So, all fishing (a.k.a. priate) boat crew members put on life jackets or, if they have it, get in life raft.
Abandon (fishing, pirate) boat.
Merchant/Commercial ship spends two hours picking them up ONE-BY-ONE.
Searching each.
Restraining each.
All are just fisherman, fed and sheltered and dropped at next port of call.
Not all "fisherman"? All pirate (oops., fishing) boat members put on trial and, if guilty, executed within 30 days of conviction.
Not hard.
Hundreds of thousands die prematurely every day in the world due to malnutrition and disease. Why not deal decisively with pirates who increase that number by increasing shipping costs for staple supplies like wheat and other grains/grasses?
But, what if AGW isn't the primary cause of such change. If so, accepting AGW as "the answer" due to faulty science motivated by political (and funding) agendas could be a big mistake. At best we would waste resources and not significantly affect the outcome; at worst, we would fail to address a problem that, had we had not erroneously attributed it to AGW, we could have mitigated or stopped.
At this point, I think we should be devoting most grant money to disproving AGW as a significant factor in climate variations. If AGW theories and predictions stand up under the resulting scientific scrutiny we can have more confidence that AGW is, indeed, a serious problem. Unfortunately, with most funding going to further prove the premise and much research being done by "scientists" who have already made up their minds (Jones et al), it's not surprising that most research supports the premise.
We need to be careful that we are not deluded by Cargo Cult Science as described by Feynman. Unfortunately, the extent of AGW isn't precisely measurable or provable and never will be - the system is too complex and there are too many external variables which can't be controlled. This makes it all the more important that the scientific debate be open and honest.
Apple has something of an advantage over notebook vendors such as HP, Acer, and Toshiba because Apple doesn't sell really low end notebooks.
CR's notebook reliability numbers are somewhat meaningless for comparing Apple to vendors which manufacture sub $500 notebooks. The low end notebooks have to cut corners which will reduce reliability (more plastic, less metal etc.). CR should really compare reliability of notebooks by vendor based on price (perhaps in two classes - $750 for base product).
The article you linked to says Therac-25 failed unit testing (on live units unfortunately).
I think the confusion is yours - the failures were in production not unit test. The point is, the Therac-25 seems to have passed the unit tests - but that clearly wasn't enough to determine that the product worked correctly.
Failing to include good comments in a complex regex in production code, esp. if the regex needs to deal with oddities in the input, should be a capital offense.
I agree for the most part - I meant to push back on the "if it passes the Unit Tests, then how is it really bad?" part - passing unit tests is almost never sufficient, although it's almost always necessary.
Also, it doesn't sound like the "quadrant coder" was writing some script to be used only once or twice -- he bothered to put comments in it which suggests to me it wasn't created in/tmp and intended to never be run again. (I actually assumed the poster was going for "funny" until I read the number of responses to his comment!)
Also, no real life complex system is ever "perfect". That's why we sometimes try to reduce the risk of catastrophic failure in life critical systems by implementing independent systems whose entire job is to detect a failure in another part and enforce safety constraints. But this solution still isn't 'perfect' - just much less likely to result in catastrophic failure.
Apparently there *is* a dark side to a high-quality unit test suite...
How true... I worked at a place where one group used agile methods (sort of). It was amusing to watch them argue to a group that their software interfaced with that "it can't be our problem, it passed unit tests". d'oh! (Of course, it was their problem - their unit tests hadn't, and couldn't, have covered everything - a code review probably would have caught it though).
I hear a bunch of folks are launching runs for position of Governor in California. You appear to be very well, indeed over, qualified for the position. If you're a resident, you should run.
But remember, the pay sucks, everyone blames you for everything, the electorate is very fickle and will vote for/against propositions seemingly randomly, and the legislators are mostly [insert term here that is even too vulgar for/.].
It might work. Google could send a letter to Murdoch (and perhaps issue a press release) explaining something like:
It has come to our attention that Mr. Murdoch believes Google doesn't have a right to show excerpts from content his companies distribute.
Although we are confident in our legal analysis which concludes that our usage falls under "fair use" and that Mr. Murdoch's position is without legal merit, the outcome of legal challenges in evolving frontiers is difficult to predict. We have a fiduciary responsibility to our shareholders to take action to mitigate damages, esp. when such actions would not measurably affect our profits. Counsel has advised us to exclude the questioned material from our indexes until this matter is resolved.
Therefore, effectively immediately we have purged all such information from our indexes and will no longer crawl Mr. Murdoch's sites. We request that Mr. Murdoc notify us immediately if he believes we have missed any of his material that he asserts we should not display in search results so we can exclude it as well.
(And, if they are really clever, they will word the message such a way that the first letter of each word in the first sentence sends a somewhat crisper message. Since Google is in California, they can just call the guber to locate a specialist in this art.)
Or, world population growth trends might just continue and the world population will stabilize. Last week's Economist has an
article
that discusses this trend. The article projects that world population will reach 9.2 billion in 2050 and stabilize at that level. Of course, it's just a projection, but that's all anything is for the 2050 time-frame.
Birthrates drop as people become more wealthy and some of the poorest areas of the world have the highest birth rates. The notion that the solution to excessive population growth is to put the "excess" bodies on space ships and send them to live somewhere else is absurd - the cost of the launch alone likely exceeds the total cost of caring for the kid here on earth for the rest of their lives.
The supply of humans, like rabbits, is nearly unlimited - only the resources to provide for them are limited. It might make sense for some reason to send a few prime "breeding pairs" (human and/or non-human) to populate other celestial bodies - for example, if one believes that forms of life on Earth are unique enough in the universe and superior in some way to other life in the universe that it's important for some moral, ethical, or religious reason to preserve and propagate Earthly species even after the Earth is uninhabitable (having been baked to a crisp by our sun for example). But, doing so won't have a measurable impact on Earth's human population.
Hmm... In almost 30 years of systems development at many companies, I've never seen a case where any individual contributor in QA made as much as the top 30% of developers. Perhaps my experience is atypical, but plenty of developers make north of $150K, few QA individual contributors do.
I'm confused... Do you think that the taxpayer should fund hobbies? What, specifically, do you disagree with in my comment?
If a degree has no visible economic value to the country, should the taxpayers fund them? Should, in the USA, that decision be left to Scientology. Bush 43, or Obama (or whoever's moral compass is in vogue that week)?
As I made clear, if the reason McFortner can't pay back his/her loans is that s/he became disabled after entering school, then address that by forgiving/mitigating the impact of the loans.
If the reason was that s/he was getting a degree in something that only a minority in a first world economy thinks is worth money, then any funding (if any) for her/his education should come out of Arts or other similar highly discretionary humanitarian funding - not something that takes away the ability of a potential physicist, chemist, doctor, or engineer to contribute to society.
Do India and China spend as large a percentage on non-economic degrees as the U.S. does? If not, then that might explain why it's hard to find an American who can write code but easy to find highly skilled developers from India or China who can and do.
And I still thought that/.'s were logical people who could do cost/benefit analysis -- but I should now realize that the Mac vs. Windows vs. Linux wars should have shown me that this is not the case.
I'm sorry that you ended up on disability and (assuming the disability was legit and was not expected at the time of granting the loan) believe that becoming (at least permanently) disabled should give some relief on Fed insured student loans.
However, I don't think we should be insuring loans for degrees or people who have little chance of getting jobs in the areas of their degrees. One way to deal with this (as has been done at the vocational school level in some states at least) is to make the EDUCATIONAL INSTITUTION somewhat responsible for students who fail to make $ after graduation. This will give the schools some motivation to (1) Screen Students Better (2) Teach Practical Skills (3) Improve the Quality of Education.
There are few jobs for experts in Old English, or Ancient Religions, or even Philosophy, outside academia (and, most of those jobs within Academia exist just because someone needs to teach the next generation of idealistic and well educated Walmart Greeters). We have a word for these things: "Hobby" (not "Practical Profession or Trade") and as a taxpayer, I have little desire to fund Hobbies.
Unfortunately, the availability of Federally insured loans and various government "scholarship" programs based on need more than merit allows universities to drive the price of an education up. When you increase the demand (via loan programs and entitlements), you increase price without increasing service.
Welcome to Higher Education.
Maybe this situation of saddling students with big loans is good -- surely some of them will figure out it's a zero-sum game where they will have to pay off not only their own loan but an entitlement grant for an unqualified person to pursue higher education until they drop out. I believe it's called the College of Hard Knocks (a.k.a. Life)
In my mind, a mandate that a business or individual provide a service to someone below cost is a form of tax (which is paid by the business' other customers or the individual).
It seems to make little practical difference if the government extracts money from the business or individual and then uses those proceeds to subsidize the "below cost" service or if the government just demands that the business or individual provides the service below cost.
I think you at least have to add the people who are already eligible for Medicaid (and similar) programs but haven't bothered to sign up to get down to 10-15 million people.
I'd be a little more sympathetic to this notion if Federal tax brackets were indexed to the cost of living in the area the taxpayer resides.
Why should a family living a pretty lower class lifestyle on $70K a year in the San Francisco pay higher taxes or higher broadband rates to get good broadband to a family living like a king on the same $70K in a nice 2500 square foot house on twenty acres in some remote burg?
People are free to move wherever they want in the US and there are consequences -- higher standard of living for the same money in a remote burg may mean one has to pay more for some services. What next, people in SF should pay for snow removal in Montana?
you can just repeat the process every few months - new scars, new pattern.
(ProTip: alternate left and right hands when rebranding - it's much easier to go about your daily life with only one hand being disabled at once - just after committing really bad crimes, might need to do both at the same time though.)
Slashdot Mirror
So, all fishing (a.k.a. priate) boat crew members put on life jackets or, if they have it, get in life raft.
Abandon (fishing, pirate) boat.
Merchant/Commercial ship spends two hours picking them up ONE-BY-ONE.
Searching each.
Restraining each.
All are just fisherman, fed and sheltered and dropped at next port of call.
Not all "fisherman"? All pirate (oops., fishing) boat members put on trial and, if guilty, executed within 30 days of conviction.
Not hard.
Hundreds of thousands die prematurely every day in the world due to malnutrition and disease. Why not deal decisively with pirates who increase that number by increasing shipping costs for staple supplies like wheat and other grains/grasses?
But, what if AGW isn't the primary cause of such change. If so, accepting AGW as "the answer" due to faulty science motivated by political (and funding) agendas could be a big mistake. At best we would waste resources and not significantly affect the outcome; at worst, we would fail to address a problem that, had we had not erroneously attributed it to AGW, we could have mitigated or stopped.
At this point, I think we should be devoting most grant money to disproving AGW as a significant factor in climate variations. If AGW theories and predictions stand up under the resulting scientific scrutiny we can have more confidence that AGW is, indeed, a serious problem. Unfortunately, with most funding going to further prove the premise and much research being done by "scientists" who have already made up their minds (Jones et al), it's not surprising that most research supports the premise.
We need to be careful that we are not deluded by Cargo Cult Science as described by Feynman. Unfortunately, the extent of AGW isn't precisely measurable or provable and never will be - the system is too complex and there are too many external variables which can't be controlled. This makes it all the more important that the scientific debate be open and honest.
Apple has something of an advantage over notebook vendors such as HP, Acer, and Toshiba because Apple doesn't sell really low end notebooks.
CR's notebook reliability numbers are somewhat meaningless for comparing Apple to vendors which manufacture sub $500 notebooks. The low end notebooks have to cut corners which will reduce reliability (more plastic, less metal etc.). CR should really compare reliability of notebooks by vendor based on price (perhaps in two classes - $750 for base product).
I think the confusion is yours - the failures were in production not unit test. The point is, the Therac-25 seems to have passed the unit tests - but that clearly wasn't enough to determine that the product worked correctly.
Failing to include good comments in a complex regex in production code, esp. if the regex needs to deal with oddities in the input, should be a capital offense.
I agree for the most part - I meant to push back on the "if it passes the Unit Tests, then how is it really bad?" part - passing unit tests is almost never sufficient, although it's almost always necessary.
/tmp and intended to never be run again. (I actually assumed the poster was going for "funny" until I read the number of responses to his comment!)
Also, it doesn't sound like the "quadrant coder" was writing some script to be used only once or twice -- he bothered to put comments in it which suggests to me it wasn't created in
Also, no real life complex system is ever "perfect". That's why we sometimes try to reduce the risk of catastrophic failure in life critical systems by implementing independent systems whose entire job is to detect a failure in another part and enforce safety constraints. But this solution still isn't 'perfect' - just much less likely to result in catastrophic failure.
How true... I worked at a place where one group used agile methods (sort of). It was amusing to watch them argue to a group that their software interfaced with that "it can't be our problem, it passed unit tests". d'oh! (Of course, it was their problem - their unit tests hadn't, and couldn't, have covered everything - a code review probably would have caught it though).
It's good to see a poster who worked on the Therac-25 - we don't get many of those here anymore.
IIRC, NonStop applications had to be very involved in FT - initiating checkpoint type operations to the "backup" etc. Absolutely nothing like Remus.
And, IIRC, NonStop SQL wasn't one of those applications - that amused me.
Hmm... Before 1979, market share for RDBMS was TINY. It really didn't begin to "serve us well" until the mid 80's.
I hear a bunch of folks are launching runs for position of Governor in California. You appear to be very well, indeed over, qualified for the position. If you're a resident, you should run.
/.].
But remember, the pay sucks, everyone blames you for everything, the electorate is very fickle and will vote for/against propositions seemingly randomly, and the legislators are mostly [insert term here that is even too vulgar for
(And, if they are really clever, they will word the message such a way that the first letter of each word in the first sentence sends a somewhat crisper message. Since Google is in California, they can just call the guber to locate a specialist in this art.)
Or, world population growth trends might just continue and the world population will stabilize. Last week's Economist has an article that discusses this trend. The article projects that world population will reach 9.2 billion in 2050 and stabilize at that level. Of course, it's just a projection, but that's all anything is for the 2050 time-frame.
Birthrates drop as people become more wealthy and some of the poorest areas of the world have the highest birth rates. The notion that the solution to excessive population growth is to put the "excess" bodies on space ships and send them to live somewhere else is absurd - the cost of the launch alone likely exceeds the total cost of caring for the kid here on earth for the rest of their lives.
The supply of humans, like rabbits, is nearly unlimited - only the resources to provide for them are limited. It might make sense for some reason to send a few prime "breeding pairs" (human and/or non-human) to populate other celestial bodies - for example, if one believes that forms of life on Earth are unique enough in the universe and superior in some way to other life in the universe that it's important for some moral, ethical, or religious reason to preserve and propagate Earthly species even after the Earth is uninhabitable (having been baked to a crisp by our sun for example). But, doing so won't have a measurable impact on Earth's human population.
...and the repairs
Overreaction... Four weeks, I plaeyd wiht over tow tbspoons taht my dad broght home frme work when i was a kid. It didnt hurt me at al.
Hmm... In almost 30 years of systems development at many companies, I've never seen a case where any individual contributor in QA made as much as the top 30% of developers. Perhaps my experience is atypical, but plenty of developers make north of $150K, few QA individual contributors do.
I'm confused... Do you think that the taxpayer should fund hobbies? What, specifically, do you disagree with in my comment?
/.'s were logical people who could do cost/benefit analysis -- but I should now realize that the Mac vs. Windows vs. Linux wars should have shown me that this is not the case.
If a degree has no visible economic value to the country, should the taxpayers fund them? Should, in the USA, that decision be left to Scientology. Bush 43, or Obama (or whoever's moral compass is in vogue that week)?
As I made clear, if the reason McFortner can't pay back his/her loans is that s/he became disabled after entering school, then address that by forgiving/mitigating the impact of the loans.
If the reason was that s/he was getting a degree in something that only a minority in a first world economy thinks is worth money, then any funding (if any) for her/his education should come out of Arts or other similar highly discretionary humanitarian funding - not something that takes away the ability of a potential physicist, chemist, doctor, or engineer to contribute to society.
Do India and China spend as large a percentage on non-economic degrees as the U.S. does? If not, then that might explain why it's hard to find an American who can write code but easy to find highly skilled developers from India or China who can and do.
And I still thought that
I'm sorry that you ended up on disability and (assuming the disability was legit and was not expected at the time of granting the loan) believe that becoming (at least permanently) disabled should give some relief on Fed insured student loans.
However, I don't think we should be insuring loans for degrees or people who have little chance of getting jobs in the areas of their degrees. One way to deal with this (as has been done at the vocational school level in some states at least) is to make the EDUCATIONAL INSTITUTION somewhat responsible for students who fail to make $ after graduation. This will give the schools some motivation to (1) Screen Students Better (2) Teach Practical Skills (3) Improve the Quality of Education.
There are few jobs for experts in Old English, or Ancient Religions, or even Philosophy, outside academia (and, most of those jobs within Academia exist just because someone needs to teach the next generation of idealistic and well educated Walmart Greeters). We have a word for these things: "Hobby" (not "Practical Profession or Trade") and as a taxpayer, I have little desire to fund Hobbies.
Unfortunately, the availability of Federally insured loans and various government "scholarship" programs based on need more than merit allows universities to drive the price of an education up. When you increase the demand (via loan programs and entitlements), you increase price without increasing service.
Welcome to Higher Education.
Maybe this situation of saddling students with big loans is good -- surely some of them will figure out it's a zero-sum game where they will have to pay off not only their own loan but an entitlement grant for an unqualified person to pursue higher education until they drop out. I believe it's called the College of Hard Knocks (a.k.a. Life)
NO!
Beer should be free - besides, that would reduce the cost of getting an education.
(And NOT Bud Light.)
In my mind, a mandate that a business or individual provide a service to someone below cost is a form of tax (which is paid by the business' other customers or the individual).
It seems to make little practical difference if the government extracts money from the business or individual and then uses those proceeds to subsidize the "below cost" service or if the government just demands that the business or individual provides the service below cost.
I think you at least have to add the people who are already eligible for Medicaid (and similar) programs but haven't bothered to sign up to get down to 10-15 million people.
From: Sir
Dear Finland Telecom Customer Service Manager,
No problem. No rush.
Actually, I'm going to be pretty busy over the cumming years so may not be able to let your techs in to do the installation for quite sometime.
I'll call you when I'm available to provide access to the installers.
Thank You
p.s. While it's on my mind, do I just call customer service for replacement if the strippers wear out or break?
I'd be a little more sympathetic to this notion if Federal tax brackets were indexed to the cost of living in the area the taxpayer resides.
Why should a family living a pretty lower class lifestyle on $70K a year in the San Francisco pay higher taxes or higher broadband rates to get good broadband to a family living like a king on the same $70K in a nice 2500 square foot house on twenty acres in some remote burg?
People are free to move wherever they want in the US and there are consequences -- higher standard of living for the same money in a remote burg may mean one has to pay more for some services. What next, people in SF should pay for snow removal in Montana?
you can just repeat the process every few months - new scars, new pattern.
(ProTip: alternate left and right hands when rebranding - it's much easier to go about your daily life with only one hand being disabled at once - just after committing really bad crimes, might need to do both at the same time though.)