You don't want to (unless, say, your organization grows rapidly and you need to migrate to a database and platform you couldn't previously afford), but what kind of leverage do you think you'll have with the vendor when it becomes known that you can't?
Sure, many databases have crummy optimization and need to be coddled with proprietary query syntax, but failing to bundle a tested ANSI-conforming equivalent is negligent.
A corporation is under ZERO obligation to inform you of your rights or even to tell the truth about what you may or may not do in the license.
If I did have the right to do X, but the license says I didn't, and their salespeople said I have to pay $Y extra for the right to do X, they'd be lying to me as an inducement to buy. Why isn't that fraud?
Good in theory, but HTML support in Office 2000 was such a debacle that there are third-party tools designed just to unmangle the markup. They compltely ignored Processing Instruction syntax, which is intended to do just what they wanted, and
Gah. Even if your document is served with charset=windows-1252 (so that a literal 0x85 byte in the document should be interpreted as U+2026 ELLIPSIS, HORIZONTAL), character entities like always refer to Unicode codepoints. … or … can only mean U+0085 NEXT LINE no matter what charset the document is in. Do browsers still screw this up?
XML allows line breaks inside tags, but since EBCDIC platforms are likely to embed NEL characters when you do this, XML 1.0 parsers should reject such documents--even though what you typed would have been well-formed if you had been working in US-ASCII. Parsers are backward compatible, but documents are not.
The marginal cost of copying software is damn near zero. When the development cost is amortized across several versions and related R&D or products, how do you decide whether the vendor is fairly recouping their cost or just poorly predicted the price elasticity of demand for it?
... which would be somewhat workable if the USPTO were acknowledged by all to be merely a registry, and questions of validity were always answered in courts. IIRC Australian patents work this way.
Instead, US courts have a strong presumption of validity about all the sh^H^Houtput from the patent office. I've even seen it argued (by RMS, I think) that prior art databases are a bad idea because only prior art that wasn't already considered by the (overworked and wrongly incented) patent examiner can really help overturn an invalid patent.
Microsoft does have a standard installer framework (if you've ever seen a MSI file, this is how it works). Many products still don't use it because it wasn't bundled until WinXP and many users haven't downloaded it as an add-on. It can stub out components and then really install them on first use, which is fairly cool over a reliable network (but an awful nuisance if you installed off a CD the BOFH then locks up).
Ashcroft probably isn't personally involved with the case at all. As pointed out upthread, you can't sue Congress so instead you sue the attorney general simply to get an injunction preventing them (and apparently all their subordinates) from enforcing that law.
If that's true, Mozilla seems to be the only team that uses Bugzilla but doesn't coordinate work on suggestions with it, and they probably ought to remove the "enhancement" bug severity as inappropriate.
Non-compete agreements signed by software developers have been held invalid/unenforceable if they leave former employees unable to practice their profession. Why isn't the same thing happening in the recording industry?
Calling an argument "semantics" is just a pejorative way of admitting that words have meanings. Governments needed to pass new laws against copyright infringement precisely because it isn't theft--if it were depriving anyone of their property, it would have been forbidden ever since our civilization adopted the rule of law.
Our society holds that private property is an inalienable right, but copyright is merely a new pragmatic bargain with creators. If you're going to rave about moral obligations, ask yourself whether society is still being well served by the variety of restrictions we are allowing creators to impose on us all, because whatever we're expecting in exchange is the only thing that makes copyright infringement immoral.
In the US, you can sue for actual damages and profits even if you register after the infringement and more than three months after publication. Once you've registered you can sue for statutory damages (amounts set by law, up to US$100,000) for any future infringement.
A decompiled binary is far from being the "preferred form for making modifications" required by the GPL--it lacks the original source's comments, variable names, inline function definitions, and clear control flow. And you'd have to port the code back to using libraries' type definitions even to get it to run on another architecture....
If there's a vendor out there willing and able to meet all your needs at a reasonable price, so that you can remain comfortable as a mere end user totally at their mercy, congratulations.
Custom development is expensive, which is why you pool your money with other clients who need the same work done. It could even be arranged through the vendor, as long as they don't set the price based on having a monopoly on the ability to do the work.
I do code, but if I didn't (and will never learn), I want to hire someone to modify the driver and make the printer do what I want or use it with a system Epson never heard of (if they're still in business, they'd probably want me to have to buy another printer instead).
<analogy class=standard> I'm not a mechanic, but I still wouldn't buy a car with the hood welded shut. </>
They don't refuse to license for $MINORITY_GROUP or for use in $DISTASTEFUL_INDUSTRY, but they aren't required to accomodate business models that rely on limiting the competing options your customers have.
The whole point [of open source] is that you do not have commercial people developing the code.
Bull! The point is that nobody has a monopoly on developing the code. Work for hire is welcome, and we're even starting to see it happen more often (and I can't wait for task markets to hit the big time).
There are wacky IA-32 opcodes like ENTER that affect several registers and memory locations. There are even circumstances (like CALL with a task gate) that affect all the registers and the page tables. The instruction set docs have huge pseudocode in places, and no reasonable pipeline can handle that without resorting to micro-ops.
The corporation holds copyright to my work, but both my manager and I have exactly as much authority as the chief officers have chosen to delegate to us--there is no legal requirement that they make every such decision personally.
There are no consequences to using GPL'd code in my work unless and until we decide to redistribute and license the work to third parties under incompatible terms.
We've all heard rumblings about age bias in programming--they'd often rather underpay an desperate, overcaffeinated twenty-year-old than have to listen to that old cuss who can explain why the project is doomed. As a colleague (hi Dan!) put it, the two career paths for a programmer are
graybeard
management
You can stay focused on coding if you're the local guru who knows the systems inside and out (in fact they might be afraid to try to go on without you), but there's always the risk your expertise will suddenly be obsolete, and many companies don't even realize how much they needed one until they get rid of them.
At 31, I'm slowly picking up project management (mostly by choosing jobs likely to let me do some, and reading), because it's easier to sell (nobody sane thinks they don't need it) and age mostly improves credibility. And somebody's got to oppose the industry's gratuitous complexity (or "cover fire", as Joel Spolsky puts it) in favor of the simplest thing that works.
Slashdot Mirror
SQL/PSM (stored procedures for ANSI SQL) was ratified six years ago, though implementations are still notably absent.
You don't want to (unless, say, your organization grows rapidly and you need to migrate to a database and platform you couldn't previously afford), but what kind of leverage do you think you'll have with the vendor when it becomes known that you can't?
Sure, many databases have crummy optimization and need to be coddled with proprietary query syntax, but failing to bundle a tested ANSI-conforming equivalent is negligent.
If I did have the right to do X, but the license says I didn't, and their salespeople said I have to pay $Y extra for the right to do X, they'd be lying to me as an inducement to buy. Why isn't that fraud?
Gah. Even if your document is served with charset=windows-1252 (so that a literal 0x85 byte in the document should be interpreted as U+2026 ELLIPSIS, HORIZONTAL), character entities like always refer to Unicode codepoints. … or … can only mean U+0085 NEXT LINE no matter what charset the document is in. Do browsers still screw this up?
XML allows line breaks inside tags, but since EBCDIC platforms are likely to embed NEL characters when you do this, XML 1.0 parsers should reject such documents--even though what you typed would have been well-formed if you had been working in US-ASCII. Parsers are backward compatible, but documents are not.
The marginal cost of copying software is damn near zero. When the development cost is amortized across several versions and related R&D or products, how do you decide whether the vendor is fairly recouping their cost or just poorly predicted the price elasticity of demand for it?
Instead, US courts have a strong presumption of validity about all the sh^H^Houtput from the patent office. I've even seen it argued (by RMS, I think) that prior art databases are a bad idea because only prior art that wasn't already considered by the (overworked and wrongly incented) patent examiner can really help overturn an invalid patent.
Where would we be if every use of the telephone or the radio could have been patented simply because they were too obvious to bother publishing about.
Microsoft does have a standard installer framework (if you've ever seen a MSI file, this is how it works). Many products still don't use it because it wasn't bundled until WinXP and many users haven't downloaded it as an add-on. It can stub out components and then really install them on first use, which is fairly cool over a reliable network (but an awful nuisance if you installed off a CD the BOFH then locks up).
Ashcroft probably isn't personally involved with the case at all. As pointed out upthread, you can't sue Congress so instead you sue the attorney general simply to get an injunction preventing them (and apparently all their subordinates) from enforcing that law.
IIRC a corporation's copyrights would have expired in 95 years (if Congress weren't going to extend them again).
If that's true, Mozilla seems to be the only team that uses Bugzilla but doesn't coordinate work on suggestions with it, and they probably ought to remove the "enhancement" bug severity as inappropriate.
Non-compete agreements signed by software developers have been held invalid/unenforceable if they leave former employees unable to practice their profession. Why isn't the same thing happening in the recording industry?
Calling an argument "semantics" is just a pejorative way of admitting that words have meanings. Governments needed to pass new laws against copyright infringement precisely because it isn't theft--if it were depriving anyone of their property, it would have been forbidden ever since our civilization adopted the rule of law.
Our society holds that private property is an inalienable right, but copyright is merely a new pragmatic bargain with creators. If you're going to rave about moral obligations, ask yourself whether society is still being well served by the variety of restrictions we are allowing creators to impose on us all, because whatever we're expecting in exchange is the only thing that makes copyright infringement immoral.
In the US, you can sue for actual damages and profits even if you register after the infringement and more than three months after publication. Once you've registered you can sue for statutory damages (amounts set by law, up to US$100,000) for any future infringement.
A decompiled binary is far from being the "preferred form for making modifications" required by the GPL--it lacks the original source's comments, variable names, inline function definitions, and clear control flow. And you'd have to port the code back to using libraries' type definitions even to get it to run on another architecture....
Custom development is expensive, which is why you pool your money with other clients who need the same work done. It could even be arranged through the vendor, as long as they don't set the price based on having a monopoly on the ability to do the work.
<analogy class=standard>
I'm not a mechanic, but I still wouldn't buy a car with the hood welded shut.
</>
They don't refuse to license for $MINORITY_GROUP or for use in $DISTASTEFUL_INDUSTRY, but they aren't required to accomodate business models that rely on limiting the competing options your customers have.
The good news is Apple wasn't foolish enough to abandon NeXT's IA-32 port. The bad news is they won't sell it.
Bull! The point is that nobody has a monopoly on developing the code. Work for hire is welcome, and we're even starting to see it happen more often (and I can't wait for task markets to hit the big time).
There are wacky IA-32 opcodes like ENTER that affect several registers and memory locations. There are even circumstances (like CALL with a task gate) that affect all the registers and the page tables. The instruction set docs have huge pseudocode in places, and no reasonable pipeline can handle that without resorting to micro-ops.
The corporation holds copyright to my work, but both my manager and I have exactly as much authority as the chief officers have chosen to delegate to us--there is no legal requirement that they make every such decision personally.
There are no consequences to using GPL'd code in my work unless and until we decide to redistribute and license the work to third parties under incompatible terms.
We've all heard rumblings about age bias in programming--they'd often rather underpay an desperate, overcaffeinated twenty-year-old than have to listen to that old cuss who can explain why the project is doomed. As a colleague (hi Dan!) put it, the two career paths for a programmer are
You can stay focused on coding if you're the local guru who knows the systems inside and out (in fact they might be afraid to try to go on without you), but there's always the risk your expertise will suddenly be obsolete, and many companies don't even realize how much they needed one until they get rid of them.
At 31, I'm slowly picking up project management (mostly by choosing jobs likely to let me do some, and reading), because it's easier to sell (nobody sane thinks they don't need it) and age mostly improves credibility. And somebody's got to oppose the industry's gratuitous complexity (or "cover fire", as Joel Spolsky puts it) in favor of the simplest thing that works.