [pdf] rockets that seem to populate the landscape.
Oh, never mind, they moved Marshall space flight center to Kenya
WTF?
That's what I was tapping about - the thought that the money "comes from somewhere else" when the burden is shifted from US taxpayers to airlines, who pass it on to passengers, who are...taxpayers (tho not all are US). This example is unusual in that airlines do so much business outside the US, but I think you understand: No matter where the money "comes from", ultimately, it's coming from individuals.
There's still some contention about where the funding will come from.
Actually, there's absolutely no doubt where the money will come from - we the people. The contention is whose hands it will go thru first before the system is complete. The "who pays for it" question is a distraction in many, many public projects, such as "who pays for a cleaner environment?" "Who pays for (existing | preventing) illegal immigration?", etc.
In a way, it can be said that governments and companies have no money at all, except that which they receive from individuals. For example, car makers objected massively to adding airbags, and one excuse they pulled up was cost. But who pays for every part of the car when it's bought? Car makers? uh, no. Every added cost to everything is always passed on to the people who buy products and use services. It must be, or the companies providing products and services would eventually go out of business.
The "who pays for it" debate is always part of the push and shove of hogs eating out of the gov't trough. Sadly, most people don't get this at all.
On my Apple ][, I changed every keyword in the BASIC interpreter, and added a few more. I read in Byte magazine how to do it, and was learning 6502 Asm at the time. Doesn't really count as a 'virus', but if you formatted a disk from one of mine, you had to learn a new 'dialect' of Basic.
That would mean that Microsoft innovated SOMETHING in the modern era. Not possible. But, it may be the SECOND virus....
Observations:
> the scan s/w itself is not FOSS.
> OpenLogic, who issued the press release (it's NOT a news story) seems to be focussed on managing FOSS dists in the enterprise.
> According to their website, OpenLogic is a Microsoft "technology partner" ("indentured servant"). ditto with Novell. But they're also partners with redhat. Tainted, but there's hope.
They do seem to be paying for and building up a developer community, but I didn't see a list of what projects they're working on or who's helping them. They may be too new to have anything going yet.
I don't see how this can work in a shop with developers compiling their own, although it would be easy to scan for pre-compiled binaries.
Given the EULAs that are being added to non-FOSS packages lately, I'd worry lots more about them than FOSS. This goes double for the properly purchased and licensed proprietary S/W. It seems like licensing some vendors' software is just a way to put yourself on their litigation radar. Screw that - I want FOSS just for the legal simplicity.
Kilogram: good example. The spec is "hard" because it is constantly refined to real-world acheivable precision, adding a few more decimal places every few years. This sometimes requires re-defining it as some real-world item that can be exactly reproduced anywhere. Precise laboratory definition provides more than a standard measure of weight, it also provides a standard measure of the quality of the laboratories that work to the standard. The whole purpose of any standard is (should be) to remove ambiguity.
Real standards are NOT hostage to the whim of a single company, but instead are guided by the whole industry. By referencing Office, MS can change the "standard" without going thru a standards process or industry body. And MS can do it without notice, since many of MS' licenses allow unnotified software updates.
Saying that the standard will "reference MS office" is no different than saying that MS Office IS the standard.
1."Consistently separating words by spaces became a general custom about the tenth century A.D., and lasted until about 1957, when FORTRAN abandoned the practice." --Sun FORTRAN Reference Manual
Absolutely not. A standards specification should stand on its own, or reference other standards. "MS office" is an implementation, not a standard. It can't be used to define a standard any more than the wheelbase of your car can define what a roadway should be.
Further, if ooxml is as "free" as MS would have politicians believe, then referring back to a proprietary product destroys that "freedom". (It's really not free, anyway, but just for the sake of discussion...)
To successfully rig an election... ...", not what it would take to do so. I agree, it would be complex. With programmable voting machines, it would first require technical hooks into the machinary. And that makes the morality of MS a relevant issue. ;) to quote Microsoft: ... disclaim[s] all ... warranties ... including ... fitness for a particular purpose
I was addressing the part that went "I don't think that Microsoft would take the chance of
It's all just a dream, anyway. I doubt elected officials would care or think the risk great enough to do anything this drastic.
And as for guarantees with Access databases, allow me
To the maximum extent permitted by law, Microsoft
(From the Access 2000 EULA)
Not just a little bit of uncertainty in their product, but the maximum uncertainty allowed by law.
Some guys (I can't find cite, sorry) came up with a really cool verification scheme. Machine prints two copies of the voter's ballot, one for the voter, and one that gets stored and counted. The voter's ballot is transparent except for the printing.
The ballot paper is printed with a dot-matrix code. Normally, dot-matrix text is (for example) 9 * 16 dots. In this dot-matrix, however, each dot is subdivided into nine or more smaller dots. The 9*16 matrix is edge to edge. Each character printed only gets a few of the smaller dots, chosen by encrypted method. If you eyeball the ballot, all you see is white noise.
The characters on the two ballots are a one's complement of each other. The two ballots must be laid one over the other to correctly form the letters of the voter's choices. The voter could then verify, over the internet, that what the central counting system received matches the ballot the voter placed. The central system's website would put up a lifesize graphic of the ballot received and counted. The voter would place his/her transparent ballot directly on the monitor, overlaying the image. Together, the dots would line up and the voter's choice is clearly spelled out.
The voter's choice is kept secret; the encrypted ballot is hard to fake, impossible to replace (if the voter verifies); the overall system is verifiable at every level. And the actual vote is still paper.
Agreed. At most, tabulating machines to count'em up, simple pulse coded numerical uploads to a central tabulator. But those machines don't even havta be programmable.
Sorry for the flame-baitish subject, but it had to be said...
...vendors allow their source code to be viewed, ... only a few special people will get to look at it
I don't think that Microsoft would take the chance of attempting to rig an election
This is the company that was caught falsifying evidence right in front of a federal judge. Would they rig an election behind the scenes? Without a second thought.
The problem isn't so much that Microsoft won't open their code
Yes, this is a problem. Microsoft's customers are its business partners, not the people who buy computers. MS has been putting 'hooks' in their code for years to allow marketing access to the millions of MS users worldwide. There's a very reasonable probability that MS has similar hooks for gov't access. Gov't access to the voting booth and the ballots is a very, very dangerous thing. And yes, non-gov't black hats are also a worry.
As a unique hardware based design (code in PROM) it'd be very difficult to compromise
This is called 'security thru obscurity' and no, it doesn't work, even a little bit. All that's required is for one copy of the machine to be made publicly available. This has been proven in real world experiences time and again.
Diebold ATM machines seem to be very, very secure
HOW do you know? Please offer citations, including links to skilled, unbiased research. Diebold fought (and still fights) so hard to prevent its voting machines from being researched that I'd be surprised to find any trustworthy research on their ATMs.
The obvious answer here of course is open source code. Don't wait on Diebold or someone else to write it; we the people need to. I have no people-organizing skills of my own, but I am a good programmer. I'd be willing to put many, many free hours into this, if there were a project out there doing it.
The point of Open Source is that the service is what is being sold, not the software. It's just like getting work done at a machine shop or auto repair shop - the parts are less significant than the service.
And Dell's service in this case sounded like it sucked.