What can I say? Three and a half years of indentured servitude got me my green card.
None of the code I wrote was part of released code (so I felt a bit better about it's proprietary nature): I wrote test automation code and server-side code for mobile services.
The physical conditions were cushy (private office, etc.), but the mental anguish was pure horror: "Ohh! Your code has no bugs? Great fix these other people's bugs -- they can't find them... Oh dear, you had lots of bugs to fix last year, tsk, tsk: bad review for you."
I suppose some people thrive in an environment that rewards the political savvy to get other people to clean up their mess, but I don't.
There actually are a few good people there, doing decent research, but, from what I saw, very little trickled down to improve day to day development, or worse, it was misinterpreted and misapplied.
Of course, that's just my experience. No doubt some people like it there -- I just attribute my experience to a bad case of culture clash (That, and the "linux fish" on my car's bumper.)
I guess I'm stupid.
Is there any reason not to have a nice neat hierarchy on the server, that is mirrored with a collapsed symlink farm, with the symlinks exposed in the web pages?
Yes, this means one has to map the long names to the short names when generating pages, but that can be an authoring-time issue or dynamic page generation issue. Heck, output-rewriting of the page can do this.
It has to be combined with some type of fuel to explode.
Surprisingly, no. It will detonate all by itself with a big enough shock. This was discovered when a large amount (about a ton) got wet, and recrystalized into a large mass, and someone got the "safe" idea of just blasting it apart with dynamite. It was always "safe" before. There have been a number of ammonium nitrate disasters.
One could be reasonably safe by simply transporting the oxidizer and the fuel separately.
True. Or even mixed, though this is obviously less safe. Ammonium nitrate is such a useful explosive precisely because it is so hard to set it off. But, with an appropriate blasting cap, and sensitizing with fuel (6% fuel oil by mass will do it, IIRC), it can be done. It will also undergo a DDT (deflagration to detonation transition) in a fire in an enclosed space.
Actually, the biggest risk is generally ensuring that one does set the charge off, and not merely disperse it over a large area.
Agreed.
So why the hell not form a posse to go down and "discuss" the misconduct with the school staff involved?
I'm sure a few tens of thousands of us holding a "seminar" about "when, if at all" to use deadly force against such vile spawn, might send the message home that such actions might get the individuals involved killed.
The school described the strip-search as 'not excessively intrusive in light of suspected infraction.'
Where the fuck does "suspected" equal the issuing of a warrant upon probable cause?
O.K., so New Jersey v. T. L. O., 469 U.S. 325 (1985), held that searches in public schools do not require warrants but only reasonable grounds for believing that the search will result in the finding of evidence of illegal activity.
What the hell were the reasonable grounds, then?
A zero-tolerance policy does not constitute "reasonable grounds" to me. Heck, this was an honors student, with presumably a clean reputation.
If some perverted teacher did that to my daughter, his or her head might quite possibly be meeting the business end of an ax, to ensure that my daughter would, in the future, be secure in her person and never risk violation of her rights by that individual again.
Most traffic offenses are civil violations, and, as such, a jury trial is not an option.
She can either pay the fine, or dispute that she committed the infraction. In some jurisdictions, she can admit that she committed the infraction, but that there were mitigating circumstances and seek the mercy of the judge. (Basically, if you refuse to pay the fine, the city automatically sues you.)
For a jury trial she has to be charged with a crime.
I suppose she could ignore the fine. In many places failing to pay or respond can result in a revocation of one's driving license, and so, she could find herself charged with driving without a license, and so could eventually escalate the issue until she is charged with a crime, and get the jury trial she wants.
The problem here is that if she did that, the question of why she did not challenge the ticket sooner would be raised.
What she should do is claim mitigating circumstances, and keep appealing until she gets the fine dismissed. However, note that there are court fees involved, and the cost to file a first level appeal could be $250 or more -- probably more than the amount of the fine. Appealing would therefore be a question of principle, and not money.
Now, the interesting thing is this: if she kept driving, and so risked her son's death, simply to comply with a parking ordinance, could she be realistically charged with having depraved indifference to human life? Generally, the law can't be contradictory: complying with one requiring the breaking of another. The scope of the law generally establishes which one has precedence (state law overriding city ordinance, for example.) When both are equal in scope, neither is applicable if they are contradictory.
Many of us decry the DMCA as making the exercise of traditional fair use rights illegal, but in the same manner that RMS turned copyright law on it's ear with the GPL "copyleft", I think we can leverage the DMCA's "anti-circumvention" provisions to our advantage:
If a piece of code is tagged as GPL, and your linker or loader lets you combine it with non-free code, without identifying that you have done this, so a recipient of the combination can't (a) know this was done, (b) have the means to identify you as the infringer and (c) the means to identify the license holders who's rights have been infringed so they can be contacted; then is not that linker or loader not a tool to enable infringement under the DMCA?
This is the digital equivalent of engraving "licenced from" (as opposed to "property of") on a piece of code, and having tools to remove it. Could not the DMCA be used to pressure OS providers to ensure their loaders and linkers respect the metadata in the code, and report possible infractions to the user?
Of course, there will always be tools that infringe. Stopping everything is not possible. But, if a tool becomes so widespread to be well-known, it can generally be tracked to the provider. Certainly, pressure to have linkers and loaders honor such metadata could be levied against providers of popular operating systems.
1. You want to dictate terms to which you want me to agree without making it possible for me to easily know whether those terms are ones to which I want to agree.
You think reading source to understand something is inconvenient for many?
How about agreeing to a potentially legally-binding license that you don't have the skill to review (because you're not a lawyer) or money to review (because you can't afford a lawyer to do it for you)?
If something says, "this is GPL," or "this is LGPL," or this is "BSD," and I understand those licenses, I can confidently accept the terms.
If something says, "this is proprietary blah blah blah," then I don't know what the terms are.
RMS might chose to reject non-free code on principle. Others might be happy to accept it. But, if you want me to run your code, that you exert control over, I want to be able to know (a) what your terms are, (b) in a manner I can understand, so (c) I can make an informed choice, and (d) preferably in a well-defined format so I can empower my browset to make the choice in my behalf.
2. Do I trust this code?
Code signing, and trusted code is something that can be handled by existing PKI mechanisms.
3. Can I substitute my own code for a portion of yours? The service license might forbid this, or it might not. The bigger problems are (a) this can't be conveniently done now, even if replacement code could be written, (b) there is no method of defining the interfaces separate from the implementation so I can determine what substitude code is acceptable.
4. If you claim to honor a free software or other open source (I view free software licenses a subset of open source ones) license, can you prove it, by providing access to source that I can (a) obtain, (b) verify matches your code?
This is something I've been pondering for a while.
As free code is built, from object modules (including obfuscated source), to libraries (shared and static), to applications, a record of (a) how it is licensed and (b) where to get source, (c) how to build it should be included in it, so that (a) licence complience can be verified, and (b) others can exercise their rights to examine and modify with ease.
This is actually harder than a combination of signatures and URLs: how code got built depends on the compiler version, flags, configuration, operating system, and a host of other factors -- some of which matter (the version of compiler), and others might not (what sectors on the hard disk the compiler executible occupied).
I've worked in shops where we produce software that is an aggregation of free and non-free components. Actually respecting the free software licenses and ensuring that appropriate source is available is a hard task. Furthermore, just providing source might not make it easy for you to replicate our build environment. It would be nice if those tools that are used to build software could (b)automatically include source distribution points and build metadata where appropriate, that (b) binaries derived from free code could be validated against the availability of said source, and (c) combinations of free and non-free code could not run except in the environments where they were built.
Granted, it is likely impossible to enforce all the above requirements (if combinations of free and non-free code are encrypted with a secret that the build machine knows, and is required at run-time, that secret could be distributed with the code to avoid the check). But, free software tools like compilers, linkers, and loaders should have capabilities to include the necessary metadata so that (a) those who wish to comply with the requirements have an easier time of ensuring that they do, (b) those that seek to subvert them have a harder time, and (c) those that want to verify complience, can.
Guess what, buckeye, you're running a marathon. 26 miles. 40k.
The thing is, a break to take a shower at that time, would KILL my productivity.
Finished the first day, struggled through the next eight to ten hours, we're about at 3:00 AM. Fatigue sets in, but that goes away by 5:00AM or so. By seven or eight, one is comfortably "in the zone", and coding against time. One is oblivious to time, actually, and sees the design clearly in one's mind. If anything, one gets frustrated that one isn't typing as fast as one wants, lest the design start to fade. (You've got not time to document anything -- this is crisis mode -- where you have to replace a horribly defective piece or implement a missing piece NOW.) The code IS the documentation, and you can extract the spec later. After all, you are doing the plumbing now and have the spec in your head. No need to persist it in anything except code just yet.
Have you ever been "in the zone"? I suspect not. You don't program, actually. Code just flows. It's almost a transcendental, hyptnotic state of extreme productivity. You can go on for another day like this, at least...
... as long as you aren't interrupted, espescially for something as mundane as a shower.
Thing is, if you fall out of "the zone" on a marathon like this, you likely will not get back.
The other thing is this: this is extremely taxing on the body and mind, and you will likely have to crash for 3-4 days afterward and not be able to repeat the performance for a few months.
Is the resulting code elegant? Likely not. It represents the extreme compromise of schedule, performance, and correctness within extremely well-defined bounds. Probably does not scale, or is not extensible. Not because it is "bad", but because that flexibility was compromised to meet schedule.
Would it have been better to do it slowly, and better? Of course, except you're stuck in this hell because (a) some other moron coultn't code correctly even when given ample time, (b) something was forgotten, (c) a late requirement came in. It could be argued that what you produced, in terms of maintainability and extensability, is no better than what said moron would have. But, you've ensured it is (a) correct, (b) on time.
It's time to stop, reflect, and consider changing jobs if management requires these kind of heroics too often. (IOW, having a fire extinguisher in the explosives factory is a good idea, but not a good place to work if you need to use it often because some moron insists on smoking there.)
Oh, I didn't say one didn't take breaks. One just doesn't take breaks long enough to sleep or go home.
Can't do it often, and the second twelve hours are often the worst, but a "second wind" effect kicks in. And, by this time, you usually know everything that is wrong, and it's crystal clear what it will take to make it right -- you just have to code it.
It's often not hard work, just lots of loose ends to tie up, or reimplement a badly written library, that you could have implemented in a week of wall time, but were too busy, so it was outsourced to someone who did it in a month, and all wrong.
You "push" because so much context is in your head, that you'd forget it before you could ever document it, and as long as you keep coding, it stays refreshed.
There's a saying that "the product of the size of the code base and the duration where it will all be crystal clear is a constant". You want that two or three golden days to be the duration when all is crystal clear. It's like being in "the zone" on steroids (or, Jolt, as the case may be).
Often, at the end, it's far from perfect, but it's good enough to meet this version's reauirements, and you can at least document what remains broken/cruftily implemented.
Except, you don't have the luxury of taking the time.
You have to fix what someone else fskd up.
It often isn't that much of a matter of re-design, so the lack of "sharpness" after the first 12 hours or so doesn't hurt as much as you'd think. And, the way things are broken is often a clue of how to repair them. It's a question of realizing that it's about two to three days of typing to be done in two to three days of wall time.
No, you can't debug effectively, except the simplest things in context, but you sure can code the simple missing bits until you drop.
Exactly. I'm always amazed by people who think that writing impenetrable code is "advanced". Any jackass can write something convoluted and obscure that nobody else can understand (or maintain) -- what takes actual talent is condensing complicated logic into code that's simple enough a ten year old would understand it.
Hmm, so when an efficient sort routine is required, I should code a bubble sort just so that a novice can understand it, instead of a shell sort that is more efficient (and, arguably, state of the art for modest sorting requirements)? [N.B. I'm using "sort" as an example: of course, it's likely that one has a black box routine availale to do this.]
Often, the problem is that the hoi poloi aren't even up to basic standards of technical competence. They should be left to areas where they can do little harm, with more advanced, and critical, work assigned to those who are skilled at it, and need only be able to explain it's workings to their peers as far as level of competence goes. Document the interface for everyone.
Do we try to make brain surgery so easy that a 10-year-old could do it?
No.
If I need to find an optimal cost path through a graph (say for a "nearest word" match), and have the memory for it, the solution screams "Viterbi Search!". It's common in the literature, but a bit non-intuative to a novice. Should I not use it? I would think that "adequate documentation" would be a cite to any one of the references on the subject.
It used to be that one could assume one's coworkers were at least familiar with the material in Donald Knuth's "The Art of Computer Programming" (which covers Viterbi Search in the first volume quite well). Sadly, that is no longer the case.
Really is it that hard to spend 10 minutes in the morning, EVERY MORNING to bathe yourself??
Sometimes, when you've spent the past 48-72 hours working to solve some crisis that some moron left and you have to clean up, yes.
You look like shit, smell about as bad, and have a cranky attitude to match.
But, shipping on time and avoiding a $250,000/day contract penalty can sometimes justify the hell (Ah, contracts with separate "code complete" and "QA Pass" dates.)
Some coders don't shower all the time because they haven't gone home.
Copyright is a federal institution in the United States, and states cannot nullify, adjust, reinterpret, or ignore it. Copyright is held by the author of the work unless they created it in the course of their official duties for a regular employer, per USC 17. Therefore, New York couldn't do this.
Can you say, "We subsidized the work, so it was a 'Work For Hire'?" The state owns it.
I've never heard of eminent domain being used with intellectual property. How would it work? They aren't actually causing you any loss, so can they just take it for free? Can they say that they get an exclusive license, or only that they get a plain old ordinary license? Can they force you to give support? Most importantly, would this run into federal supremacy problems? Congress created copyright, so it could impose whatever limitations it liked, but states might be on sketchy ground here.
There's a first time for everything.
I think eminent domain here would be in the form of state control of use of the intellectual property, first that the state does not have to respect the GPL for "public purposes", and later, to order support to repair "defects" caused by the writer, for example in the same way that a city can order a landownder who restricted a drainage swail on their property, lest the owner be billed to correct the problem they caused.
About the only defense would be the unconstitutionality of slavery, but the counter argument could take the form of either restriction of use by the writer, or the argument that they weren't forced to write it, but having done so, it is a public menace.
And, even if copyright is under Federal jurisdiction, the state can arguably position itself as copyright holder, or at least one of several copyright holders. Furthermore, states do have powers of eminent domain within their borders (which would likely apply to use restrictions there): NY funded it, so in NY, you have to pay a use tax to use it, for example.
I have learned to not question the creativity that the state can use to take and tax.
Taxing the internet would be collecting all the existing taxes on corporate profits (rather like the telcos get taxed), as well as a tax on the traffic volume, and/or end-point identification (Domain names or static IP address blocks), etc. Yes, domain names and static IP addresses cost money now, but they are not a direct source of revenue for the government. IOW, a new tax.
Another example: oil refineries are taxed on profits and gasoline is taxed at the pump separately.
Yet another example: the "Universal Access Service Fee" on telephone endpoints.
Except, (a) Canada does not have as strong a constitution, (b) jurisdiction over tax matters is via a special tax court, and (c) the burder of proof based on a balance or probabilities is upon the taxpayer to establish their innocence. The burden to establish civil penalties is upon the state.
Basically, if the taxing body says you're guilty of tax evasion, you have to prove otherwise. And, generally your assets can be siezed, until you can prove your innocence, often depriving you of the means to afford adequate legal representation.
Contrast areas of criminal jurisdiction not related to state revenue: the burden of proof likes on the prosecution.
Then, if those Americans don't like it, they should leave, watch that the door doesn't whack them on the ass when they do, and if they decide to return, stand in line behind all those that actually want to be here.
Some of us actually like it here, because, as imperfect as it is, we find it far better than the alternative that we left. I usually hear the most griping from those that are fat, lazy, spoiled, and think the world owes them a life.
As for me, I've been treated far better as a foreigner coming to the U.S. legally through "the front door" than a citizen when I lived in Canada. I was shat upon for being able to (barely) support a family on my income alone: "How DARE you deny child care workers a job by not puting them in daycare?" I was berated for hiring a kid to mow my lawn in the summer of 2003: "Who the hell are you to be able to afford that?" My kid was punished in her Ontario school for "showing off" for show and tell by displaying her yearbook from her last year in Texas: it was "too ostentatious because it was hardcover".
I make no bones about hiding my utter disgust for "liberals". They are like fleas upon the societal dog who multiply and suck the breath and blood out of the productive. We should not suffer them just because they sometimes take down a corrupt and greedy individual who leverages money to buy law for self-serving purposes.
If you don't have "connections" you wait forever, often longer than if you could have paid the tax dollars you did toward health care toward a private policy. I discovered this very quickly when I moved from Canada to the U.S.: my taxes went way down (as I was married at the time and owned a home, there being no mortgage interest deduction in Canada, or the option to file jointly (and the tax credit for a spouse was a token amount)), and the premium I paid for health insurance was far less than the tax dollars I save, and the care I received was far, far, better.
Granted, there are many who lack health insurance in the U.S., the prospect of malpractice litigation, insurance, and plethora of various insurers, make for inefficiencies and insurance more expensive than it needs to be.
But, I've found if you're willing to work, you get some level of health insurance at least on a par with what you'd get in a socialized nation: the worst insured care in the U.S. may suck compared to the best, but it is often better than what is available to all in a socialized medicine nation.
I've lived 36 years in Canada, seen the nationalization of health care become complete there in the course of my life, amd 11 in the U.S. I write from expereience. There is a thriving healthcare industry just south of the Canada/U.S. border that caters to Canadians who can cross the border.
Canada is like North Korea and Cuba as far as it's national health care system goes. It isn't a matter of paying taxes to provide basic care for everyone, with the option to purchase better care for yourself once you've paid your "tax share". This is illegal, arguably because it is "unfair" for "the rich" to have better health care than "the rest". They can buy better cars, though. So, if they chose to be materially frugal, and invest in gold-plated health care instead, why question it?
Of course, this results in long waits in the public health care system, where doctors are paid a mediocre wage, can't command higher salaries on the basis of competentence, and have procedural quotas imposed.
Private care is an option but only if (a) the provider does not participate in the public system, (b) the patient is not eligible for the public system (IOW is not a citizen). Thus, only providers that cater to foreign athletes operate privately.
I am a lawful permanent resident of the U.S. since 2006, and can apply to become a citizen in 2011. I had previously resided in the U.S. from 1997 to 2003 on various visas, returned to Canada for a year in 2003/4, and came back to the U.S. on work visas in 2004, immigrating in 2006. I have an American son. When we were in Canada in 2003 and he was sick, I had the choice of applying for Canadian citizenship for him du sanguis, and he could wait and get "free" health care. Insted, having produced his American passport, I paid and he went to the head of the queue.
There has been pressure for a two-tier system, as in other socialized nations like Great Britian, but one hears tales of woe from their citizenry as well.
In the face of mounting healthcare costs, what the U.S. really needs is a system of cooperatively-ownend health insurers, rather like cooperatively-owned life insurers, and standardization of policies, and practices, in the healthcare field.
Geez. The government won't sue for use. They'll tax you on your own code, and prosecute for criminal income tax evasion if you don't pay.
Dunno about the U.S., but in Canada, while the police can't bust down your door just because they think you killed someone, they can certainly do so, and sieze all your assets and property, if they think you're guilty of tax evasion. I hear the IRS is about the same, though my dealings with them have been reasonable.
This is effectively an open source government subsidy, albeit a small one.
Government subsidies generally are used to either (a) encourage a behavior (mortgate interest deducture designed to encourage home ownership, for example, though what it really does is raise home prices), or (b) create a claim to that subsidiesd, in hopes of exploiting it later, which is likely the case here.
IOW, "You could not have developed this code if the taxpayer did not subsidize it, therefore the taxpayer owns it, not you, and you now have to pay a $50/year tax to use it. Obviously, since you did not own it, you could not copyright it, and the GPL is null and void, except where we say otherwise. If you need further convincing, we will just apply the doctrine of eminent domain to own it."
Basically, the government creates the slimest of justifications for an ownership interest in something, on terms likely never acceptable in a free market, and the uses it's force to exploit that supposed ownership interest.
I'd expect that RMS might actually like this, but I also think he is a bit naive about how evil and incompetent governments can be.
He supports universal healthcare, for example, but can not accept that it's general failure is due to a design and not implementation flaw.
Slashdot Mirror
Machines should use the short ones, and people who care about the structure, the long ones, and there should be some kind of mapping.
None of the code I wrote was part of released code (so I felt a bit better about it's proprietary nature): I wrote test automation code and server-side code for mobile services.
The physical conditions were cushy (private office, etc.), but the mental anguish was pure horror: "Ohh! Your code has no bugs? Great fix these other people's bugs -- they can't find them... Oh dear, you had lots of bugs to fix last year, tsk, tsk: bad review for you."
I suppose some people thrive in an environment that rewards the political savvy to get other people to clean up their mess, but I don't.
There actually are a few good people there, doing decent research, but, from what I saw, very little trickled down to improve day to day development, or worse, it was misinterpreted and misapplied.
Of course, that's just my experience. No doubt some people like it there -- I just attribute my experience to a bad case of culture clash (That, and the "linux fish" on my car's bumper.)
I guess I'm stupid. Is there any reason not to have a nice neat hierarchy on the server, that is mirrored with a collapsed symlink farm, with the symlinks exposed in the web pages? Yes, this means one has to map the long names to the short names when generating pages, but that can be an authoring-time issue or dynamic page generation issue. Heck, output-rewriting of the page can do this.
Er, Quebec French and the French spoken in France are separated by about 350 years of linguistic evolution.
Yes, that is a better cite. Thanks.
Yes.
It has to be combined with some type of fuel to explode.
Surprisingly, no. It will detonate all by itself with a big enough shock. This was discovered when a large amount (about a ton) got wet, and recrystalized into a large mass, and someone got the "safe" idea of just blasting it apart with dynamite. It was always "safe" before. There have been a number of ammonium nitrate disasters.
One could be reasonably safe by simply transporting the oxidizer and the fuel separately.
True. Or even mixed, though this is obviously less safe. Ammonium nitrate is such a useful explosive precisely because it is so hard to set it off. But, with an appropriate blasting cap, and sensitizing with fuel (6% fuel oil by mass will do it, IIRC), it can be done. It will also undergo a DDT (deflagration to detonation transition) in a fire in an enclosed space.
Actually, the biggest risk is generally ensuring that one does set the charge off, and not merely disperse it over a large area.
Thank you for reminding me why I left that hell hole.
Agreed. So why the hell not form a posse to go down and "discuss" the misconduct with the school staff involved? I'm sure a few tens of thousands of us holding a "seminar" about "when, if at all" to use deadly force against such vile spawn, might send the message home that such actions might get the individuals involved killed.
Where the fuck does "suspected" equal the issuing of a warrant upon probable cause?
O.K., so New Jersey v. T. L. O., 469 U.S. 325 (1985), held that searches in public schools do not require warrants but only reasonable grounds for believing that the search will result in the finding of evidence of illegal activity.
What the hell were the reasonable grounds, then?
A zero-tolerance policy does not constitute "reasonable grounds" to me. Heck, this was an honors student, with presumably a clean reputation.
If some perverted teacher did that to my daughter, his or her head might quite possibly be meeting the business end of an ax, to ensure that my daughter would, in the future, be secure in her person and never risk violation of her rights by that individual again.
Most traffic offenses are civil violations, and, as such, a jury trial is not an option. She can either pay the fine, or dispute that she committed the infraction. In some jurisdictions, she can admit that she committed the infraction, but that there were mitigating circumstances and seek the mercy of the judge. (Basically, if you refuse to pay the fine, the city automatically sues you.) For a jury trial she has to be charged with a crime. I suppose she could ignore the fine. In many places failing to pay or respond can result in a revocation of one's driving license, and so, she could find herself charged with driving without a license, and so could eventually escalate the issue until she is charged with a crime, and get the jury trial she wants. The problem here is that if she did that, the question of why she did not challenge the ticket sooner would be raised. What she should do is claim mitigating circumstances, and keep appealing until she gets the fine dismissed. However, note that there are court fees involved, and the cost to file a first level appeal could be $250 or more -- probably more than the amount of the fine. Appealing would therefore be a question of principle, and not money. Now, the interesting thing is this: if she kept driving, and so risked her son's death, simply to comply with a parking ordinance, could she be realistically charged with having depraved indifference to human life? Generally, the law can't be contradictory: complying with one requiring the breaking of another. The scope of the law generally establishes which one has precedence (state law overriding city ordinance, for example.) When both are equal in scope, neither is applicable if they are contradictory.
Many of us decry the DMCA as making the exercise of traditional fair use rights illegal, but in the same manner that RMS turned copyright law on it's ear with the GPL "copyleft", I think we can leverage the DMCA's "anti-circumvention" provisions to our advantage:
If a piece of code is tagged as GPL, and your linker or loader lets you combine it with non-free code, without identifying that you have done this, so a recipient of the combination can't (a) know this was done, (b) have the means to identify you as the infringer and (c) the means to identify the license holders who's rights have been infringed so they can be contacted; then is not that linker or loader not a tool to enable infringement under the DMCA?
This is the digital equivalent of engraving "licenced from" (as opposed to "property of") on a piece of code, and having tools to remove it. Could not the DMCA be used to pressure OS providers to ensure their loaders and linkers respect the metadata in the code, and report possible infractions to the user?
Of course, there will always be tools that infringe. Stopping everything is not possible. But, if a tool becomes so widespread to be well-known, it can generally be tracked to the provider. Certainly, pressure to have linkers and loaders honor such metadata could be levied against providers of popular operating systems.
And, I can chose to accept them or not.
But, the bigger issues here are:
1. You want to dictate terms to which you want me to agree without making it possible for me to easily know whether those terms are ones to which I want to agree.
You think reading source to understand something is inconvenient for many?
How about agreeing to a potentially legally-binding license that you don't have the skill to review (because you're not a lawyer) or money to review (because you can't afford a lawyer to do it for you)?
If something says, "this is GPL," or "this is LGPL," or this is "BSD," and I understand those licenses, I can confidently accept the terms.
If something says, "this is proprietary blah blah blah," then I don't know what the terms are.
RMS might chose to reject non-free code on principle. Others might be happy to accept it. But, if you want me to run your code, that you exert control over, I want to be able to know (a) what your terms are, (b) in a manner I can understand, so (c) I can make an informed choice, and (d) preferably in a well-defined format so I can empower my browset to make the choice in my behalf.
2. Do I trust this code?
Code signing, and trusted code is something that can be handled by existing PKI mechanisms.
3. Can I substitute my own code for a portion of yours? The service license might forbid this, or it might not. The bigger problems are (a) this can't be conveniently done now, even if replacement code could be written, (b) there is no method of defining the interfaces separate from the implementation so I can determine what substitude code is acceptable.
4. If you claim to honor a free software or other open source (I view free software licenses a subset of open source ones) license, can you prove it, by providing access to source that I can (a) obtain, (b) verify matches your code?
This is something I've been pondering for a while.
As free code is built, from object modules (including obfuscated source), to libraries (shared and static), to applications, a record of (a) how it is licensed and (b) where to get source, (c) how to build it should be included in it, so that (a) licence complience can be verified, and (b) others can exercise their rights to examine and modify with ease.
This is actually harder than a combination of signatures and URLs: how code got built depends on the compiler version, flags, configuration, operating system, and a host of other factors -- some of which matter (the version of compiler), and others might not (what sectors on the hard disk the compiler executible occupied).
I've worked in shops where we produce software that is an aggregation of free and non-free components. Actually respecting the free software licenses and ensuring that appropriate source is available is a hard task. Furthermore, just providing source might not make it easy for you to replicate our build environment. It would be nice if those tools that are used to build software could (b)automatically include source distribution points and build metadata where appropriate, that (b) binaries derived from free code could be validated against the availability of said source, and (c) combinations of free and non-free code could not run except in the environments where they were built.
Granted, it is likely impossible to enforce all the above requirements (if combinations of free and non-free code are encrypted with a secret that the build machine knows, and is required at run-time, that secret could be distributed with the code to avoid the check). But, free software tools like compilers, linkers, and loaders should have capabilities to include the necessary metadata so that (a) those who wish to comply with the requirements have an easier time of ensuring that they do, (b) those that seek to subvert them have a harder time, and (c) those that want to verify complience, can.
Yeah, after a 10k run, I'd want to shower too.
Guess what, buckeye, you're running a marathon. 26 miles. 40k.
The thing is, a break to take a shower at that time, would KILL my productivity.
Finished the first day, struggled through the next eight to ten hours, we're about at 3:00 AM. Fatigue sets in, but that goes away by 5:00AM or so. By seven or eight, one is comfortably "in the zone", and coding against time. One is oblivious to time, actually, and sees the design clearly in one's mind. If anything, one gets frustrated that one isn't typing as fast as one wants, lest the design start to fade. (You've got not time to document anything -- this is crisis mode -- where you have to replace a horribly defective piece or implement a missing piece NOW.) The code IS the documentation, and you can extract the spec later. After all, you are doing the plumbing now and have the spec in your head. No need to persist it in anything except code just yet.
Have you ever been "in the zone"? I suspect not. You don't program, actually. Code just flows. It's almost a transcendental, hyptnotic state of extreme productivity. You can go on for another day like this, at least...
Thing is, if you fall out of "the zone" on a marathon like this, you likely will not get back.
The other thing is this: this is extremely taxing on the body and mind, and you will likely have to crash for 3-4 days afterward and not be able to repeat the performance for a few months.
Is the resulting code elegant? Likely not. It represents the extreme compromise of schedule, performance, and correctness within extremely well-defined bounds. Probably does not scale, or is not extensible. Not because it is "bad", but because that flexibility was compromised to meet schedule.
Would it have been better to do it slowly, and better? Of course, except you're stuck in this hell because (a) some other moron coultn't code correctly even when given ample time, (b) something was forgotten, (c) a late requirement came in. It could be argued that what you produced, in terms of maintainability and extensability, is no better than what said moron would have. But, you've ensured it is (a) correct, (b) on time.
It's time to stop, reflect, and consider changing jobs if management requires these kind of heroics too often. (IOW, having a fire extinguisher in the explosives factory is a good idea, but not a good place to work if you need to use it often because some moron insists on smoking there.)
Can't do it often, and the second twelve hours are often the worst, but a "second wind" effect kicks in. And, by this time, you usually know everything that is wrong, and it's crystal clear what it will take to make it right -- you just have to code it.
It's often not hard work, just lots of loose ends to tie up, or reimplement a badly written library, that you could have implemented in a week of wall time, but were too busy, so it was outsourced to someone who did it in a month, and all wrong.
You "push" because so much context is in your head, that you'd forget it before you could ever document it, and as long as you keep coding, it stays refreshed.
There's a saying that "the product of the size of the code base and the duration where it will all be crystal clear is a constant". You want that two or three golden days to be the duration when all is crystal clear. It's like being in "the zone" on steroids (or, Jolt, as the case may be).
Often, at the end, it's far from perfect, but it's good enough to meet this version's reauirements, and you can at least document what remains broken/cruftily implemented.
You have to fix what someone else fskd up.
It often isn't that much of a matter of re-design, so the lack of "sharpness" after the first 12 hours or so doesn't hurt as much as you'd think. And, the way things are broken is often a clue of how to repair them. It's a question of realizing that it's about two to three days of typing to be done in two to three days of wall time.
No, you can't debug effectively, except the simplest things in context, but you sure can code the simple missing bits until you drop.
Hmm, so when an efficient sort routine is required, I should code a bubble sort just so that a novice can understand it, instead of a shell sort that is more efficient (and, arguably, state of the art for modest sorting requirements)? [N.B. I'm using "sort" as an example: of course, it's likely that one has a black box routine availale to do this.]
Often, the problem is that the hoi poloi aren't even up to basic standards of technical competence. They should be left to areas where they can do little harm, with more advanced, and critical, work assigned to those who are skilled at it, and need only be able to explain it's workings to their peers as far as level of competence goes. Document the interface for everyone.
Do we try to make brain surgery so easy that a 10-year-old could do it?
No.
If I need to find an optimal cost path through a graph (say for a "nearest word" match), and have the memory for it, the solution screams "Viterbi Search!". It's common in the literature, but a bit non-intuative to a novice. Should I not use it? I would think that "adequate documentation" would be a cite to any one of the references on the subject.
It used to be that one could assume one's coworkers were at least familiar with the material in Donald Knuth's "The Art of Computer Programming" (which covers Viterbi Search in the first volume quite well). Sadly, that is no longer the case.
Sometimes, when you've spent the past 48-72 hours working to solve some crisis that some moron left and you have to clean up, yes.
You look like shit, smell about as bad, and have a cranky attitude to match.
But, shipping on time and avoiding a $250,000/day contract penalty can sometimes justify the hell (Ah, contracts with separate "code complete" and "QA Pass" dates.)
Some coders don't shower all the time because they haven't gone home.
Can you say, "We subsidized the work, so it was a 'Work For Hire'?" The state owns it.
I've never heard of eminent domain being used with intellectual property. How would it work? They aren't actually causing you any loss, so can they just take it for free? Can they say that they get an exclusive license, or only that they get a plain old ordinary license? Can they force you to give support? Most importantly, would this run into federal supremacy problems? Congress created copyright, so it could impose whatever limitations it liked, but states might be on sketchy ground here.
There's a first time for everything.
I think eminent domain here would be in the form of state control of use of the intellectual property, first that the state does not have to respect the GPL for "public purposes", and later, to order support to repair "defects" caused by the writer, for example in the same way that a city can order a landownder who restricted a drainage swail on their property, lest the owner be billed to correct the problem they caused.
About the only defense would be the unconstitutionality of slavery, but the counter argument could take the form of either restriction of use by the writer, or the argument that they weren't forced to write it, but having done so, it is a public menace.
And, even if copyright is under Federal jurisdiction, the state can arguably position itself as copyright holder, or at least one of several copyright holders. Furthermore, states do have powers of eminent domain within their borders (which would likely apply to use restrictions there): NY funded it, so in NY, you have to pay a use tax to use it, for example.
I have learned to not question the creativity that the state can use to take and tax.
Taxing the internet would be collecting all the existing taxes on corporate profits (rather like the telcos get taxed), as well as a tax on the traffic volume, and/or end-point identification (Domain names or static IP address blocks), etc. Yes, domain names and static IP addresses cost money now, but they are not a direct source of revenue for the government. IOW, a new tax.
Another example: oil refineries are taxed on profits and gasoline is taxed at the pump separately.
Yet another example: the "Universal Access Service Fee" on telephone endpoints.
Except, (a) Canada does not have as strong a constitution, (b) jurisdiction over tax matters is via a special tax court, and (c) the burder of proof based on a balance or probabilities is upon the taxpayer to establish their innocence. The burden to establish civil penalties is upon the state.
Basically, if the taxing body says you're guilty of tax evasion, you have to prove otherwise. And, generally your assets can be siezed, until you can prove your innocence, often depriving you of the means to afford adequate legal representation.
Contrast areas of criminal jurisdiction not related to state revenue: the burden of proof likes on the prosecution.
Then, if those Americans don't like it, they should leave, watch that the door doesn't whack them on the ass when they do, and if they decide to return, stand in line behind all those that actually want to be here.
Some of us actually like it here, because, as imperfect as it is, we find it far better than the alternative that we left. I usually hear the most griping from those that are fat, lazy, spoiled, and think the world owes them a life.
As for me, I've been treated far better as a foreigner coming to the U.S. legally through "the front door" than a citizen when I lived in Canada. I was shat upon for being able to (barely) support a family on my income alone: "How DARE you deny child care workers a job by not puting them in daycare?" I was berated for hiring a kid to mow my lawn in the summer of 2003: "Who the hell are you to be able to afford that?" My kid was punished in her Ontario school for "showing off" for show and tell by displaying her yearbook from her last year in Texas: it was "too ostentatious because it was hardcover".
I make no bones about hiding my utter disgust for "liberals". They are like fleas upon the societal dog who multiply and suck the breath and blood out of the productive. We should not suffer them just because they sometimes take down a corrupt and greedy individual who leverages money to buy law for self-serving purposes.
Rather like talk of taxing the internet because DARPA funded it's original development?
Get sick in Canada, and tell me what you think.
If you don't have "connections" you wait forever, often longer than if you could have paid the tax dollars you did toward health care toward a private policy. I discovered this very quickly when I moved from Canada to the U.S.: my taxes went way down (as I was married at the time and owned a home, there being no mortgage interest deduction in Canada, or the option to file jointly (and the tax credit for a spouse was a token amount)), and the premium I paid for health insurance was far less than the tax dollars I save, and the care I received was far, far, better.
Granted, there are many who lack health insurance in the U.S., the prospect of malpractice litigation, insurance, and plethora of various insurers, make for inefficiencies and insurance more expensive than it needs to be.
But, I've found if you're willing to work, you get some level of health insurance at least on a par with what you'd get in a socialized nation: the worst insured care in the U.S. may suck compared to the best, but it is often better than what is available to all in a socialized medicine nation.
I've lived 36 years in Canada, seen the nationalization of health care become complete there in the course of my life, amd 11 in the U.S. I write from expereience. There is a thriving healthcare industry just south of the Canada/U.S. border that caters to Canadians who can cross the border.
Canada is like North Korea and Cuba as far as it's national health care system goes. It isn't a matter of paying taxes to provide basic care for everyone, with the option to purchase better care for yourself once you've paid your "tax share". This is illegal, arguably because it is "unfair" for "the rich" to have better health care than "the rest". They can buy better cars, though. So, if they chose to be materially frugal, and invest in gold-plated health care instead, why question it?
Of course, this results in long waits in the public health care system, where doctors are paid a mediocre wage, can't command higher salaries on the basis of competentence, and have procedural quotas imposed.
Private care is an option but only if (a) the provider does not participate in the public system, (b) the patient is not eligible for the public system (IOW is not a citizen). Thus, only providers that cater to foreign athletes operate privately.
I am a lawful permanent resident of the U.S. since 2006, and can apply to become a citizen in 2011. I had previously resided in the U.S. from 1997 to 2003 on various visas, returned to Canada for a year in 2003/4, and came back to the U.S. on work visas in 2004, immigrating in 2006. I have an American son. When we were in Canada in 2003 and he was sick, I had the choice of applying for Canadian citizenship for him du sanguis, and he could wait and get "free" health care. Insted, having produced his American passport, I paid and he went to the head of the queue.
There has been pressure for a two-tier system, as in other socialized nations like Great Britian, but one hears tales of woe from their citizenry as well.
In the face of mounting healthcare costs, what the U.S. really needs is a system of cooperatively-ownend health insurers, rather like cooperatively-owned life insurers, and standardization of policies, and practices, in the healthcare field.
It DOES NOT need government interferance.
Geez. The government won't sue for use. They'll tax you on your own code, and prosecute for criminal income tax evasion if you don't pay.
Dunno about the U.S., but in Canada, while the police can't bust down your door just because they think you killed someone, they can certainly do so, and sieze all your assets and property, if they think you're guilty of tax evasion. I hear the IRS is about the same, though my dealings with them have been reasonable.
Great. Just great.
This is effectively an open source government subsidy, albeit a small one.
Government subsidies generally are used to either (a) encourage a behavior (mortgate interest deducture designed to encourage home ownership, for example, though what it really does is raise home prices), or (b) create a claim to that subsidiesd, in hopes of exploiting it later, which is likely the case here.
IOW, "You could not have developed this code if the taxpayer did not subsidize it, therefore the taxpayer owns it, not you, and you now have to pay a $50/year tax to use it. Obviously, since you did not own it, you could not copyright it, and the GPL is null and void, except where we say otherwise. If you need further convincing, we will just apply the doctrine of eminent domain to own it."
Basically, the government creates the slimest of justifications for an ownership interest in something, on terms likely never acceptable in a free market, and the uses it's force to exploit that supposed ownership interest.
I'd expect that RMS might actually like this, but I also think he is a bit naive about how evil and incompetent governments can be.
He supports universal healthcare, for example, but can not accept that it's general failure is due to a design and not implementation flaw.