I think credit card companies charging 2 to 3% fees from the retailers for credit card transactions is fair, I think.
I think 2-3% is extremely high for what they are doing. The credit card processors have very low marginal costs and typically operate at very large scales. Without thinking too hard about this, I agree that it would be reasonable to model the fees as a function of short-term interest rates (for the loan) plus a markup. But remember, we're talking about 2-3% out of every dollar spent: that's a slice of the total revenue, not just profits; so that eats directly into the profit margin of the seller. Risk-wise, it is diversified across more or less the entire population of consumers and market sectors, so really the biggest problem is when consumer spending goes down overall, which is when everyone is feeling the hurt. Interest-wise, it's a percentage of the total value spent regardless of the loan period. Assuming everyone pays their bills on time (no late fees: worst-case scenario for the credit card company), that means 2-3% for a month or two, which makes the annualized rate more like 12-18%. That sounds absurdly high, so I would welcome any corrections. I'm just saying that 2-3% sounds like a lot.
I'd say most retailers do very little tracking via the bankcard numbers since they would be exposed to PCI issues if they stored the data and it was hacked.
It's pretty easy to salt + hash every card number and then just track the hashes. In fact, because this type of tracking is imprecise to begin with, there's probably some trade-offs that can be made which improve security (make reversal or brute-forcing harder) at the low low cost of increased hash collisions.
Some of us were listening long before Portal came along.
...but most of us were introduced to him by Portal. That certainly makes the song more noteworthy, though arguing about importance is probably pissing into the wind. But in the context of the GP's statement (what JoCo is known for), Still Alive takes the cake.
Write the worst code you possibly can.
<useful context snipped>
Used to be quite easy in the past. With C, you just needed let your pointer variables uninitialized or free twice the same memory.
What do you mean? Do you mean that C is a dead language, or that these problems no longer occur in C languages? Or maybe you mean that the types of developers who frequently had uninitialized-variable or double-free bugs no longer use C? The GP's list is all about bad data-structure design, and that can be done equally well in all languages. Maybe you're joking with the C comment, but I don't see the humor in inaccuracy—and I think it detracts from an otherwise excellent list of ways to totally screw with your maintenance programmers.
In this case, the overriding opinion is that the acts of the prosecutors are responsible for the death of Swartz.
However, in at least two other cases... the overriding opinion... was that the person or people accused of bullying were not responsible for the deaths of the victims, as suicide victims usually have underlying issues.
As I see it, the difference is that the prosecutor is in a position of privilege granted by public mandate. Public servants acting in their "work" capacity should absolutely be held to a higher standard than individuals in their personal lives. This is consistent with Slashdot's group desire for extreme personal privacy, but continuous videotaping of police officers. Or with the prevailing cultural opinion here that sex with your direct report abusive and potentially less-than-consensual, but consensual sex with another co-worker is just fine.
Sorry, upon re-reading my own statement I realize that I failed to include a substantial assumption. Let me amend:
Some mechanism should exist which allows laws to be removed from the books without requiring them to be tested in court or an about-face from the legislature.
It is my opinion that the legislative process is currently skewed towards adding more laws than are removed. I think this is a natural problem: there simply isn't any incentive to remove laws, because the legislative branch wants to look "tough on crime" and because they know that out-dated laws (e.g. state sodomy laws) can simply be ignored by the executive branch. To counterbalance this, I think it should be easier for laws to be removed from the books on the judicial side. The existing powers in the judicial side are presently unpopular with TPTB: Jury nullification has the potential to land you in contempt of court. Grand juries seem to be rubber-stamps now for whatever reason. Elected judges in Illinois have a very low rate of recall: in Cook County, even skipping out on work to sunbathe won't get you kicked out. Cook County also just re-elected—by a landslide—a prosecutor who has a reputation for her "Law & Order" approach to crime-fighting, violent or not.
Ugh. The problem is that on the judicial side, all of the powers that you listed lie entirely with the masses, and so the only way to change is through massive advertising campaigns. Civil-rights organizations can hire lawyers to work within the court system, but they can't make people care.
Prosecutors offer a 6 month plea deal instead of the maximum sentence of 35 years.
Yeah... and you don't see that as problematic at all? When I see "6 month plea bargain... or up to 35 years in prison" I see the subtext of coercion by the entire judicial system. Because, no matter how slim the chance, it's still possible that he'll get nailed in the courtroom and end up with the maximum sentence. If 6 months was the reasonable number offered for "haggling", then nearly two orders of magnitude more is surely unreasonable.
The violent crime rate in the UK is twice what it is in the US
I don't have a good citation, but when one of the recent "GUNS!" articles came around it was mentioned that UK has lower standards for counting violent crime than the US. Looking at The Gospel, the only clear difference I see is that the UK counts "all sexual offences" as violent crimes, but the US only counts "rape". And the US reports are filtered by victim age, but I don't know about the UK. And the US has two separate statistics, both self-reported (one to the police, one to a survey). So, I'm not sure how much I would trust a comparison of the raw violent crime rates, as each nation has its own incentive to "cook" those statistics. It does look like violent crime is going down in both the US and the UK, which means that the lower incarceration rate is not necessarily a bad trade-off.
Or, had he accepted a plea bargin, 6 months in prison.
If what you say is true, then the prosecution was asking for a 70-fold increase in prison time in exchange for the constitutional luxury of a trial by jury. This is a solid indication that 35 years should have qualified as "cruel or unusual". IOW, your facts seem to support the popular narrative as well: Schwartz's blood is on the hands of the prosecutor.
And which legal system do you think works better and is more responsive to citizens?
Well, since the legal system isn't picked off of a menu, how about we just sound off some changes which would "work better and [be] more responsive to citizens"?
Some mechanism should exist which allows laws to be removed from the books without requiring them to be tested in court. For example, an organization like the EFF or ACLU should be able to raise the unconstitutionality of a law without waiting for some poor schmuck to be putting up a defense to it. This would allow for bad laws to be repealed more efficiently, quickly, and with less risk.
Laws which can be demonstrated to be selectively enforced, or which have had no enforcement within a pre-ordained time frame, should be pulled off of the books. Such laws are open to abuse because they can be used as political weapons in a courtroom, even though the selective enforcement implies that they are not valued by society at large.
Ultimately, if they're human enough to demand rights they'll get them
It is fitting to see this on the weekend before Martin Luther King Day. No, they won't just "get them"... everything worthwhile is fought for. A Neanderthal born today will not see "human(ish) rights" applied in his lifetime.
Virginia, huh? Hmm... What else could be in Virginia, I wonder...?
NASDAQ's backup data center. Huh, maybe that means that Ashburn, VA is a reasonably good location for data centers. Spooks can install taps for local snooping in any data center, but getting connectivity to the rest of the world is expensive.
I can't speak to the forcing of huge loans. For that, you can probably just look at anything Bain Capital has touched. I did look a little into KB Toys and Border's (book store), both of which were shuttered after being more-or-less purchased by private equity firms. In both cases, the company was turning a profit, but the ROI wasn't sufficient for the private equity firm (so-called "vulture capitalists", another term to search for). Being the leading owner but without much stake compared to their net value, it made the most sense to kill the entire company and liquidate the assets rather than liquidate their ownership of the company (sell *most* of the stocks for a company, see its value plummet as the #1 owner has publicly given up). It's a sorry state, and IMHO is one of the fundamental downsides of our current system of public ownership.
how's you knowledge about creating and maintaining fish ponds or basic elevator maintenance?
Fish pond maintenance is probably one step further than necessary in order to make a better-than-average decision on fish purchases. For the environmentally-conscious, it would be useful just knowing (1) which fish are local, (2) which fish are farmed vs. caught wild, and (3) which fish are amenable to sustainable farming/hunting. For the health-conscious, it would probably also be useful knowing (1) toxicity information (e.g. mercury in apex predators) and (2) nutritional information.
I fail to see the economic benefit for elevator maintenance for most people. Owners of larger buildings probably do know a thing or two about elevator maintenance, since they have to be certified and inspected by municipalities.
it seems that your snide over-the-top responses match the "fundamentalism" of your original position:
In todays society we need to specialize and not everyone can learn a little of everything.
It is true that in "today's society" there is more specialization due to the economic benefit of efficiency and the capabilities of automation. However, most specialization occurs in the workplace, and most of us end up using generalist knowledge regularly in the other 1/2 of our waking lives. But that doesn't mean that everyone needs to learn a little of everything. Many people will see benefits by learning a little bit about a handful of topics: home and car owners learning about basic maintenance and repairs to save money, consumers learning about product differences while comparison-shopping, family heads learning about cooking and efficient/healthy menu planning, etc.
By the same token you should also know a little about raising pigs as you won't be held a****** farmer, or mechanic
Well yeah, that does make sense. If I roughly know local growing seasons, I can rule out some produce-vending bullshitters. If I know that "kobe beef" is not a USDA-protected term, I can rule out bullshit on a restaurant menu. If I roughly know the material properties of copper, I can rule out stupid copper kitchenware which has nothing to do with heat transfer. The list goes on... the average person has much incentive to develop generalist knowledge to avoid getting screwed over. In fact, the "invisible hand" of the market assumes perfect information availability as a mechanism of consumer-driven price correction. So yeah, you should strive to learn a little bit about everything that you interact with, even if you make a living as a specialist. Because there are plenty of specialists out there whose specialties are identifying and opening the wallets of the clueless.
OTOH, this straw man is ridiculous:
and lets not forget those a****** astrophysicists, clearly everybody should be able to calculate the amount of redshift from a distant star
Yeah, because getting hoodwinked by astrophysicists is something many people fight on a regular basis.
objectively describe what is wrong using with references to either standard or internal best practices or conventions
are you implying that those criteria are objective?
I think the GP's "objectively" is referring to the analysis of code given references. Objectively describe which conventions have been violated, with the implication that the violations make the code bad.
Regarding the "best practices or conventions", any such document really needs to have thorough rationale and examples. Many conventions are very subtle or specific at first glance (or pointless and badong, in the case of the poorer conventions used). Every developer who didn't write that document is going to have a mental argument with most of the bullet points on such a document, and hopefully the rationale and examples can win back some of those arguments. Just like in math class, the authors need to show their work to get credit.
one could even pretty much do away with the shell's redirection operators, with the pipe operator being the only one that's fundamentally necessary (if you really want to be ideologic about this).
You can have my redirect operators when you pry them from my cold, dead fingers. Many redirects can occur within a single command, which allows for some interesting shorthand. Eliminating redirects forces many interactive one-liners to be written as scripts with temp files, which is a PITA for exploration.
Perform diffs on subsets of data
vimdiff <(zgrep 'ugly_pattern' file1.gz) <(zgrep 'ugly_pattern' file2.gz)
Fork off output for debugging while still generating some processed result
lsof -n -p $(pgrep -d, -P2) | tee >(grep ssh > ssh-sockets.txt) | more | stuff
Selectively enable interactive verbose output from some filters
ypcat passwd | cut -d: -f1
| lookup_user_stuff_things --verbose 2>(egrep -i 'error|ldap|nis')
| grep -i admin | awk '{ print $3, $2; }' | sort -u This one is especially useful when using in-house software which has crappy debug/verbose logging.
But you don't have to statically link anyway. You can just copy the.so's too.
Do you even need to do that? As I understand it, the only requirement is that the user be able to replace the functionality. Wouldn't it be sufficient to provide a single binary with all LGPL functionality provided as externally-visible PIC symbols (e.g. memcpy@plt)? It is still a single executable so there's no futzing with LD_* for default users, and power users still have the freedom to override functionality with LD_PRELOAD.
Slashdot Mirror
I think 2-3% is extremely high for what they are doing. The credit card processors have very low marginal costs and typically operate at very large scales. Without thinking too hard about this, I agree that it would be reasonable to model the fees as a function of short-term interest rates (for the loan) plus a markup. But remember, we're talking about 2-3% out of every dollar spent: that's a slice of the total revenue, not just profits; so that eats directly into the profit margin of the seller. Risk-wise, it is diversified across more or less the entire population of consumers and market sectors, so really the biggest problem is when consumer spending goes down overall, which is when everyone is feeling the hurt. Interest-wise, it's a percentage of the total value spent regardless of the loan period. Assuming everyone pays their bills on time (no late fees: worst-case scenario for the credit card company), that means 2-3% for a month or two, which makes the annualized rate more like 12-18%. That sounds absurdly high, so I would welcome any corrections. I'm just saying that 2-3% sounds like a lot.
It's pretty easy to salt + hash every card number and then just track the hashes. In fact, because this type of tracking is imprecise to begin with, there's probably some trade-offs that can be made which improve security (make reversal or brute-forcing harder) at the low low cost of increased hash collisions.
Whoa, that's wrong. I have been using 64-bit Opera for months.
Seen in the warehouse break room:
I tried that, but the rules were too complex!
What do you mean? Do you mean that C is a dead language, or that these problems no longer occur in C languages? Or maybe you mean that the types of developers who frequently had uninitialized-variable or double-free bugs no longer use C? The GP's list is all about bad data-structure design, and that can be done equally well in all languages. Maybe you're joking with the C comment, but I don't see the humor in inaccuracy—and I think it detracts from an otherwise excellent list of ways to totally screw with your maintenance programmers.
As I see it, the difference is that the prosecutor is in a position of privilege granted by public mandate. Public servants acting in their "work" capacity should absolutely be held to a higher standard than individuals in their personal lives. This is consistent with Slashdot's group desire for extreme personal privacy, but continuous videotaping of police officers. Or with the prevailing cultural opinion here that sex with your direct report abusive and potentially less-than-consensual, but consensual sex with another co-worker is just fine.
Sorry, upon re-reading my own statement I realize that I failed to include a substantial assumption. Let me amend:
It is my opinion that the legislative process is currently skewed towards adding more laws than are removed. I think this is a natural problem: there simply isn't any incentive to remove laws, because the legislative branch wants to look "tough on crime" and because they know that out-dated laws (e.g. state sodomy laws) can simply be ignored by the executive branch. To counterbalance this, I think it should be easier for laws to be removed from the books on the judicial side. The existing powers in the judicial side are presently unpopular with TPTB: Jury nullification has the potential to land you in contempt of court. Grand juries seem to be rubber-stamps now for whatever reason. Elected judges in Illinois have a very low rate of recall: in Cook County, even skipping out on work to sunbathe won't get you kicked out. Cook County also just re-elected—by a landslide—a prosecutor who has a reputation for her "Law & Order" approach to crime-fighting, violent or not.
Ugh. The problem is that on the judicial side, all of the powers that you listed lie entirely with the masses, and so the only way to change is through massive advertising campaigns. Civil-rights organizations can hire lawyers to work within the court system, but they can't make people care.
Yeah... and you don't see that as problematic at all? When I see "6 month plea bargain... or up to 35 years in prison" I see the subtext of coercion by the entire judicial system. Because, no matter how slim the chance, it's still possible that he'll get nailed in the courtroom and end up with the maximum sentence. If 6 months was the reasonable number offered for "haggling", then nearly two orders of magnitude more is surely unreasonable.
I don't have a good citation, but when one of the recent "GUNS!" articles came around it was mentioned that UK has lower standards for counting violent crime than the US. Looking at The Gospel, the only clear difference I see is that the UK counts "all sexual offences" as violent crimes, but the US only counts "rape". And the US reports are filtered by victim age, but I don't know about the UK. And the US has two separate statistics, both self-reported (one to the police, one to a survey). So, I'm not sure how much I would trust a comparison of the raw violent crime rates, as each nation has its own incentive to "cook" those statistics. It does look like violent crime is going down in both the US and the UK, which means that the lower incarceration rate is not necessarily a bad trade-off.
If what you say is true, then the prosecution was asking for a 70-fold increase in prison time in exchange for the constitutional luxury of a trial by jury. This is a solid indication that 35 years should have qualified as "cruel or unusual". IOW, your facts seem to support the popular narrative as well: Schwartz's blood is on the hands of the prosecutor.
Well, since the legal system isn't picked off of a menu, how about we just sound off some changes which would "work better and [be] more responsive to citizens"?
It is fitting to see this on the weekend before Martin Luther King Day. No, they won't just "get them"... everything worthwhile is fought for. A Neanderthal born today will not see "human(ish) rights" applied in his lifetime.
They are one in the same. Ward Cunningham is a mastermind dispatched from Planet X.
NASDAQ's backup data center. Huh, maybe that means that Ashburn, VA is a reasonably good location for data centers. Spooks can install taps for local snooping in any data center, but getting connectivity to the rest of the world is expensive.
I can't speak to the forcing of huge loans. For that, you can probably just look at anything Bain Capital has touched. I did look a little into KB Toys and Border's (book store), both of which were shuttered after being more-or-less purchased by private equity firms. In both cases, the company was turning a profit, but the ROI wasn't sufficient for the private equity firm (so-called "vulture capitalists", another term to search for). Being the leading owner but without much stake compared to their net value, it made the most sense to kill the entire company and liquidate the assets rather than liquidate their ownership of the company (sell *most* of the stocks for a company, see its value plummet as the #1 owner has publicly given up). It's a sorry state, and IMHO is one of the fundamental downsides of our current system of public ownership.
Fish pond maintenance is probably one step further than necessary in order to make a better-than-average decision on fish purchases. For the environmentally-conscious, it would be useful just knowing (1) which fish are local, (2) which fish are farmed vs. caught wild, and (3) which fish are amenable to sustainable farming/hunting. For the health-conscious, it would probably also be useful knowing (1) toxicity information (e.g. mercury in apex predators) and (2) nutritional information.
I fail to see the economic benefit for elevator maintenance for most people. Owners of larger buildings probably do know a thing or two about elevator maintenance, since they have to be certified and inspected by municipalities.
it seems that your snide over-the-top responses match the "fundamentalism" of your original position:
It is true that in "today's society" there is more specialization due to the economic benefit of efficiency and the capabilities of automation. However, most specialization occurs in the workplace, and most of us end up using generalist knowledge regularly in the other 1/2 of our waking lives. But that doesn't mean that everyone needs to learn a little of everything. Many people will see benefits by learning a little bit about a handful of topics: home and car owners learning about basic maintenance and repairs to save money, consumers learning about product differences while comparison-shopping, family heads learning about cooking and efficient/healthy menu planning, etc.
Well yeah, that does make sense. If I roughly know local growing seasons, I can rule out some produce-vending bullshitters. If I know that "kobe beef" is not a USDA-protected term, I can rule out bullshit on a restaurant menu. If I roughly know the material properties of copper, I can rule out stupid copper kitchenware which has nothing to do with heat transfer. The list goes on... the average person has much incentive to develop generalist knowledge to avoid getting screwed over. In fact, the "invisible hand" of the market assumes perfect information availability as a mechanism of consumer-driven price correction. So yeah, you should strive to learn a little bit about everything that you interact with, even if you make a living as a specialist. Because there are plenty of specialists out there whose specialties are identifying and opening the wallets of the clueless.
OTOH, this straw man is ridiculous:
Yeah, because getting hoodwinked by astrophysicists is something many people fight on a regular basis.
It's like a babby without the pragnency.
I think the GP's "objectively" is referring to the analysis of code given references. Objectively describe which conventions have been violated, with the implication that the violations make the code bad.
Regarding the "best practices or conventions", any such document really needs to have thorough rationale and examples. Many conventions are very subtle or specific at first glance (or pointless and badong, in the case of the poorer conventions used). Every developer who didn't write that document is going to have a mental argument with most of the bullet points on such a document, and hopefully the rationale and examples can win back some of those arguments. Just like in math class, the authors need to show their work to get credit.
You can have my redirect operators when you pry them from my cold, dead fingers. Many redirects can occur within a single command, which allows for some interesting shorthand. Eliminating redirects forces many interactive one-liners to be written as scripts with temp files, which is a PITA for exploration.
Perform diffs on subsets of data vimdiff <(zgrep 'ugly_pattern' file1.gz) <(zgrep 'ugly_pattern' file2.gz) Fork off output for debugging while still generating some processed result lsof -n -p $(pgrep -d, -P2) | tee >(grep ssh > ssh-sockets.txt) | more | stuff Selectively enable interactive verbose output from some filters ypcat passwd | cut -d: -f1| lookup_user_stuff_things --verbose 2>(egrep -i 'error|ldap|nis')
| grep -i admin | awk '{ print $3, $2; }' | sort -u
This one is especially useful when using in-house software which has crappy debug/verbose logging.
Do you even need to do that? As I understand it, the only requirement is that the user be able to replace the functionality. Wouldn't it be sufficient to provide a single binary with all LGPL functionality provided as externally-visible PIC symbols (e.g. memcpy@plt)? It is still a single executable so there's no futzing with LD_* for default users, and power users still have the freedom to override functionality with LD_PRELOAD.
Uh, uh!!!! Fart, drop the control module into the water, and then blow up!
He's right. See fopencookie, which provides a way to abstract an arbitrary stream as a FILE*.