A college degree in what? Frankly, with the exception of history or econ majors, I don't know what I'd do with graduate with a liberal arts major. I can plug someone with a science or engineering background into a lot of slots, but the share of US graduates in science, engineering, and mathematics has been in rapid decline for some time now. Maybe that's why you are seeing college grads in jobs you feel are beneath them.
Actually, the point is that XML is both human and machine readable. If all you cared about is machine readable structured data then why didn't ASN.1 and all of it's ultra efficient encoding schemes take off? I would maintain it was because they were human opaque.
With XML, I can machine parse it easily, but I can also sit down and interact with it as a human if I have to. This is a huge leap forward (at least from my perspective).
Expos facto laws are unconsitutional in the US, and unlike other aspects of our constitution that our legislatures seem to have decided don't exist (and the, fortunately smaller, elements of our constitution that the *courts* have forgotten exists) the expos facto restriction is still alive and well.
No, actually 1 is a valid reason. Open sourcing code takes energy. I've got tons of code I write for work that isn't in shape to make sense even to release it through our internal processes at work (which are on an order of difficulty with starting a sourceforge project). The code was written to help me do my job, quick and dirty, and it was never worth while to massage it into utility for anyone else.
Think of it this way. Should all publically produced code be open sourced? Think really hard about that and I think you will discover at least two cases where it shouldn't:
1) No one else could possible care about this code.
The majority of the code written anywhere is this way. It's a one off to solve a particular problem that is not shared by others in a way that would allow reusing your code.
2) This code contains information that for a variety of reasons is confidential (security, personal privacy, etc).
A lot of code is probably also disqualifed from open sourcing because of this one.
If you accept that not all government code should be open sourced (either for the reasons given above, or others) then the question becomes:
How do you decide what to open source?
This can be done a lot of ways. You can allow anyone in the organization to make the decision themselves. In a small tight knit org where you trust everyone, this might be good, but think about some of the federal employees you've dealt with...
So you are left with trying to produce a policy as to what should be open sourced and what shouldn't. This then gives rise to a mechanism for at very least communicating the policy, and probably also requiring reporting on compliance with the policy so you can audit the policy and figure out what went wrong when problems inevidably occur.
So where are we now. We've got policy. We've got policy communications mechanism. We've got reporting requirements. We've got auditing mechanism... and there we are:)
In healthy (usually private sector) orgs, the policy is lightweight, the policy mechanism is an easy to use web tool which handles reporting and allows auditing.
This sort of sanity would be truely surprising in a government org.
I agree. There's nothing special about it. I consider myself a medium (not high) grade tech. I don't even want to know some of the languages (like Tcl), I just happened to need to learn them and use them in certain unfortunate circumstances.
Languages I've written code in thats still in production:
Perl, Python, Tcl, C, C++, Java, PHP
Things some might disagree with calling programming languages that I've written code in that's still in production:
sh, csh
Languages I've dabbled at, but don't consider myself to actually understand:
scheme, postscript,fortran
I don't tend to count awk,sed, or regular expressions, Makefile, but am low level facile with all of those.
I'm currently trying to fit in learning Ruby, as I don't see why people would use it in preference to Python, but there are enough smart people who swear by it that there must be something there.
I'd love to have the time to learn Smalltalk and ObjectiveC, as I keep hearing REALLY cool things about them.
I've found that near complete ignorance of Windows is my best defense. I've not been a serious Windows user since 1994. So when someone asks me for assistance with their Windows problem I can quite truthfully say:
"I'm sorry, I don't know how that works."
Don't get me wrong, I make my living in tech. I code in between 8 and 12 languages (depending on how good my memory is that day), can play a medium grade Linux/Solaris guru when necessary, write web apps, architect large distributed systems, operate a wide variety of service provider and enterprise networking equipment, etc. I also like helping people who are having technical problems. But there's a big difference between being the IM of last resort for various Linux/Python,etc problems and having to deal with Windows users.
Popn GWP CO2 CO2/GWP China 20.0% 12.5% 15.2% 1.22 Russia 2.3% 2.5% 6.7% 2.68 Japan 2.0% 7.0% 5.3% 0.75 France 0.9% 3.2% 1.6% 0.5 Germany 1.3% 4.4% 3.7% 0.84 UK 0.9% 3.2% 2.5% 0.78 USA 4.6% 21.0% 25.2% 1.2
CO2/GWP is important because what it shows is the pollution efficiency of a nation. If you presume that we still need to produce at least as much Gross World Product (GWP) as we are currently doing, and you want to minimize pollution then you would seek to favor countries that are more pollution efficient (ie lower CO2/GWP).
Actually... the idea is to get the same cost per unit output. No one in business really cares how much they pay their employees (or at least no one with any brains). What they care about is what they have to pay per unit output. If hiring a rural-american to costs $30/hour, and they produce 10 widgets/hour, and hiring an Indian costs $5/hour and they produce 1 widget/hour, you'd have to be incredibly dim to hire the Indian ($5/widget) vs the rural-american ($3/widget). It's all about costs and productivity.
Outsourcing can incur serious productivity losses due to time differences, distance, coordination difficulties etc.
What's really killing a lot of states is business unfriendly laws and taxes. If you add costs to employeeing someone through taxes and regulation, then they either have to take a lower wage (thus bringing costs back down to parity) or have greater productivity (thus bringing cost/widget back in line). A lot of states seem to completely fail to realized this.
You do realize that if you systematically increase a corporations costs, then the corporation either passes those costs on to consumers or gets out of the high cost business, right?
So if I make a whole industry systematically prone to expensive lawsuits, then I either make that industries products more expensive, or I eliminate that industry.
I don't see making the tort system rational as corporate welfare. I see it as repealing the trial lawyer tax on consumers.
1) Do not limit punative damages. They are there for a purpose, to punish egregious behavior on the part of the defendant. 2) Do not pay punative damages to the plaintiff or plaintiff's attorneys. Pay them to a randomly selected peer juristiction.
Punative damages are there to punish egregiously bad behavior. They need to remain, and they need to remain uncapped, so that less scrupulous parties will have the threat of unlimited liability for malfesance. However, most behavior falls far short of malfesance, and is probably not a good candidate for punative damages. In order for punative damages to function effectively though non-punative incentives for awarding them (like sympathy for the plaintiff) need to be removed.
Actually... I've run the numbers for effective tax rates for earners in Single, Head of Household, and Married filing statuses including the payroll taxes, Earned Income Tax Credit, Child Tax Credit, Standard Deductions, Exemptions (including phase out), for various household sizes and incomes here:
http://fairtaxblog.blogspot.com/
in an attempt to bring together a factual discussion of the FairTax proposal. Taking all of these into account the poor are generally still paying a negative effective tax rate (thanks to the Earned Income Tax Credit).
Please note though, as I've assumed (for purposes of my comparison) that 100% of income is spent, I've not included the Federal Saving Tax Credit (for low income savers). Also note that when you get to higher incomes you gain a lot of extra latitude in avoiding taxes that I can't really account for systematically, so the effective tax rates for higher earnings are probably somewhat exaggerated in ways I can't correct for.
You do realize that historically it is normal for around 35-40% of the non-institutionalized working age (ie not imprisoned or commited) population not be employed?
And if you'd like to see about part-time workers you can get recent statistics here:
http://www.bls.gov/webapps/legacy/cpsatab5.htm Please note, there's nothing wrong with a part of the population being employed part time by choice. The BLS statistics differentiate these as people employed part time for economic/non-economic reasons... if you look at the stats you'll see around 4-5 million people are employed part-time for economic reasons. You can find those stats here:
http://www.bls.gov/webapps/legacy/cpsatab5.htm According to the BLS the labor force is around 150 million:
http://www.bls.gov/webapps/legacy/cpsatab1.htm Out of a labor force of around 150 million that means that we have about 3-4% of the labor force working part time because they can't get full time work. While I'd prefer this to be a lower percentage, it doesn't seem to be a great and shocking problem.
Also, would you please source your data above. Absent primary sourcing I've absolutely no reaon to believe your facts.
There's plenty of spectrum. The problem is that we squander it in stupid ways. Take for instance broadcast television. 80%+ of households in the US have cable, but we still dedicate vast swaths of really good spectrum to the exclusive use of television broadcasters who didn't even pay for it originally in most cases. Then we compound it by giving them more spectrum for digital broadcast. Let's do the math: a 6Mhz channel can carry 38Mb per second, broadcast quality TV can be transmitted in about 3 Mb per second. So even if broadcasting video over the air were a good idea (which I maintain it isn't), we are using 10+ times as much spectrum as necessary to do it. I'll start taking people seriously who claim we are running out of spectrum when you can't get TV with rabbit ears anymore...
Actually, in some places it's worse. My brother at one point wanted no outgoing LD from his phone. The phone company charged him a fee for access to the LD network, and then a fee to block LD. When queried they maintained that everyone who had a phone line had to pay the LD access fee, but blocking LD was then a service. So you paid to both have and not have long distance.
And those dollars get spent to buy US Treasuries. Which keeps US interest rates low. So not only are the Chinese sending us artificially cheap goods, they are lending us money at artificially low rates.
As near as I can tell only the citizens of China are getting screwed here. US consumers are making out like bandits.
Isn't it kind of the Chinese government to subsidize US consumers this way? What you are basically saying is that everytime I chip in $40 for a DVD player, the Chinese government chips in between $40 and $120 towards the cost of that DVD player.
Yes, now all they need to do is spend the 6 billion in cash they have and stop grossing 1+ bill in revenue per quarter.
Umm... if you check their financials you'll
see they only have about 2.7 billion in cash:
http://finance.yahoo.com/q/ks?s=SUNW
And they are carrying about 1.47 Billion in debt.
Given that they ran a negative cashflow of -20 million dollars last year, they could keep this up for some time. You revenue is irrelavent, it's your earnings and cash flow that count.
Do 400,000+ transactions per hour 24/7 on your home built pc and get back to me.
That must be why there are no Sun boxes in the TPC-C top ten.
http://www.tpc.org/tpcc/results/tpcc_perf_results. asp
I guess if you want your corporate IT department to rival that of a medium sized College, you could squeeze extra performance out of Linux.
Or you could field twice that many people managing your relationships with the proprietary providers of the software you need. In most cases it's the proprietary code that is the bottleneck (in my experience in industry). Or you could also field extra sysadmins to work on compiling and integrating all the FOSS programs that your users actually need that Solaris doesn't ship with.
Oh... and working around the issues and problems with Solaris, like the fact that they screwed up their version of BPF so badly that the libpcap folks found it was faster to filter in user space. Or the fact that their packet sniffing interface doesn't hand over the whole frame received, but trunkates it for you to the size indicated by the ethernet header making them useless for certain kinds of tasks. Or the million and one other little things that are broken in Solaris that will NEVER be fixed.
So the Opteron is about twice as fast at int and 30% faster at float. So while you can get more processors from Sun than an x86 base, you may not get more performance.
Their OS is less feature rich, but has more bugs, and doesn't perform as well in most cases as Linux.
Look around, everyone who possibly can is getting off of Suns and onto Linux x86. The major things holding most of Suns customers back in this regard are proprietary software support, and that's improving all the time.
And as to Java... I'm not sure exactly how they intend to make money there... IBM does the Java services market SOOOO much better than Sun does.
Re:Something good may yet come out of this
on
Out of Gas
·
· Score: 1
It is true that there are externalities involved with oil consumption, but I'm aware of NO objective way to quantify them and thus include them in the cost of gas. It should be noted that many of the externalities you list are peculiar to America, as only the environmental one is born by Europe.
Re:Why energy and food are frequently excluded.
on
Out of Gas
·
· Score: 1
You will note that we seem to have three kinds of things in the market:
1) Commodities, which flucuate wildly in price around their equillibrium price. 2) Government induces market failures (health care, education, cable tv and other gov guaranteed monopolies), which increase in price. 3) Most other products, which decrease in price.
So almost every non-commodity good where government hasn't caused a market failure is experiencing rapid supply side deflation... interesting...
Slashdot Mirror
A college degree in what? Frankly, with the exception of history or econ majors, I don't know what I'd do with graduate with a liberal arts major. I can plug someone with a science or engineering background into a lot of slots, but the share of US graduates in science, engineering, and mathematics has been in rapid decline for some time now. Maybe that's why you are seeing college grads in jobs you feel are beneath them.
Actually, the point is that XML is both human and machine readable. If all you cared about is machine readable structured data then why didn't ASN.1 and all of it's ultra efficient encoding schemes take off? I would maintain it was because they were human opaque.
With XML, I can machine parse it easily, but I can also sit down and interact with it as a human if I have to. This is a huge leap forward (at least from my perspective).
Expos facto laws are unconsitutional in the US, and unlike other aspects of our constitution that our legislatures seem to have decided don't exist (and the, fortunately smaller, elements of our constitution that the *courts* have forgotten exists) the expos facto restriction is still alive and well.
No, actually 1 is a valid reason. Open sourcing code takes energy. I've got tons of code I write for work that isn't in shape to make sense even to release it through our internal processes at work (which are on an order of difficulty with starting a sourceforge project). The code was written to help me do my job, quick and dirty, and it was never worth while to massage it into utility for anyone else.
Think of it this way. Should all publically produced code be open sourced? Think really hard about that and I think you will discover at least two cases where it shouldn't:
:)
1) No one else could possible care about this code.
The majority of the code written anywhere is this way. It's a one off to solve a particular problem that is not shared by others in a way that would allow reusing your code.
2) This code contains information that for a variety of reasons is confidential (security, personal privacy, etc).
A lot of code is probably also disqualifed from open sourcing because of this one.
If you accept that not all government code should be open sourced (either for the reasons given above, or others) then the question becomes:
How do you decide what to open source?
This can be done a lot of ways. You can allow anyone in the organization to make the decision themselves. In a small tight knit org where you trust everyone, this might be good, but think about some of the federal employees you've dealt with...
So you are left with trying to produce a policy as to what should be open sourced and what shouldn't. This then gives rise to a mechanism for at very least communicating the policy, and probably also requiring reporting on compliance with the policy so you can audit the policy and figure out what went wrong when problems inevidably occur.
So where are we now. We've got policy. We've got policy communications mechanism. We've got reporting requirements. We've got auditing mechanism... and there we are
In healthy (usually private sector) orgs, the policy is lightweight, the policy mechanism is an easy to use web tool which handles reporting and allows auditing.
This sort of sanity would be truely surprising in a government org.
I agree. There's nothing special about it. I consider myself a medium (not high) grade tech. I don't even want to know some of the languages (like Tcl), I just happened to need to learn them and use them in certain unfortunate circumstances.
Languages I've written code in thats still
in production:
Perl, Python, Tcl, C, C++, Java, PHP
Things some might disagree with calling programming languages that I've written code in that's still in production:
sh, csh
Languages I've dabbled at, but don't consider myself to actually understand:
scheme, postscript,fortran
I don't tend to count awk,sed, or regular expressions, Makefile, but am low level facile with all of those.
I'm currently trying to fit in learning Ruby, as I don't see why people would use it in preference to Python, but there are enough smart people who swear by it that there must be something there.
I'd love to have the time to learn Smalltalk and ObjectiveC, as I keep hearing REALLY cool things about them.
I've found that near complete ignorance of Windows is my best defense. I've not been a serious Windows user since 1994. So when someone asks me for assistance with their Windows problem I can quite truthfully say:
"I'm sorry, I don't know how that works."
Don't get me wrong, I make my living in tech. I code in between 8 and 12 languages (depending on how good my memory is that day), can play a medium grade Linux/Solaris guru when necessary, write web apps, architect large distributed systems, operate a wide variety of service provider and enterprise networking equipment, etc. I also like helping people who are having technical problems. But there's a big difference between being the IM of last resort for various Linux/Python,etc problems and having to deal with Windows users.
Just to carry this out one more column:
Popn GWP CO2 CO2/GWP
China 20.0% 12.5% 15.2% 1.22
Russia 2.3% 2.5% 6.7% 2.68
Japan 2.0% 7.0% 5.3% 0.75
France 0.9% 3.2% 1.6% 0.5
Germany 1.3% 4.4% 3.7% 0.84
UK 0.9% 3.2% 2.5% 0.78
USA 4.6% 21.0% 25.2% 1.2
CO2/GWP is important because what it shows is
the pollution efficiency of a nation. If you
presume that we still need to produce at least
as much Gross World Product (GWP) as we are
currently doing, and you want to minimize
pollution then you would seek to favor countries
that are more pollution efficient (ie lower CO2/GWP).
Actually... the idea is to get the same cost per unit output. No one in business really cares how much they pay their employees (or at least no one with any brains). What they care about is what they have to pay per unit output. If hiring a rural-american to costs $30/hour, and they produce 10 widgets/hour, and hiring an Indian costs $5/hour and they produce 1 widget/hour, you'd have to be incredibly dim to hire the Indian ($5/widget) vs the rural-american ($3/widget). It's all about costs and productivity.
Outsourcing can incur serious productivity losses due to time differences, distance, coordination difficulties etc.
What's really killing a lot of states is business unfriendly laws and taxes. If you add costs to employeeing someone through taxes and regulation, then they either have to take a lower wage (thus bringing costs back down to parity) or have greater productivity (thus bringing cost/widget back in line). A lot of states seem to completely fail to realized this.
You do realize that if you systematically increase a corporations costs, then the corporation either passes those costs on to consumers or gets out of the high cost business, right?
So if I make a whole industry systematically prone to expensive lawsuits, then I either make that industries products more expensive, or I eliminate that industry.
I don't see making the tort system rational as corporate welfare. I see it as repealing the trial lawyer tax on consumers.
My suggestion in general:
1) Do not limit punative damages. They are there for a purpose, to punish egregious behavior on the part of the defendant.
2) Do not pay punative damages to the plaintiff or plaintiff's attorneys. Pay them to a randomly selected peer juristiction.
Punative damages are there to punish egregiously bad behavior. They need to remain, and they need to remain uncapped, so that less scrupulous parties will have the threat of unlimited liability for malfesance. However, most behavior falls far short of malfesance, and is probably not a good candidate for punative damages. In order for punative damages to function effectively though non-punative incentives for awarding them (like sympathy for the plaintiff) need to be removed.
Actually... I've run the numbers for effective tax rates for earners in Single, Head of Household, and Married filing statuses including the payroll taxes, Earned Income Tax Credit, Child Tax Credit, Standard Deductions, Exemptions (including phase out), for various household sizes and incomes here:
http://fairtaxblog.blogspot.com/
in an attempt to bring together a factual discussion of the FairTax proposal. Taking all of these into account the poor are generally still paying a negative effective tax rate (thanks to the Earned Income Tax Credit).
Please note though, as I've assumed (for purposes of my comparison) that 100% of income is spent, I've not included the Federal Saving Tax Credit (for low income savers). Also note that when you get to higher incomes you gain a lot of extra latitude in avoiding taxes that I can't really account for systematically, so the effective tax rates for higher earnings are probably somewhat exaggerated in ways I can't correct for.
You do realize that historically it is normal for around 35-40% of the non-institutionalized working age (ie not imprisoned or commited) population not be employed?
t xt
ftp://ftp.bls.gov/pub/special.requests/lf/aat1.
And if you'd like to see about part-time workers you can get recent statistics here:
http://www.bls.gov/webapps/legacy/cpsatab5.htm
Please note, there's nothing wrong with a part of the population being employed part time by choice. The BLS statistics differentiate these as people employed part time for economic/non-economic reasons... if you look at the stats you'll see around 4-5 million people are employed part-time for economic reasons. You can find those stats here:
http://www.bls.gov/webapps/legacy/cpsatab5.htm
According to the BLS the labor force is around 150 million:
http://www.bls.gov/webapps/legacy/cpsatab1.htm
Out of a labor force of around 150 million that means that we have about 3-4% of the labor force working part time because they can't get full time work. While I'd prefer this to be a lower percentage, it doesn't seem to be a great and shocking problem.
Also, would you please source your data above. Absent primary sourcing I've absolutely no reaon to believe your facts.
There's plenty of spectrum. The problem is that we squander it in stupid ways. Take for instance broadcast television. 80%+ of households in the US have cable, but we still dedicate vast swaths of really good spectrum to the exclusive use of television broadcasters who didn't even pay for it originally in most cases. Then we compound it by giving them more spectrum for digital broadcast. Let's do the math: a 6Mhz channel can carry 38Mb per second, broadcast quality TV can be transmitted in about 3 Mb per second. So even if broadcasting video over the air were a good idea (which I maintain it isn't), we are using 10+ times as much spectrum as necessary to do it. I'll start taking people seriously who claim we are running out of spectrum when you can't get TV with rabbit ears anymore...
Actually, in some places it's worse. My brother at one point wanted no outgoing LD from his phone. The phone company charged him a fee for access to the LD network, and then a fee to block LD. When queried they maintained that everyone who had a phone line had to pay the LD access fee, but blocking LD was then a service. So you paid to both have and not have long distance.
And if they loose, the each member of the class could provide a coupon to the RIAA for $5 off a CD...
And those dollars get spent to buy US Treasuries. Which keeps US interest rates low. So not only are the Chinese sending us artificially cheap goods, they are lending us money at artificially low rates.
As near as I can tell only the citizens of China are getting screwed here. US consumers are making out like bandits.
Isn't it kind of the Chinese government to subsidize US consumers this way? What you are basically saying is that everytime I chip in $40 for a DVD player, the Chinese government chips in between $40 and $120 towards the cost of that DVD player.
How is this a bad thing again?
Umm... if you check their financials you'll see they only have about 2.7 billion in cash:
http://finance.yahoo.com/q/ks?s=SUNW
And they are carrying about 1.47 Billion in debt. Given that they ran a negative cashflow of -20 million dollars last year, they could keep this up for some time. You revenue is irrelavent, it's your earnings and cash flow that count.
Do 400,000+ transactions per hour 24/7 on your home built pc and get back to me.
That must be why there are no Sun boxes in the TPC-C top ten.
http://www.tpc.org/tpcc/results/tpcc_perf_results. asp
I guess if you want your corporate IT department to rival that of a medium sized College, you could squeeze extra performance out of Linux.
Or you could field twice that many people managing your relationships with the proprietary providers of the software you need. In most cases it's the proprietary code that is the bottleneck (in my experience in industry). Or you could also field extra sysadmins to work on compiling and integrating all the FOSS programs that your users actually need that Solaris doesn't ship with.
Oh... and working around the issues and problems with Solaris, like the fact that they screwed up their version of BPF so badly that the libpcap folks found it was faster to filter in user space. Or the fact that their packet sniffing interface doesn't hand over the whole frame received, but trunkates it for you to the size indicated by the ethernet header making them useless for certain kinds of tasks. Or the million and one other little things that are broken in Solaris that will NEVER be fixed.
Since another poster was kind enough to address most of your software points, I'll address the hardware:
2 004q1 /cpu2000-20040112-02710.html
s 2004q1 /cpu2000-20040112-02709.html
s 2004q1 /cpu2000-20040209-02854.htmls 2004q1 /cpu2000-20040112-02709.html
Chip for chip the UltraSparc is slower:
1.28Ghz Ultra IIIi (the newest Sun chip for which
I can get spec benchmarks):
Specint: 704
http://www.specbench.org/cpu2000/results/res
Specfp: 1063
http://www.specbench.org/cpu2000/results/re
Opteron 146:
Specint: 1354
http://www.specbench.org/cpu2000/results/re
Specfp: 1394
http://www.specbench.org/cpu2000/results/re
So the Opteron is about twice as fast at int and 30% faster at float. So while you can get more processors from Sun than an x86 base, you may not get more performance.
Sun's pretty much a dead company walking.
Their hardware is more expensive, and slower.
Their OS is less feature rich, but has more bugs, and doesn't perform as well in most cases as Linux.
Look around, everyone who possibly can is getting off of Suns and onto Linux x86. The major things holding most of Suns customers back in this regard are proprietary software support, and that's improving all the time.
And as to Java... I'm not sure exactly how they intend to make money there... IBM does the Java services market SOOOO much better than Sun does.
Because 2.6.6 is in unstable :)
It is true that there are externalities involved with oil consumption, but I'm aware of NO objective way to quantify them and thus include them in the cost of gas. It should be noted that many of the externalities you list are peculiar to America, as only the environmental one is born by Europe.
You will note that we seem to have three kinds of things in the market:
1) Commodities, which flucuate wildly in price around their equillibrium price.
2) Government induces market failures (health care, education, cable tv and other gov guaranteed monopolies), which increase in price.
3) Most other products, which decrease in price.
So almost every non-commodity good where government hasn't caused a market failure is experiencing rapid supply side deflation... interesting...