I suspect your domain hasn't been out there long enough yet.
My company's primary domain is registerd with technical contacts of "hostmaster@[our_domain.com]" and for years we never got a spam. Then about 2 years ago, somebody must have included it in a big master list; now it takes about 30-50 spams a day on average, mostly true "bottom feeder" crap like cialis and vicodin and *adult* crap.
My work email's been out there a lot longer, but doesn't draw nearly the number of spams and about 80% of them are financial/economic scams - mortgage and stock touts, lottery, 419, etc.
Upstream filters are blocking emails with virus attachments; I have no idea how many of those are coming in...
I don't think it has anything to do with intelligence, per se. I've seen an MD/PhD with an annual reseach grant total of $100 million struggle with this; I had to go to train his office manager on how to update his mail aliases, 'cause the mail client he liked was funky. If it wasn't in his alias list, and therefore clickable, he'd fubar it about 10% of the time and force the manager to fix it right now, 24/7, and he never understood anti-spam obfuscation (his staff filtered his incoming email for him).
His time was very valuable and he just wanted it to work.
Of course, the odds are good that nearly 50% of the people out there are of below-average intelligence, so any plan has to deal with both ends of the bell curve.
Sue Microsoft, your class stands to make a pretty penny.
So far as am aware, the only lawsuits wherein someone has (potentially) "made a pretty penny" from Microsoft have either been IP infringement cases (e.g. Eolas), or unlawful busines practice cases (e.g. Dr. DOS). If you know of *ANY* cases where MS has lost on a product liability issue, please share with the rest of the class.
Personally, considering that you would be suing someone with over a billion dollars cash on hand, who would have strong incentive to litigate to the death, I'd say your class more likely stands to piss away great sums in legal fees, with the prospect of zero return on investment. This and the "no warranty" hurdle is why, despite being one of the "deepest pockets" in the U.S., MS does not draw very many product liability nuisance suits. There's just no return it; MS won't settle, and it's very very very hard for the plaintiff to win.
I was not talking about the Gitmo detainees. I was talking about a U.S. citizen, arrested on U.S. soil in May of 2002, and transferred to military custody after civilan authorities could not present a legal charge.
Moore's business agent quoted Eisner, according to the NY Times. That that isn't "proper quoting" in journo-speak doesn't mean its not "quoting" in the everday vernacular. And yes, the reliability of that quote is subject to plenty of argument and interpretation, but that doesn't meant it's not a quote, just a questionable one.
"Hearsay" is a rule for the courtroom, not the court of public opinion. And in court, it's still hearsay when the NY Times quotes Eisner directly, rather than indirectly. In general, most of us have trusted the Times (and most other "respectable" journalistic sources) with being honest and more or less unbiased, but that seems to be changing, so I suspect we need to be critical of both direct and indirect quotes.
I agree that honest debate requires proper sourcing, and Moore's agent is not exactly a "reliable source" on the subject, in that he's highly interested in the success of the movie; however, we don't require witnesses in court to be disinterested either - we go with what we've got available. We listen to them, try to assess their biases, and then decide if the statements are credible or not.
Eisner is too smart to make a public statement on the issue; the best evidence we're going to get on Disney and Eisner's motives on the issue is going to be 2nd hand at best, and blatantly speculative at worst.
Personally, I find the statements attributed to Eisner fairly plausible, but the source is somewhat discreditable given his interests. My biases make me suspect it's a fairly accurate quote, but I wouldn't bet more then a buck on it...
And frankly, if true I don't blame Eisner for watching out for his stockholders, even though it wasn't in the best interests of open debate - that's not his job. (That major US media as a whole is "beholden to the stockholders" is a seperate problem/thread.)
The New York Times, on May 5th, 2004. It was a 2nd-hand quote, via Michael Moore's business agent, who was dealing with Disney and Mirimax on the film.
I know how our government's *supposed* to work, from the legislative process to little things like only Congress having the authority to declare war, not to mention that U.S. citizens are entitled to:
In all criminal prosecutions, the accused shall enjoy the right to a speedy and public trial...
instead of being held incommunicado for over two years, without even regular access to a lawyer.
Unfortunately, government process seems to be subject to a lot of "exceptions" lately.
Because a mediocre programmer will write code that's both slower and more expensive to maintain. In fact, a huge percentage of internal programming projects are *still* total failures, with absolutely no return on the investment.
In other words, cost is irrelevant, cost/benefit is what matters.
While the maxim is true, the "who cares" is not. With *poor* OO architectures, it's a lot easier to write programs that won't complete until well after the heat-death of the Universe. In such cases, a processor-derived halving of the runtime doesn't really help much.
Which is not to say that we should go back to stone knives and bearskins, just to avoid a wasted cycles, but rather to say that todays' latest toys work better with a thorough understanding of the basics, so that the OO solution is in fact "well architected", and that any definition of "well architected" has to take performance efficiencies into account, not just object elegance.
And by the way, there's a number of well-documented cases in the real world wherein increasing CPU performance actually degraded overall system throughput. This is particularly "not-uncommon" with complex database systems. See this PDF, originally published by Oracle.
You remember wrong on details, although you're right on the relevant principle. The "Berne" changes were generally more minor technical issues...
In 1976, effective Jan. 1, 1978, U.S. copyright law was substantially overhauled. Some of the major changes were to coordinate with international treaty, which required that you couldn't lose copyright over technicalities (e.g. failure to register).
Under current U.S. law:
A copyrightable work is copyrighted as soon as it is created.
You don't have to register a copyright, normally.
You DO have to register a copyright before you can sue someone for copyright infringement.
If you register your published work before someone infringes it (or within 3 months of publication), you can sue for attornies fees and "statutory damages" (an amount between $200 and $150,000, "as the court considers just").
If you haven't registered in time, you can only sue for "actual damages and profits". "Actual damages" is usually considered to be the lost business stolen by the infringer; it would be curious to see someone argue for a lost salary if the infringing headhunter cost someone a job.
A copyright owner can also request an injunction to prohibit infringment, and can ask the court to order impoundment of infringing copies or anything used to make infringing copies.
As usual, Eblen was spot-on in his comments. I'm a little disappointed he didn't attack the "only one backup allowed by copyright" urban legend as well, although that's the less pointed of the two reasons why the SCO position was complete, utter, total bushwa.
U.S. copyright law allows for only one backup copy
U.S. copyright law says no such thing; it explicitly (to those who know how courts interpret legal statutes) allows for multiple copies "for archival purposes only".
When a court looks at a law to decide what it means, the legal standard is that they are to presume that every word is meaningful. Since the law in question, 17 USC 117, says, in part:
all archival
copies are destroyed in the event that continued possession of the computer program should cease to be rightful.
and
Any exact
copies prepared in accordance with the provisions of this section may be leased, sold, or otherwise transferred, along with the copy from which such copies were prepared...
In other words, the law clearly expects people to have multiple backups and says they all have to be destroyed or transferred appropriately
SCO should refuse to pay that lawyer; he's clearly contradicting the plain meaning of the law he's attempting to reference.
Of course, when the goal is FUD, it doesn't have to make sense, it just has to "sound good".
Heise is claiming that the GPL contract (a license is a subspecies of contract) is invalid because it allows *more* copying than the (alleged) default limit of 1 backup in copyright law.
If his theory is true, then the obvious corollary is that all publishing contracts between a software author and a software publisher are also invalid, as they also allow the publishers to make more than one copy. This, of course, is stupid and wrong.
Actually, the law doesn't state how many backups you can make, the legal staute (17 USC 117) states:
Section 117. Limitations on exclusive rights: Computer programs
(a) Making of Additional Copy or Adaptation by Owner of Copy. -- Notwithstanding the provisions of section 106, it is not an infringement for the owner of a copy of a computer program to make or authorize the making of another copy or adaptation of that computer program provided:
(1) that such a new copy or adaptation is created as an essential step in the utilization of the computer program in conjunction with a machine and that it is used in no other manner, or
(2) that such new copy or adaptation is for archival purposes only and that all archival copies are destroyed in the event that continued possession of the computer program should cease to be rightful.
[... other stuff...]
In other words, it says, you can make "a copy", but it doesn't say "one copy", you can make a copy and then make a copy and then make a copy...; it also says "all archival copies [must be] destroyed". The legal standards for interpreting a statute say that all elements are presumed to have meaning; the "all archival copies" clause clearly envisions an owner with multiple archival copies. The omitted sub-section B also clearly states that multiple backup copies are permitted.
In other, other words, SCO is blowing really, really, really weak smoke.
Licenses are a sub-species of contracts; contracts are a formulation of a "meeting of the minds", i.e. an agreement between 2 or more parties. You can't have an agreement without intent.
That said, there are some exceptions, and there are some rules for digging out what the "intent" is behind a license or contract, and there are some principles for resolving ambiguity. For example, when there is an ambiguity in a contract between a very resource-rich entity and a resource-poor entity, the reasonable interpretation most favorable to the resource-poor entity is favored; the theory is that the rich/big/sophistacted entity more likely new what they were doing...
There are probably also some laws/rules about "intent" in a corporation. If SCO/Caldera had policies about keeping the Unix code out of Linux and enforced them on any developers they employed, but some "disgruntled" employee made significant efforts to break the rules (e.g. checked them in under an alias to an external project), then there was no "intent"; if they didn't have rules and told their Linux developers "just get it done", then they either had "intent" or were negligent with their property, and they lose.
For the case where some other developer took "their" code and put it in Linux, it's more complicated, I think. The "negligence" argument above is plausible; it'll definitely get a hearing in court. However, it's not clear to me that they had a duty to monitor Linux for "stolen" code coming in from the outside, I wouldn't be shocked if the courts distinguish re-distribution of the code from maintenance of the code, with redistributers not having as much responsibility for the content.
By analogy, if I accidentally buy a stolen computer in a lot of used PCs, the original owner can sue and get the machine back, but he can't sue me for any damages or costs because I accidentally ended up with his PC (unless I damaged it after I had notice that it wasn't legally mine). I don't have a positive duty to check that the PCs aren't stolen. However, if I sell that stolen PC in true good faith, the original owner in theory can't successfully sue me for anything (although I could be dragged into proceedings as a witness, and could be sued with the claim I knew or should have known the PC was stolen).
Extending my analogy a bit though, if I sell that stolen PC and the buyer loses posession of it to the original owner because it was stolen goods, then legally it wasn't mine to sell to the final buyer and he can sue me to recover his money.
This analogy leads me to wonder if:
assuming that "illegal" "Unix" code is found in Linux
assuming SCO/Caldera distributed that code
then users/buyers of SCO/Caldera linux could sue SCO
The theory is that SCO represented the product as a "legal" one and their customers who built businesses around that representation of their product would be hurt by that mis-representation. They in turn could sue their "suppliers" though...
The law is pretty clear, yes, the owner is entitled to recover stolen goods. In fact, in many cases an owner is entitled to recover stolen goods even if they've been resold honestly several times in the meantime. This comes up sometimes in the art world, with valuable works surfacing and returning to their original owners. For example, a couple of years ago, a stolen Wyeth painting came up at auction; it was seized and returned to Sears, the parent company of the gallery from which it was stolen. It had apprecieted from 30k to 500k in that time.
I'm not a lawyer and there are probably some exceptional cases; in particular, the line between "looting" and archeology (particularly when the archelogists are from another country/culture) seems to be moving around in the last 20 years or so, although this seems to mostly be handled at the diplomatic level...
In natural populations species is generally defined in terms of a naturally interbreeding population; two populations that can interbreed successfully, but normally don't are usually considered different species (if they're distinguishable).
The definition is not perfect; there are some amusing anomolous cases, e.g. where sub-group A breeds with B which breeds with C wich breeds with D, but A and D don't ever seem to interbreed, even given the chance.
Similar problems of classification come up in linguistics, e.g. the rhenish fan (an area of Germany/France/Belgium/Holland where the languages/dialects shade into one another). There's an inside joke that "a language is a dialect with an army" which contrasts popular notions of what a language is with the technical definition of a language as a set of mutually intelligeble dialects. By the technical standard, Swedish and Norwegian are one language... However, the technical definition breaks in the face of A nice survey of this language stuff is here
Thanks for the reference, I was vaguely aware of the transition from "grant+17" to "filing+20" but was never clear on the details.
As I read it, "old" patents such as the LZW get "greater of" and "new" patents get "filing + 20" (plus "injury time" if the PTO takes over 3 years to process the patent). Oy...
Personally, I don't see Unisys in the same class as SCO or the RIAA (or Rambus).
They owned a real patent to a relatively real invention (ignoring the argument about patentability of computational algorithms); their "sleazy" behaviour was to let people use the patent for free for a while, then change the rules and ask for money. A bit low, yes, but not true bottom-feeder behaviour, in my book. (Of course, I wasn't a developer who's livelihood was effected.)
For the newbies out there, the issue was that Unisys owned a patent to LZW compression. They initially issued free, liberal licenses and LZW became widely used, particularly in GIF software and compression tools. They then rescinded the old license and demanded royalties. Nobody challenged them legally (so far as I know there was no legitimate basis to do so). They successfully demanded patent royalties from a number of software makers.
The end of the story seems to be that the patent supposedly expires sometime about now, either last December or this June, depending on who you ask.
Slashdot Mirror
test@test.c_o_m probably sees a lot of action like this.
My company's primary domain is registerd with technical contacts of "hostmaster@[our_domain.com]" and for years we never got a spam. Then about 2 years ago, somebody must have included it in a big master list; now it takes about 30-50 spams a day on average, mostly true "bottom feeder" crap like cialis and vicodin and *adult* crap.
My work email's been out there a lot longer, but doesn't draw nearly the number of spams and about 80% of them are financial/economic scams - mortgage and stock touts, lottery, 419, etc.
Upstream filters are blocking emails with virus attachments; I have no idea how many of those are coming in...
His time was very valuable and he just wanted it to work.
Of course, the odds are good that nearly 50% of the people out there are of below-average intelligence, so any plan has to deal with both ends of the bell curve.
So far as am aware, the only lawsuits wherein someone has (potentially) "made a pretty penny" from Microsoft have either been IP infringement cases (e.g. Eolas), or unlawful busines practice cases (e.g. Dr. DOS). If you know of *ANY* cases where MS has lost on a product liability issue, please share with the rest of the class.
Personally, considering that you would be suing someone with over a billion dollars cash on hand, who would have strong incentive to litigate to the death, I'd say your class more likely stands to piss away great sums in legal fees, with the prospect of zero return on investment. This and the "no warranty" hurdle is why, despite being one of the "deepest pockets" in the U.S., MS does not draw very many product liability nuisance suits. There's just no return it; MS won't settle, and it's very very very hard for the plaintiff to win.
I was not talking about the Gitmo detainees.
I was talking about a U.S. citizen, arrested on U.S. soil in May of 2002, and transferred to military custody after civilan authorities could not present a legal charge.
Moore's business agent quoted Eisner, according to the NY Times. That that isn't "proper quoting" in journo-speak doesn't mean its not "quoting" in the everday vernacular. And yes, the reliability of that quote is subject to plenty of argument and interpretation, but that doesn't meant it's not a quote, just a questionable one.
I agree that honest debate requires proper sourcing, and Moore's agent is not exactly a "reliable source" on the subject, in that he's highly interested in the success of the movie; however, we don't require witnesses in court to be disinterested either - we go with what we've got available. We listen to them, try to assess their biases, and then decide if the statements are credible or not.
Eisner is too smart to make a public statement on the issue; the best evidence we're going to get on Disney and Eisner's motives on the issue is going to be 2nd hand at best, and blatantly speculative at worst.
Personally, I find the statements attributed to Eisner fairly plausible, but the source is somewhat discreditable given his interests. My biases make me suspect it's a fairly accurate quote, but I wouldn't bet more then a buck on it...
And frankly, if true I don't blame Eisner for watching out for his stockholders, even though it wasn't in the best interests of open debate - that's not his job. (That major US media as a whole is "beholden to the stockholders" is a seperate problem/thread.)
The New York Times, on May 5th, 2004. It was a 2nd-hand quote, via Michael Moore's business agent, who was dealing with Disney and Mirimax on the film.
Unfortunately, government process seems to be subject to a lot of "exceptions" lately.
In other words, cost is irrelevant, cost/benefit is what matters.
Which is not to say that we should go back to stone knives and bearskins, just to avoid a wasted cycles, but rather to say that todays' latest toys work better with a thorough understanding of the basics, so that the OO solution is in fact "well architected", and that any definition of "well architected" has to take performance efficiencies into account, not just object elegance.
And by the way, there's a number of well-documented cases in the real world wherein increasing CPU performance actually degraded overall system throughput. This is particularly "not-uncommon" with complex database systems. See this PDF, originally published by Oracle.
Integer math took another bit for sign, and performed operations with decimal arithmetic; binary 00101 + 00101 = 00000 (plus overflow).
Hard to believe you could get a playable 3-d tic-tac-toe in 20k...
Sheer genius!
You remember wrong on details, although you're right on the relevant principle. The "Berne" changes were generally more minor technical issues...
In 1976, effective Jan. 1, 1978, U.S. copyright law was substantially overhauled. Some of the major changes were to coordinate with international treaty, which required that you couldn't lose copyright over technicalities (e.g. failure to register).
Under current U.S. law:
As usual, Eblen was spot-on in his comments. I'm a little disappointed he didn't attack the "only one backup allowed by copyright" urban legend as well, although that's the less pointed of the two reasons why the SCO position was complete, utter, total bushwa.
U.S. copyright law says no such thing; it explicitly (to those who know how courts interpret legal statutes) allows for multiple copies "for archival purposes only".
When a court looks at a law to decide what it means, the legal standard is that they are to presume that every word is meaningful. Since the law in question, 17 USC 117, says, in part:
and In other words, the law clearly expects people to have multiple backups and says they all have to be destroyed or transferred appropriatelySCO should refuse to pay that lawyer; he's clearly contradicting the plain meaning of the law he's attempting to reference. Of course, when the goal is FUD, it doesn't have to make sense, it just has to "sound good".
If his theory is true, then the obvious corollary is that all publishing contracts between a software author and a software publisher are also invalid, as they also allow the publishers to make more than one copy. This, of course, is stupid and wrong.
Actually, the law doesn't state how many backups you can make, the legal staute (17 USC 117) states:
In other words, it says, you can make "a copy", but it doesn't say "one copy", you can make a copy and then make a copy and then make a copy...; it also says "all archival copies [must be] destroyed". The legal standards for interpreting a statute say that all elements are presumed to have meaning; the "all archival copies" clause clearly envisions an owner with multiple archival copies. The omitted sub-section B also clearly states that multiple backup copies are permitted.In other, other words, SCO is blowing really, really, really weak smoke.
This is turning into a laugher...
Intent is (almost) everything with licenses.
Licenses are a sub-species of contracts; contracts are a formulation of a "meeting of the minds", i.e. an agreement between 2 or more parties. You can't have an agreement without intent.
That said, there are some exceptions, and there are some rules for digging out what the "intent" is behind a license or contract, and there are some principles for resolving ambiguity. For example, when there is an ambiguity in a contract between a very resource-rich entity and a resource-poor entity, the reasonable interpretation most favorable to the resource-poor entity is favored; the theory is that the rich/big/sophistacted entity more likely new what they were doing...
There are probably also some laws/rules about "intent" in a corporation. If SCO/Caldera had policies about keeping the Unix code out of Linux and enforced them on any developers they employed, but some "disgruntled" employee made significant efforts to break the rules (e.g. checked them in under an alias to an external project), then there was no "intent"; if they didn't have rules and told their Linux developers "just get it done", then they either had "intent" or were negligent with their property, and they lose.
For the case where some other developer took "their" code and put it in Linux, it's more complicated, I think. The "negligence" argument above is plausible; it'll definitely get a hearing in court. However, it's not clear to me that they had a duty to monitor Linux for "stolen" code coming in from the outside, I wouldn't be shocked if the courts distinguish re-distribution of the code from maintenance of the code, with redistributers not having as much responsibility for the content.
By analogy, if I accidentally buy a stolen computer in a lot of used PCs, the original owner can sue and get the machine back, but he can't sue me for any damages or costs because I accidentally ended up with his PC (unless I damaged it after I had notice that it wasn't legally mine). I don't have a positive duty to check that the PCs aren't stolen. However, if I sell that stolen PC in true good faith, the original owner in theory can't successfully sue me for anything (although I could be dragged into proceedings as a witness, and could be sued with the claim I knew or should have known the PC was stolen).
Extending my analogy a bit though, if I sell that stolen PC and the buyer loses posession of it to the original owner because it was stolen goods, then legally it wasn't mine to sell to the final buyer and he can sue me to recover his money.
This analogy leads me to wonder if:
- assuming that "illegal" "Unix" code is found in Linux
- assuming SCO/Caldera distributed that code
- then users/buyers of SCO/Caldera linux could sue SCO
The theory is that SCO represented the product as a "legal" one and their customers who built businesses around that representation of their product would be hurt by that mis-representation. They in turn could sue their "suppliers" though...The law is pretty clear, yes, the owner is entitled to recover stolen goods. In fact, in many cases an owner is entitled to recover stolen goods even if they've been resold honestly several times in the meantime. This comes up sometimes in the art world, with valuable works surfacing and returning to their original owners. For example, a couple of years ago, a stolen Wyeth painting came up at auction; it was seized and returned to Sears, the parent company of the gallery from which it was stolen. It had apprecieted from 30k to 500k in that time.
I'm not a lawyer and there are probably some exceptional cases; in particular, the line between "looting" and archeology (particularly when the archelogists are from another country/culture) seems to be moving around in the last 20 years or so, although this seems to mostly be handled at the diplomatic level...
The definition is not perfect; there are some amusing anomolous cases, e.g. where sub-group A breeds with B which breeds with C wich breeds with D, but A and D don't ever seem to interbreed, even given the chance.
Similar problems of classification come up in linguistics, e.g. the rhenish fan (an area of Germany/France/Belgium/Holland where the languages/dialects shade into one another). There's an inside joke that "a language is a dialect with an army" which contrasts popular notions of what a language is with the technical definition of a language as a set of mutually intelligeble dialects. By the technical standard, Swedish and Norwegian are one language... However, the technical definition breaks in the face of A nice survey of this language stuff is here
So is, like, gravity, dude...
Thanks for the reference, I was vaguely aware of the transition from "grant+17" to "filing+20" but was never clear on the details.
As I read it, "old" patents such as the LZW get "greater of" and "new" patents get "filing + 20" (plus "injury time" if the PTO takes over 3 years to process the patent). Oy...
SCO has the 43 million cap; it's tripled since last summer, most of that recently...
They owned a real patent to a relatively real invention (ignoring the argument about patentability of computational algorithms); their "sleazy" behaviour was to let people use the patent for free for a while, then change the rules and ask for money. A bit low, yes, but not true bottom-feeder behaviour, in my book. (Of course, I wasn't a developer who's livelihood was effected.)
The end of the story seems to be that the patent supposedly expires sometime about now, either last December or this June, depending on who you ask.
For the curious, here's a link and here's another