I'm pretty far left of centre, but have always been deeply suspicious of "group rights."
The government of India's system of quotas for various "backward" castes, the dysfunctional kludges that are called the Belgian and Lebanese governments, and the Quebecker's continuing deranged attempt to legislate language, are prime examples of how disastrous these efforts can become. In the US, initiatives of this sort tend to be racially-based, but the bad idea has been tried all over the world for any number of reasons and seems to have failed everywhere. This is probably an indication that we need to achieve fairness by other means.
Tell me, would the language police shut down an English-language site that called for repeal of the province's law forcing French down its citizens' throats?
Sounds like you may have confused two stages in the "ecological succession" of the Web.
The growth of the Internet (which certainly did cause some change, though far less than electricity, the telegraph or the telephone) happened due largely to the efforts of people in the public sector and universities. This was the Pioneer phase.
The dotcomistas represent the second ("Land Grab") phase. Their goals in life were to get a free ride off the pioneers, make a fast buck, and lock everything down by litigation before competitors showed up to spoil the party. They didn't sow, and it's a pity any of them ever reaped.
So excuse me for not mourning the failure of these sleazebags' ambitions of being the next M$oft. One M$ is one too many. Now these dot-CEOs can fall back to their natural talents at selling timeshare condos and devising multi-level marketing schemes.
And forgive me for lacking sympathy for the suckers (both employees and investors) who believed the get-rich-quick pitches.
How about using the appropriate paradigm for the application at hand (which is not always OO), the right paradigm for the data in the database (which may be relational, OO, etc.), and establishing a sensible protocol between the two?
Because the OO model and the relational model aren't really either/or alternatives.
The idea that there's one correct paradigm for the code, and another for the data, presupposes a model that doesn't view data and behavior as being intimately connected. That is, your initial assumption denies the validity of the OO approach.
Object models are at a higher level of abstraction. I think it makes better sense to consider use of an RDBMS as an implementation detail, rather than a methodology. I'm painfully familiar with DB-centric methodologies like Oracle CDM. They presuppose a particular solution, then force the designers and implementers to deal with the complexities of the chosen solution, rather than finding an optimal way to represent the problem space. OOD, on the other hand, can lead to a variety of implementations, some using an OO language, others something else.
Bottom line: my beef with RDBMS's is that the usual approach is to pick the tool first, then figure out the requirements after you've already put on the straitjacket. That's a recipe for failed implementations.
Sounds like you're a dedicated technical guy, but I don't think I'd enjoy having you as a manager. To expand a well-known acronym, There's More Than One Way To Do It. It's important to recognize that about people.
I'm pretty enthusiastic about software myself, and am constantly exploring. But, when hiring, I don't give a toss about how excited the applicant gets over language features, as long as they're good at their job. I don't expect them to share my enthusiasm (though it might be nice if I'm looking for someone to talk to over lunch). I've known some people who do fine work, but see computing as just a job, and don't lie awake at night obsessing about queuing algorithms. I find that hard to understand, but accept it. "There's nowt so queer as folk." Her being low-key and not "hairy and unpredictable" might tell you something about her cognitive style, but little about her competency. And, when forming teams, I like to have some predictable people along with us jackrabbits, just to make sure someone's there to sweep up after the party.
Ever noticed how most massive, high-profile public works projects are undertaken by corrupt regimes with inferiority complexes? Albert Speer's brainfarts, Chicago skyscrapers, the Aswan Dam, Brasilia, the Three Gorges Dam, Petronas. Costly attempts to deflect attention from the slimy (in some cases, murderous) activities that got the rulers where they are.
Similar to the old investor's rule of thumb: look at the corporate headquarters. If it's grandiose, better put your money somewhere else.
Oracle has recently been pissing off customers in a big way with a processor-power-based pricing model. More CPUs, more MHz, pay Oracle more $$$. This is a transparent attempt to capture Moore's Law in a revenue stream, and has caused everyone who can take the re-engineering hit to migrate towards DB2, which still has a per-node license that's about half as costly for big installations.
Then there's middleware. Oracle has been moving for some time to become an end-to-end solutions provider, and this means that they're trying to engulf app server and other middleware functionality with Oracle proprietary products that nobody but Oracle takes seriously. When telling Oracle engineers that we'll use well-proven non-Oracle middle-tier products (say, EJB), they start spreading FUD. "Performance! Stability!" We know enough to ignore them, but I'm sure there are plenty of firms out there who have to rely more heavily on vendor advice, and that's how Ellison gets to buy those big boats.
So, I'd say that they'd be a lot more like M$ if they could get away with it.
I worked developing software in an aerospace company. The DOD was one of our customers. The aerospace company retained copyright to the source we developed. The DOD got the rights to use the source internal to their operations, but not to resell it.
So, the correct answer must be that it depends on the terms of your contract. Standard procurement guidelines probably assign ownership to the government by default, but those terms are negotiable.
Incidentally, the same aerospace company once sold source rights for another system to a foreign military organization, with no restrictions on resale. The customer improved it, then sold the rights to a private company, and they're now competitors. I'd like to know what happened to the bozo who let that contract slip through.
Re:Why isn't XML-RPC considered bloat?
on
ESR On XML-RPC
·
· Score: 1
I agree. Part of the reason the web is so easy to deal with is the use of non-binary protocols. Think of http compared to something opaque and binary that might be theoretically more efficient.
I'm happy to trade some plentiful and growing bandwidth for the ability to debug by eyeball. If you have a real need to maximize bandwidth utilization, this probably isn't the solution. But most of the time when I see people optimizing in this way, they're focusing on something that doesn't really matter in the grand scheme of things. Coders' time is expensive, as is debug and test time. Hardware and bandwidth are relatively cheap.
I recently visited a site that dumped you to an error page if you were using anything but IE5 or higher on Windows. It's my employer's 401K provider, so they have a captive audience. The "market" will not sort this out, unless you personally are the buyer. The same goes for health insurance, public services, and other cases where the user of the service isn't the same as the person who pays the bill. Employers or the government will usually contract for for the cheapest service they can get away with. Quality of service for user is a secondary priority.
Remember the old phone company joke: "We don't care. We don't have to."
Wait a minute... government collecting revenues for businesses is "socialism"? Plutocracy or corporate welfare, maybe. But these are capitalist firms that are attempting to subvert the famous free market to get their snouts deep into a new subsidy trough. Oppressive, centralist, unfair and corrupt, yes. Socialist, no. Incidentally, IANAS myself. But I also have seen a number of instances when Schumpeter has been proven right. Firms will suck up to the government to duck competition whenever they can, and if that is not resisted, free markets disappear.
Incidentally, this should prove advantageous to the US if the Europeans go further down this road and we have the sense not to.
The Guinness family were pro-independence, but were and are Protestant. And they're big supporters of the Conservative Party in England. Highly unlikely that they've ever backed the IRA.
I do, however, prefer Beamish or Murphy's. Not quite so heavy.
Currently, in the US, if a company has a lawyer on the payroll, this is a more or less risk-free strategy for them, but it's risky for the defendant, since it costs you money.
Maybe the Brits have the right idea. In some civil suits (for example, libel), if the judge finds in favor of the defendant, the plaintiff can be ordered to pay the defendant's court costs. This is to deter exactly this kind of tactical lawsuit.
What you're saying sounds a bit like "We tried using perl and the project failed miserably."
But I do agree that most of the benefit from Extreme comes when you have experienced programmers who can work without a safety net.
Frequent unit testing should catch the coding errors, but the project scope will spiral out of control unless there's someone there who knows what's feasible and what isn't, and you need someone who can actually make the code work.
It's a good methodology in the right situations. Pity that it has become "flavor of the month" at some sites. But it sounds like your girlfriend's experience had more to do with it being a badly-staffed, dysfunctional company than with the methodology they adopted.
Software engineering is a good thing, and I've seen it really pay off. But certification is either unnecessary or counter-productive.
Certification works best in technically stagnant fields like architecture and civil engineering, and there, mainly to restrict the supply of architects and civil engineers. The pace of change in software (and, to a lesser extent, software engineering) is too rapid for that. In any case, certification is more often a barrier to entry to a profession than a means of consumer protection.
I've got the CS degree myself, and lots of pieces of paper saying I know how to do this and that. Their only real value is to soothe nervous clients. Most of what I've learned that's of value was learned outside the degree and certification programs, and could not be learned in such programs. Colleagues of mine with degrees in other fields, or no degrees at all, are just as good, or better, software engineers than I am.
Remember, NOBODY who founded the CS field had a CS degree. There weren't any then. That includes Turing, Von Neumann, Wiener, Hamming, Knuth. I could go on.
In fact, in the UK, a suspect's refusal to testify can be used as evidence of guilt.
Jack Straw is also attempting to abolish trial by jury for most criminal cases, on the grounds that juries sometimes refuse to convict suspects. And the British law allowing the police to require anyone involved in a criminal investigation to decrypt any information the police ask for (and this can be a "fishing expedition"-- the American concept of probable cause is much weaker in the UK; all they need is a magistrate to rubber-stamp it) is as near totalitarian as anything in the golden age of the USSR or Hoxha's Albania.
Further evidence that there is really no distinction between authoritarian center-left and authoritarian right-wing governments. The only difference is that between those who have power, and the rest of us. The British public is too cowed or apathetic to kick back, and the only gains in human rights in that country have come from the (not particularly democratic) EU, particularly the incorporation of the European Convention on Human Rights into British law.
A sad situation, and if we see signs of any of this happening in the US, we should resist it.
When I started managing a sizeable team for the first time, someone I trusted told me that, if I really wanted to be successful, I should start by getting rid of the indispensible people.
This seemed at the time to be evil, cynical advice. But it had some truth in it. Two absolutely key employees left a few months later (better offer, not anything to do with the project) and we did just fine without them. Indispensible = self-marketing plus refusal to share knowledge. Both these individual traits are destructive to teams.
The flip side is that, when that better offer comes, manage the succession as best you can, and walk out that door . Anyway, if you're really that good, you will already have been sharing knowledge with others, and encouraging their development. In that case, this is just another growth opportunity for you and them.
You owe the company the same loyalty that the company would show to you. For 99.9% of companies, that's nil. Loyalty is a human quality, not one of companies.
I agree with the sentiment but not the conclusion. Given half a chance, I'll happily reassign bad coders to jobs making coffee or some other socially useful task.
But browsers shouldn't crash when they encounter bad html, any more than compilers should crash if they find a syntax error. It's not a question of weakening standards, it's a question of robustness of the tool. The extent to which they try to render malformed html is another question.
We just changed 401K plans, and our new provider (a large CPA firm) REQUIRES IE. The reason: the implementer thought that having CSS was more important than supporting Netscape. She drank the M$ kool-aid, the whole pitcher. Interestingly, the site is just as much of a human-factors nightmare as if that ten-year-old handcoded it. This kind of crap seems to be happening more and more, and it's bad for everyone. Whining about the way naive users write code won't solve this very real problem. Now imagine what it will be like when government agencies, monopoly utilities, etc, start playing hardball this way. To pay your bills, you'll have to pay Bill. If you can't do e-commerce on it, the utility of your Linux box has just declined significantly.
I think it was vigilance, rather than diligence. As a vigilant but not particularly diligent guy, at least I hope so...
Either way, I fully agree with your sentiment. I was just reminding other/.-ers of how powerfully the current flows the other way (flock intertia can be a powerful force too), and was questioning your assertion of what laws are for. But I'll be damned if I'm going to wear a tracking collar for the benefit of Frito-Lay, or Sony, even if it's a virtual one using IPv6 optional features... In that, I think we are in total concurrence.
Since it was from a "2000 in Review" article, it's unsurprising that some of it was old news. Still, it's a nice summary.
I concur with your view that corporations often impose more intrusive, repressive regimes on their employees than even some totalitarian governments have managed to force onto their subjects. So this is why the US opposed Communism?
Incidentally, companies started snooping on their employees long ago. It has just gotten worse lately because it's cheaper to do so, so it's more widely practiced.
Let's hope the mice really do get smarter, and don't just breed faster to compensate for the attrition caused by the traps...
With all due respect, I cannot agree with this highly principled view.
Your description of the origins and purpose of government is a recent, rationalist justification. It explains nicely the way that a small number of somewhat democratic governments have evolved in rich countries. At least, that's how they're now supposed to work in principle.
But, in the more general case, the behavior of governments is better explained by a model combining features of macroparasitism and protection rackets. They bleed you, but it's not in their economic interest to take every drop of blood. Think of ranching or fisheries management for other examples. And they claim that the alternative protection rackets are even worse in order to maintain their position with minimum resistance.
I'm not an anarchist. But let's not kid ourselves about the fact that government, through most of history, has been of absolutely no benefit to the governed. On the contrary, they have been massively destructive.
Consider the oft-cited example of the early Middle Eastern societies, and the "wonderful" benefits of irrigation. Results: thousands of years of slavery, genocide, war, human sacrifice (those charming temples!), malnutrition, ecological destruction through soil salinity, famine and malnutrition due to monoculture, schistosomiasis and malaria from the irrigation systems.
So there is little precedent for government operating in the interest of the people. It is not the default state (pardon the pun). The best we've seen are periods of less effective oppression. Therefore, every opportunity to discard existing precedents, if carried out properly, is an opportunity to prolong and possibly expand freedom. But, without continued resistance, the system will revert to despotism. This is what is now happening on the Web. Commercial interests are aligning with government to find better ways of counting the flock. The better to fleece them...
It's not the maturity of the Net that we're seeing: it's the persistence of destructive and repressive behavior by the powerful. The idea that this is just a choice that people make goes against historical evidence, and implies that the prospects of resisting counter-productive regulation are far simpler than they will be.
Pardon my "idiocy," but it seems that you failed to get the point of my last post. Property rights, and contractual rights, can only exist where the system is not biased in favor of some parties over others. Oligopolies have enough power to subvert the system, and therefore can impose whatever contractual terms they choose: buyers then only have the choice of putting up with them or doing without. For example, choosing not to buy gasoline, or electricity. In cases of inelastic demand, this isn't really much of a choice.
Look at how the railroads dealt with property rights in the golden age of capitalism. They induced the government to seize individuals' property using the power of eminent domain. This is where the railways' rights of way came from in many cases: the little guys' property rights were an inconvenience to the railroads, and the law was on the side of the strong. Now tell me how this proves that capitalism is some "natural" state of affairs that doesn't need to be enforced by rule of law and protected from being hijacked by special interests.
Incidentally, I think that even Friedman would agree with me on this point. Asserting that "rights" will take care of these problems ducks the difficult question of how these rights can be enforced, since it is in the interest of some sellers to subvert the market (and the system that regulates it) at the first opportunity.
You probably won't know about compliance and defection strategies, but they're pertinent here. As soon as a player in the market achieves dominance, if it's cheaper to change the rules to keep out new players than it is to keep competing, they'll do that unless they are prevented. That is, they'll cash in their chips and start playing a new game, with rules that favor them. If that's no longer "capitalism" as you define it, so what? Monopolies don't value your opinion anymore than I do.
BTW, assuming you can read English, this is what the.sig says:
"I steer against them. The force, to which all others must yield, has no effect on me. I move in a direction contrary to the rapidly-whirling world." Notice that the English wouldn't have fit on the.sig line. Latin, in this way, is like Perl: it's concise and powerfully expressive, but not everyone can make sense if it.
Having the CFO be over the CIO guarantees that IT will be viewed as a cost center instead of a strategic asset. Therefore, the bean-counters will starve IT of investment until something crumps in a horrible way. Sometimes this can take months or years. Meanwhile the CFO meets his/her/its financial targets and becomes very rich. But in the medium term, the company fails to compete and everyone finds a new future in the fast-food industry.
You can call it a tragedy, or you can call it natural selection.
Free markets have to be enforced by regulation. Otherwise sellers will collude to rip off the buyers. This has happened since time immemorial. Adam Smith observed: "People of the same trade seldom meet together, even for a merriment and diversion, but the conversation ends in a conspiracy against the public or some contrivance to raise prices."
If that involves sellers purchasing legislation to lock up the market, that's what they'll do. And, again, that's what they have always done. Look at the way the guilds worked. It's all about market segmentation and increased barriers to entry. In the real world, sellers hate competition and will do anything they can to find shelter from it. Without a referee, the capitalist boxing match doesn't turn into a free-for-all. It turns into a rigged fight.
Slashdot Mirror
I'm pretty far left of centre, but have always been deeply suspicious of "group rights."
The government of India's system of quotas for various "backward" castes, the dysfunctional kludges that are called the Belgian and Lebanese governments, and the Quebecker's continuing deranged attempt to legislate language, are prime examples of how disastrous these efforts can become. In the US, initiatives of this sort tend to be racially-based, but the bad idea has been tried all over the world for any number of reasons and seems to have failed everywhere. This is probably an indication that we need to achieve fairness by other means.
Tell me, would the language police shut down an English-language site that called for repeal of the province's law forcing French down its citizens' throats?
Sounds like you may have confused two stages in the "ecological succession" of the Web.
The growth of the Internet (which certainly did cause some change, though far less than electricity, the telegraph or the telephone) happened due largely to the efforts of people in the public sector and universities. This was the Pioneer phase.
The dotcomistas represent the second ("Land Grab") phase. Their goals in life were to get a free ride off the pioneers, make a fast buck, and lock everything down by litigation before competitors showed up to spoil the party. They didn't sow, and it's a pity any of them ever reaped.
So excuse me for not mourning the failure of these sleazebags' ambitions of being the next M$oft. One M$ is one too many. Now these dot-CEOs can fall back to their natural talents at selling timeshare condos and devising multi-level marketing schemes.
And forgive me for lacking sympathy for the suckers (both employees and investors) who believed the get-rich-quick pitches.
The article is dead-on, I really enjoyed it.
Because the OO model and the relational model aren't really either/or alternatives.
The idea that there's one correct paradigm for the code, and another for the data, presupposes a model that doesn't view data and behavior as being intimately connected. That is, your initial assumption denies the validity of the OO approach.
Object models are at a higher level of abstraction. I think it makes better sense to consider use of an RDBMS as an implementation detail, rather than a methodology. I'm painfully familiar with DB-centric methodologies like Oracle CDM. They presuppose a particular solution, then force the designers and implementers to deal with the complexities of the chosen solution, rather than finding an optimal way to represent the problem space. OOD, on the other hand, can lead to a variety of implementations, some using an OO language, others something else.
Bottom line: my beef with RDBMS's is that the usual approach is to pick the tool first, then figure out the requirements after you've already put on the straitjacket. That's a recipe for failed implementations.
Sounds like you're a dedicated technical guy, but I don't think I'd enjoy having you as a manager. To expand a well-known acronym, There's More Than One Way To Do It. It's important to recognize that about people.
I'm pretty enthusiastic about software myself, and am constantly exploring. But, when hiring, I don't give a toss about how excited the applicant gets over language features, as long as they're good at their job. I don't expect them to share my enthusiasm (though it might be nice if I'm looking for someone to talk to over lunch). I've known some people who do fine work, but see computing as just a job, and don't lie awake at night obsessing about queuing algorithms. I find that hard to understand, but accept it. "There's nowt so queer as folk." Her being low-key and not "hairy and unpredictable" might tell you something about her cognitive style, but little about her competency. And, when forming teams, I like to have some predictable people along with us jackrabbits, just to make sure someone's there to sweep up after the party.
Ever noticed how most massive, high-profile public works projects are undertaken by corrupt regimes with inferiority complexes? Albert Speer's brainfarts, Chicago skyscrapers, the Aswan Dam, Brasilia, the Three Gorges Dam, Petronas. Costly attempts to deflect attention from the slimy (in some cases, murderous) activities that got the rulers where they are. Similar to the old investor's rule of thumb: look at the corporate headquarters. If it's grandiose, better put your money somewhere else.
I think it's a little worse than you say.
Oracle has recently been pissing off customers in a big way with a processor-power-based pricing model. More CPUs, more MHz, pay Oracle more $$$. This is a transparent attempt to capture Moore's Law in a revenue stream, and has caused everyone who can take the re-engineering hit to migrate towards DB2, which still has a per-node license that's about half as costly for big installations.
Then there's middleware. Oracle has been moving for some time to become an end-to-end solutions provider, and this means that they're trying to engulf app server and other middleware functionality with Oracle proprietary products that nobody but Oracle takes seriously. When telling Oracle engineers that we'll use well-proven non-Oracle middle-tier products (say, EJB), they start spreading FUD. "Performance! Stability!" We know enough to ignore them, but I'm sure there are plenty of firms out there who have to rely more heavily on vendor advice, and that's how Ellison gets to buy those big boats.
So, I'd say that they'd be a lot more like M$ if they could get away with it.
This makes sense.
I worked developing software in an aerospace company. The DOD was one of our customers. The aerospace company retained copyright to the source we developed. The DOD got the rights to use the source internal to their operations, but not to resell it.
So, the correct answer must be that it depends on the terms of your contract. Standard procurement guidelines probably assign ownership to the government by default, but those terms are negotiable.
Incidentally, the same aerospace company once sold source rights for another system to a foreign military organization, with no restrictions on resale. The customer improved it, then sold the rights to a private company, and they're now competitors. I'd like to know what happened to the bozo who let that contract slip through.
I agree. Part of the reason the web is so easy to deal with is the use of non-binary protocols. Think of http compared to something opaque and binary that might be theoretically more efficient.
I'm happy to trade some plentiful and growing bandwidth for the ability to debug by eyeball. If you have a real need to maximize bandwidth utilization, this probably isn't the solution. But most of the time when I see people optimizing in this way, they're focusing on something that doesn't really matter in the grand scheme of things. Coders' time is expensive, as is debug and test time. Hardware and bandwidth are relatively cheap.
I recently visited a site that dumped you to an error page if you were using anything but IE5 or higher on Windows. It's my employer's 401K provider, so they have a captive audience. The "market" will not sort this out, unless you personally are the buyer. The same goes for health insurance, public services, and other cases where the user of the service isn't the same as the person who pays the bill. Employers or the government will usually contract for for the cheapest service they can get away with. Quality of service for user is a secondary priority.
Remember the old phone company joke: "We don't care. We don't have to."
Wait a minute... government collecting revenues for businesses is "socialism"? Plutocracy or corporate welfare, maybe. But these are capitalist firms that are attempting to subvert the famous free market to get their snouts deep into a new subsidy trough. Oppressive, centralist, unfair and corrupt, yes. Socialist, no. Incidentally, IANAS myself. But I also have seen a number of instances when Schumpeter has been proven right. Firms will suck up to the government to duck competition whenever they can, and if that is not resisted, free markets disappear.
Incidentally, this should prove advantageous to the US if the Europeans go further down this road and we have the sense not to.
The Guinness family were pro-independence, but were and are Protestant. And they're big supporters of the Conservative Party in England. Highly unlikely that they've ever backed the IRA.
I do, however, prefer Beamish or Murphy's. Not quite so heavy.
Currently, in the US, if a company has a lawyer on the payroll, this is a more or less risk-free strategy for them, but it's risky for the defendant, since it costs you money.
Maybe the Brits have the right idea. In some civil suits (for example, libel), if the judge finds in favor of the defendant, the plaintiff can be ordered to pay the defendant's court costs. This is to deter exactly this kind of tactical lawsuit.
What you're saying sounds a bit like "We tried using perl and the project failed miserably."
But I do agree that most of the benefit from Extreme comes when you have experienced programmers who can work without a safety net.
Frequent unit testing should catch the coding errors, but the project scope will spiral out of control unless there's someone there who knows what's feasible and what isn't, and you need someone who can actually make the code work.
It's a good methodology in the right situations. Pity that it has become "flavor of the month" at some sites. But it sounds like your girlfriend's experience had more to do with it being a badly-staffed, dysfunctional company than with the methodology they adopted.
Not to mention JPython/Jython, which is an implementation of Python running on the JVM.
Software engineering is a good thing, and I've seen it really pay off. But certification is either unnecessary or counter-productive.
Certification works best in technically stagnant fields like architecture and civil engineering, and there, mainly to restrict the supply of architects and civil engineers. The pace of change in software (and, to a lesser extent, software engineering) is too rapid for that. In any case, certification is more often a barrier to entry to a profession than a means of consumer protection.
I've got the CS degree myself, and lots of pieces of paper saying I know how to do this and that. Their only real value is to soothe nervous clients. Most of what I've learned that's of value was learned outside the degree and certification programs, and could not be learned in such programs. Colleagues of mine with degrees in other fields, or no degrees at all, are just as good, or better, software engineers than I am.
Remember, NOBODY who founded the CS field had a CS degree. There weren't any then. That includes Turing, Von Neumann, Wiener, Hamming, Knuth. I could go on.
In fact, in the UK, a suspect's refusal to testify can be used as evidence of guilt.
Jack Straw is also attempting to abolish trial by jury for most criminal cases, on the grounds that juries sometimes refuse to convict suspects. And the British law allowing the police to require anyone involved in a criminal investigation to decrypt any information the police ask for (and this can be a "fishing expedition"-- the American concept of probable cause is much weaker in the UK; all they need is a magistrate to rubber-stamp it) is as near totalitarian as anything in the golden age of the USSR or Hoxha's Albania.
Further evidence that there is really no distinction between authoritarian center-left and authoritarian right-wing governments. The only difference is that between those who have power, and the rest of us. The British public is too cowed or apathetic to kick back, and the only gains in human rights in that country have come from the (not particularly democratic) EU, particularly the incorporation of the European Convention on Human Rights into British law.
A sad situation, and if we see signs of any of this happening in the US, we should resist it.
When I started managing a sizeable team for the first time, someone I trusted told me that, if I really wanted to be successful, I should start by getting rid of the indispensible people.
This seemed at the time to be evil, cynical advice. But it had some truth in it. Two absolutely key employees left a few months later (better offer, not anything to do with the project) and we did just fine without them. Indispensible = self-marketing plus refusal to share knowledge. Both these individual traits are destructive to teams.
The flip side is that, when that better offer comes, manage the succession as best you can, and walk out that door . Anyway, if you're really that good, you will already have been sharing knowledge with others, and encouraging their development. In that case, this is just another growth opportunity for you and them.
You owe the company the same loyalty that the company would show to you. For 99.9% of companies, that's nil. Loyalty is a human quality, not one of companies.
I agree with the sentiment but not the conclusion. Given half a chance, I'll happily reassign bad coders to jobs making coffee or some other socially useful task.
But browsers shouldn't crash when they encounter bad html, any more than compilers should crash if they find a syntax error. It's not a question of weakening standards, it's a question of robustness of the tool. The extent to which they try to render malformed html is another question.
We just changed 401K plans, and our new provider (a large CPA firm) REQUIRES IE. The reason: the implementer thought that having CSS was more important than supporting Netscape. She drank the M$ kool-aid, the whole pitcher. Interestingly, the site is just as much of a human-factors nightmare as if that ten-year-old handcoded it. This kind of crap seems to be happening more and more, and it's bad for everyone. Whining about the way naive users write code won't solve this very real problem. Now imagine what it will be like when government agencies, monopoly utilities, etc, start playing hardball this way. To pay your bills, you'll have to pay Bill. If you can't do e-commerce on it, the utility of your Linux box has just declined significantly.
I think it was vigilance, rather than diligence. As a vigilant but not particularly diligent guy, at least I hope so...
/.-ers of how powerfully the current flows the other way (flock intertia can be a powerful force too), and was questioning your assertion of what laws are for. But I'll be damned if I'm going to wear a tracking collar for the benefit of Frito-Lay, or Sony, even if it's a virtual one using IPv6 optional features... In that, I think we are in total concurrence.
Either way, I fully agree with your sentiment. I was just reminding other
Since it was from a "2000 in Review" article, it's unsurprising that some of it was old news. Still, it's a nice summary.
I concur with your view that corporations often impose more intrusive, repressive regimes on their employees than even some totalitarian governments have managed to force onto their subjects. So this is why the US opposed Communism?
Incidentally, companies started snooping on their employees long ago. It has just gotten worse lately because it's cheaper to do so, so it's more widely practiced.
Let's hope the mice really do get smarter, and don't just breed faster to compensate for the attrition caused by the traps...
With all due respect, I cannot agree with this highly principled view.
Your description of the origins and purpose of government is a recent, rationalist justification. It explains nicely the way that a small number of somewhat democratic governments have evolved in rich countries. At least, that's how they're now supposed to work in principle.
But, in the more general case, the behavior of governments is better explained by a model combining features of macroparasitism and protection rackets. They bleed you, but it's not in their economic interest to take every drop of blood. Think of ranching or fisheries management for other examples. And they claim that the alternative protection rackets are even worse in order to maintain their position with minimum resistance.
I'm not an anarchist. But let's not kid ourselves about the fact that government, through most of history, has been of absolutely no benefit to the governed. On the contrary, they have been massively destructive.
Consider the oft-cited example of the early Middle Eastern societies, and the "wonderful" benefits of irrigation. Results: thousands of years of slavery, genocide, war, human sacrifice (those charming temples!), malnutrition, ecological destruction through soil salinity, famine and malnutrition due to monoculture, schistosomiasis and malaria from the irrigation systems.
So there is little precedent for government operating in the interest of the people. It is not the default state (pardon the pun). The best we've seen are periods of less effective oppression. Therefore, every opportunity to discard existing precedents, if carried out properly, is an opportunity to prolong and possibly expand freedom. But, without continued resistance, the system will revert to despotism. This is what is now happening on the Web. Commercial interests are aligning with government to find better ways of counting the flock. The better to fleece them...
It's not the maturity of the Net that we're seeing: it's the persistence of destructive and repressive behavior by the powerful. The idea that this is just a choice that people make goes against historical evidence, and implies that the prospects of resisting counter-productive regulation are far simpler than they will be.
Pardon my "idiocy," but it seems that you failed to get the point of my last post. Property rights, and contractual rights, can only exist where the system is not biased in favor of some parties over others. Oligopolies have enough power to subvert the system, and therefore can impose whatever contractual terms they choose: buyers then only have the choice of putting up with them or doing without. For example, choosing not to buy gasoline, or electricity. In cases of inelastic demand, this isn't really much of a choice.
.sig says: .sig line. Latin, in this way, is like Perl: it's concise and powerfully expressive, but not everyone can make sense if it.
Look at how the railroads dealt with property rights in the golden age of capitalism. They induced the government to seize individuals' property using the power of eminent domain. This is where the railways' rights of way came from in many cases: the little guys' property rights were an inconvenience to the railroads, and the law was on the side of the strong. Now tell me how this proves that capitalism is some "natural" state of affairs that doesn't need to be enforced by rule of law and protected from being hijacked by special interests.
Incidentally, I think that even Friedman would agree with me on this point. Asserting that "rights" will take care of these problems ducks the difficult question of how these rights can be enforced, since it is in the interest of some sellers to subvert the market (and the system that regulates it) at the first opportunity.
You probably won't know about compliance and defection strategies, but they're pertinent here. As soon as a player in the market achieves dominance, if it's cheaper to change the rules to keep out new players than it is to keep competing, they'll do that unless they are prevented. That is, they'll cash in their chips and start playing a new game, with rules that favor them. If that's no longer "capitalism" as you define it, so what? Monopolies don't value your opinion anymore than I do.
BTW, assuming you can read English, this is what the
"I steer against them. The force, to which all others must yield, has no effect on me. I move in a direction contrary to the rapidly-whirling world." Notice that the English wouldn't have fit on the
I've seen this situation.
Having the CFO be over the CIO guarantees that IT will be viewed as a cost center instead of a strategic asset. Therefore, the bean-counters will starve IT of investment until something crumps in a horrible way. Sometimes this can take months or years. Meanwhile the CFO meets his/her/its financial targets and becomes very rich. But in the medium term, the company fails to compete and everyone finds a new future in the fast-food industry.
You can call it a tragedy, or you can call it natural selection.
Free markets have to be enforced by regulation. Otherwise sellers will collude to rip off the buyers. This has happened since time immemorial. Adam Smith observed: "People of the same trade seldom meet together, even for a merriment and diversion, but the conversation ends in a conspiracy against the public or some contrivance to raise prices."
If that involves sellers purchasing legislation to lock up the market, that's what they'll do. And, again, that's what they have always done. Look at the way the guilds worked. It's all about market segmentation and increased barriers to entry. In the real world, sellers hate competition and will do anything they can to find shelter from it. Without a referee, the capitalist boxing match doesn't turn into a free-for-all. It turns into a rigged fight.