What if for half the cost of a single person, you could instead have 500 martian probes distributed across the planet? Besides, NASA scientists have made statements contradicting what you just said. Robots are slower, but can operate nonstop indefinitely where humans need long rest periods.
Humans advocate humans; roboticists advocate robots. Film at 11.
You'd need 1,500 people on Earth working 8 hour shifts, for a much longer period of time, analyzing visual data, and saying "Hmmm... that rock that the probe passed 4 hours of analysis, and 24 minutes ago looked interesting; let's send a command now to the probe, which it will receive in 24 minutes, to go back and look at it. The elapsed time to deciding if it was 'just a shadow' should only be another time to compose command + 24 minutes + 5 hours travel + 24 minutes of transmitting back new pictures of the rock. Then we can decide what to do next, which will only take decision time + compose message time + 24 minutes transmit time + ??? doing time + 24 minutes transmitting back the result."
Low Specific Impulse requirements make the Martian moons a good place for permanent basing, but not so good for staging a Mars mission, and not so great an idea if we are just going to go there to go there, rather than go there to stay there.
Mostly, they are a great staging area for asteroid missions, given that the escape velocity is generally achievable with spring-loaded catapults and electric winches, rather than the more expensive and hard to construct mass drivers that you'd have to build to get mass off Earth's moon.
Can't or won't? I would have thought that it would be possible to create a habitat in either that would require nothing incoming.
So far we cannot. We've tried several times and haven't cracked the problem yet.
We can and we did. It was euphemistically called "The U.S. Government Relocation Facility", but it's code name was "Project Greek Island", and it was capable of sustaining a fairly large population and support staff for 30 years, in the event of a nuclear war.
"The Raven Rock Mountain Complex" was built as a similar "relocation facility" intended for the Pentagon.
"The Cheyenne Mountain Complex" was another facility, for SAC/NORAD.
"The Mount Weather Emergency Operations Center" was intended for use by FEMA; it's sometimes called "The High Point Special Facility".
There are a bunch of U.S. COG facilities (Continuation Of Government) besides that, and most major countries have their own equivalents.
Generally most of them have nuclear power plants, of the type used on U.S. nuclear submarines. Several of the facilities have more than one of them.
We really cannot build a "sustainable habitat" anywhere, "biosphere 2" has the longest record of about 2 years, the experiment ended when they ran out of oxygen, food, and patience with each other.
Biosphere2 failed for many reasons, including the fact that they didn't take into account the fact that curing concrete sequestered CO2 at a high enough rate that the plants were unable to survive, and that in not surviving, they failed to produce O2 from the CO2 they were getting, which made it harder on the animals (including humans).
In the second series of experiments, it was also very common for the door alarms to be bypassed, and they would order out for pizza. The pizza delivery guy who used to go out there lived in the apartment building next to mine at 2000E Roger Road in Tucson. So it was seriously not a sealed environment, and their ambitions exceeded their technical capability, since the thing was more or less funded with the eventual goal of it being a tourist attraction anyway.
I personally could live in a small group, or even on my own, for quite a long period of time, and there are a lot of other people who could do the same; not everyone has to sleep in a "puppy pile" to maintain their sanity or their patience with other people of the same bent.
P.S.: In case you care: the pizza guy's apartment was in the building on the left in the picture located here: https://media.apts247.info/63/...
Mod it to -1 if you like, people will never *be* on Mars. The closest they'll get is to see it through a visor or a monitor. And if that's the case, a monitor on a different planet is more convenient.
You are obviously not a geologist. A person, even in a suit, and wielding a rock hammer, and equipped with a rather small lab can do more geology in one day than all of the Mars probes ever sent have done, combined.
Not to mention the fricking communications latency of using RPVs, or depending on the cleverness of remotely targeted semi-autonomous robots.
No, they only mentioned the Linux Foundation code, the 115M lines. And they did it to emphasize what they perceived as the value of the code. My Point here was that this code neeeds to be broken down into different categories of complexity. I fear they only used a "Default category"
Complexity distinctions are rather specious; let me explain.
Yes, a coding error in an OS can crash the whole machine. But at the time the lines of code stats were written, most computers were not running a protected mode OS, and therefore, a coding error in *any* program could potentially crash the entire machine.
By that token, they've overestimated the value, by treating everything as if it could crash the machine. Unless we relax the criteria to "could crash the machine and/or result in a security vulnerability", in which case we are back to all code having the same level of impact on the machine as all other code. Security was pretty much not an issue back then, because there were no such things as "bad actors who did not just crash their individual machine, or, on a multiuser system, had their account suspended until they talked to someone with a scowl on their face, and received a stern warning".
Also, I will claim that programmers who learned to code when anything they did could crash a machine (and yes, I've crashed mainframes dozens of times with user space code, back in the day, taking everyone's programs with mine -- and had mine shot down the same way) learned to be a hell of a lot more careful coders than the generation who grew up with "if it crashes, it'll only kill itself, and we can just look at the core dump". In other words, they were better coders because they had to be.
Let's turn that back on you - is Uber violating the law where you live but no action has been taken against them? What does that make where you live by your own (very stupid) definition?
No, Uber, the B2C *ride sharing* negotiation service, which takes a cut from the *contractor* offering to *share* their car for $$$ via the Uber *negotiation service* is/NOT/ breaking the law in San Francisco.
Or San Francisco would be collecting $100 a pop for any Uber logo'ed car with passengers other than the driver, because that the fine for operating a gypsy cab in San Francisco.
Therefore, the city of San Francisco is/tacitly stating/ that Uber is/NOT/ a Taxi service.
And believe me, the traffic and parking enforcement people in San Francisco are greedy bastards; if they felt they could legally collect those fines -- they'd do it. An expired parking meter costs you $76 + fees (which are not cheap): https://www.sfmta.com/sites/de...
Perhaps you should write a book about your ideas how software is written.
When COCOMO was developed we had ideas about productivity in software development like this, per day a developer writes about: 25 - 100 lines of code in user applications 5 - 10 in service software, like specialized editors, TCP/IP stacks etc. < 1 line of code in system software, especially kernels. And this is independent of programming language used
These stats are largely inaccurate for modern coders, who are much more productive than if they were writing their code in IBM BAL. If you have a modern coder writing at this rate (on average), then you should likely fire them, and hire someone who can code, instead. The "on average" is because you should spend 90% of your time planning and 10% of your time coding.
On a project (The Whistle InterJet), I wrote a Fetchmail replacement to work around a number of issues that the author felt were unnecessary to fix, and refused to incorporate as bug fixes. As gravy on the side, it got us out from under the GPL on the code, so we didn't have to carry around local patches.
The major specific issue was not reverse time-ordering of the "Received" timestamp line being used to identify an addressee -- Fetchmail overrode earlier "for" values as it moved forward through the message, which frequently lost information, and allowed faked headers to override a destination, in order to abuse Fetchmail as a remailer for SPAM.
The resulting production code was a little over 22,000 lines of code, and was written in a period of 2 weeks of normal 5 day work weeks. That was 1,100 lines of code a day. And yes, I ended up with tendonitis, and had to wear wrist braces off and on for a while after that, after that much typing.
Anyway, your idea how to burn down the costs is just ridiculous.
And yet... Mac OS X is UNIX certified, and Linux is not.
And yet... I was tech lead on the team of 6 full time people that did it, and, in aggregate, wrote more lines of code for Mac OS X Tiger than the rest of the Core OS Kernel team and the Core OS user space team wrote for that same OS release, combined.
Yes, we cut through an metric assload of red tape in order to do it,, and we had the dictatorial hammer of a $200M lawsuit to pound people on the head to enable us to do it. In other words, in a lot of cases, with the aid of Apple Legal,, and the backing of upper management: we were dictatorial asses. If we had to change an API, and it broke your project as a consequence of you using -Werror: you got to fix your project, because the consequences of us not making the change were $200M out of Apple's wallet, and so our hammer was bigger than yours.
I have a number of very, very good friends who were hit by this hammer; but when you work in a corporate environment, one of the things you are paid for is to put the company's good above your personal sensibilities. The other major thing you are paid for is to get along amicably with people you would never work with, were it not for the fact that you were being paid to work with them.
In other words, unlike in volunteer Open Source, where you bike-shed, bicker, and kingdom-build to the point of process ossification, you have your part, you do what you are paid to do, and the product ships, as the product was intended to ship.
The point is: there are assloads of red tape between doing a mediocre job, and doing great things, and you merely need to enable people to do them.
And I bet my ass and my balls that the original estimate is off by a factor of ten, minimum. Because they likely did not judge the projects into the required "category" mentioned above.
No, they only mentioned the Linux Foundation code, the 115M lines. And they did it to emphasize what they perceived as the value of the code.
The "required" ancillary code doesn't matter, because you could just
Hospital emergency rooms go offline during strikes?
How to recognise a dumb shill - he will use an utterly ridiculous hyperbole.
How do you *get to* the hospital emergency room, when your only transportation is public transportation?
How do you *get to* the hospital emergency room *in time* when the French Taxi drivers go "on strike" by way of intentional traffic jams for everyone, because they are pissed that no one wants to ride in French Taxis, if there is any reasonable alternative, whatsoever?
How to recognize a dumb shill: they answer only your direct argument, out of context, rather than taking the consequences of the situation into account.
P.S.: During the 1979 nurses strike in Montreal, Canada, several hospitals had to close their emergency rooms to incoming patients. The Baton Rouge General Medical Center’s mid city emergency room was closed at 7AM on 31 March of 2015 due to a strike by doctors and nurses.
P.P.S.: Why don't you do a google search before you call something "hyperbole" next time?
A freedom to do something is no protection from social enforcement of the consequences of you exercising that freedom.
Ridiculous. By that argument, everyone has the freedom to do absolutely anything they want—they might just get thrown in jail if it happens to violate a law. This makes a mockery of the very word "freedom".
Nevertheless.
Daniel Ellsberg with conspiracy, espionage and theft of government property for his disclosure of the Pentagon Papers. Ellsberg later got off on a procedural technicality resulting from disclosures by the "Plumbers" group members during the investigation of the Watergate scandal.
If you want similar cases where journalists have been jailed for what they've published, look no further than Josh Wolf, Judith Miller, Jim Taricani, Vanessa Leggett, Timothy Crews, James Risen, Glenn Greenwald, Janine Gibson, Eric Lichtlau, and so on (some were merely threatened with censure; others were actually jailed -- all of them since 2000.
The thing that's currently protecting Glenn Greenwald is the backlash that the government would suffer now that he's won the Pultizer.
There's a big difference between "can lead to negative consequences" and "will lead to negative consequeces", or even "would a reasonable person *expect* it to lead to negative consequences -- which is the actual measure taken by the courts.
Yeah, there should be shield laws. There are some at the state level, but for the most part, journalistic speech is *not* protected from government suppression.
Writing hardware drivers / kernel code is a bit more time consuming than writing a database application. Did you take that into account?
You mean like when I personally wrote nearly a million lines in the Mac OS X kernel, over a period of several years, in order to get it UNIX certified?
Yeah, I took that into account.
Perhaps not enough? I should probably reduce that cost estimate, given that pretty much any dink can write user space code and get it to work, and we are mostly talking about user space code...
It is being fixed where I live and Uber is getting fined and forced to play by the rules or get out so your insult is misplaced. What was the point of your insult anyway? If you are just doing it to try to prove you are better than some stranger on the internet then that is a very pitiful state of affairs.
No, the point was to correctly place "bribery works" as a measure of "shithole-ness" in context relative to the designations "first world vs. second world vs. third world". Being in the "third world" doesn't make some place a "shithole"; bribery working there as a normal part of the process *does*, most assuredly, make some place a "shithole".
They are biased towards gifted students, and biased against students that are not gifted.
We may argue about of the definition of the word "gifted", but the most commonly accepted definition throughout the history of gifted programs has been "of most future economic or strategic benefit to society".
They are not aimed at producing more Pablo Picassos, Walt Whitmanns, or Maya Angelous, they are aimed at producing more John Glenns, Albert Einsteins, Robert Goddards, and so on. They are aimed at providing adults who would be of most benefit to winning the Cold War, and any similar future conflicts which may arise, in order to ensure the continued existence of the larger society.
What constitutes "gifted" is "more naturally inclined to further the goals of society as a whole". In other words, what society values is what we then term as "gifted", when it is evinced at an earlier age.
If you want to change the definition of "gifted", then first change what society values. You can get a rough idea of current valuations by comparing a two axis graph of "economic reward" vs. "job category", and then look at the top earners in each category. Thus, you get that society values functional plumbing (plumbers), but values substantially less correct grammar and spelling (english majors).
[...]the H1-B system is totally broken and is being used to help decimate the American middle class.
Dec.i.mate: kill one in every ten of (a group of soldiers or others) as a punishment for the whole group.
As long as it's only one in ten, I'm kind of OK with this. Also, I'm kind of OK with the idea that such punishment is actually deserved, since it implies 90% "good apples" and 10% "bad apples", which, if you've ever worked a middle class job, is very easy to credit as an underestimation.
My calculation on that comes out closer to $1.3B for a 5 year project to replace all of it.
With a much smaller number of highly dedicated people who are 3X as expensive as the average software engineer in Silicon Valley, I think it'd be possible to drive the number down closer to $790M and 2 years.
The people would need to be dedicated, and the project would need to be driven by (in effect) a dictatorial ass whom everyone has agreed to follow to the ends of the Earth. In other words, a rather strict hierarchy.
Oh yeah... the project would have to be operated Cathedral style, and not use most "agile" techniques.
Freedom of the press is not the same as civil disobedience which is not the same as a company ignoring laws.
Point of order: there are no federal shield laws for journalists in the U.S.. Just because there is freedom of the press written into the bill of rights, does not mean that you can not be held legally accountable for what you print.
The common argument is "yelling 'fire!' in a crowded theater", which is commonly misinterpreted as stating that the yelling itself is illegal; it is not, it is protected by "freedom of speech"; the consequences *may* however be something you can have your ass thrown into jail over.
A freedom to do something is no protection from social enforcement of the consequences of you exercising that freedom. That is pretty much the very definition of "civil disobedience".
Rosa Parks was, in fact, arrested for her act of civil disobedience on 1 December 1955.
Err, why do you think that Uber is superior? Surge pricing during a Tube strike is a real bitch, as is the difficulty in arranging for a guaranteed 5am pick-up for the airport arranged the night before.
Wont work unless the HDMI matrix has EDID management so the computers do not see the displays appearing and disappearing.
You need a good quality one, not the junk from monoprice. About $1100 for a good 4X4 with full edid management.
THIS, x1000... with caveats...
The caveats are:
(1) Many monitors won't negotiate an EDID on an inactive channel (i.e. one not selected as the primary channel at the time the computer attempts to negotiate). This is mostly because the firmware in these monitors is not running a finite state automaton and/or is not multithreaded, and so ignores channels which are not currently selected when negotiating. A *lot* of monitors and television sets being used as monitors, particularly Samsung models, have this problem.
(2) When using left/right or top/bottom negotiation, your KVM needs to present EDID information for the effective aggregate monitor size for the negotiation. And most KVMs, even those with EDID management, can't do that. In other words, the KVM has to present one *or more* monitors to the computer as if they were a single, higher resolution monitor. Right there, you've blown the price of the KVM hardware out of most price ranges. Even then, you may find that the KVM suffers from the same "inactive channel negotiation" problem described as a monitor problem in #1.
(3) Most operating systems fail to implement active resolution renegotiation. This is a problem in two ways:
(3)(a) For the proposed use model, when doing mode switching, i.e. when switching between which PC "owns" a given display, the EDID management that negotiated with the OS over the display resolution is no longer valid... the virtual resolutions may have changed out from under them, based on what the EDID management presented as possible display resolutions, vs. what the monitors that ignored the KVM because it wasn't currently active present as the possible resolutions, vs. the difference between presenting one monitor, or more than one monitor, as the virtual monitor.
(3)(b) The OS renegotiation problem may prevent it from working anyway. Specifically, a lot of "media center" PCs are built on top of Linux, and Linux is really poor at renegotiating an EDID with a device when it goes from the "device not connected" to the device connected" state. This problem tends to find itself exacerbated by the fact that a lot of (stupid) monitors will not switch to a non-default input channel, unless they see that there is active input on the port.
Apple products (*not* Hackint0sh!) tend to have much fewer problems in this regard; Windows, it really depends on the hardware (Toshiba tends to be bad; Sony tends to be better at it), the video driver (which depends on the card in use), and which version of Windows.
P.S. Linux really needs to fix its EDID negotiation on these axises -- but FreeBSD is generally worse, even though both these platforms inherit their EDID negotiation from their windowing systems.
If we really wanted to stop CC fraud, we could almost eliminate it. It's pretty simple, but we've abandoned this in favor of convenience.
100% guaranteed that there would not be any credit card fraud, if there were no credit cards. In other news, people who have their appendix removed don't get appendicitis, which is why it was SOP for the Mercury, Gemini, and Apollo astronauts.
I visited USA last year, and was horrified when my transaction went through when the merchant swiped the mag strip on my Australian chip-and-PIN card, without requiring my PIN or signature.
Signatures are not required on charges of $25 or less, since the store is indemnified against a loss up to that amount, when they fail to collect a signature. Over that amount, it usually requires a signature, and then your signature is floating around as a digital copy for a forger to use.
Chip-and-PIN has reduced brick and mortar fraud, but online fraud is alive and well, as is ATM fraud. Just expect that, as in Europe, the U.S. incidence of card skimming, card trapping, and cash trapping to go through the roof. Expect also to see a rise in chip-and-PIN "preplay" attacks. Expect also that the "foreign currency loophole" will be scammed on contactless cards. Martin Emms of Newcastle University in the UK demonstrated this attack at a conference at the end of last year.
NB: Your NZ card would still be susceptible to the "foreign currency loophole", even if there is wide scale deployment of chip-and-PIN everywhere -- anywhere the transaction isn't taking place in the local currency... no PIN is required.
PS: Doug Johnson of the American Bankers Association has stated that banks prefer that we move to one time token systems, like ApplePay and SamsungPay, in any case.
But the simplest way is to use the stolen card to buy gift cards, use these to purchase merchandise, and fence that via reship or whatever, even eBay.
Once the gift card is used, the link to the original cardholder is lost, AVS is useless. In fact, use out of town mules to use the gift cards, bus them in and out, and even the video of them at the register is useless.
Have you tried buying a gift card with a credit card? There are a few mall locations which allow you to do this, on camera, but if you try it at a grocery store, they'll deny the purchase. They'll let you buy it with a debit card, but not a credit card (I got to watch an insistent lady in front of me in line try very very hard to throw a hissy fit until they let her get away with it; it was like watching a 19 year old trying to buy alcohol).
The more straightforward solution seems to be to simply make CC fraud much more difficult. We have the technology to do so, but seem unwilling to implement it. The new CCs in the US are better in that they have chips, but inexplicably still use signatures rather than PINs.
I personally know of 9 methods of scamming a chip-and-PIN system. The only real value they have are to the credit card company, and the merchant, both of which get to blame you, instead of being blamed themselves, for when one of these scams is run. The intent is clearly to offload the cost of fraud onto the consumer, rather than keeping it the problem of the large financial markets that have some hope of being able to curb the abuses, by virtue of economy of scale approaches to the problem.
The typical Russian Mobster who gets scammed simply has the scammer killed.
We could implement two factor authentication for large purchases and when the CC isn't swiped. Simply send an authorization code to the phone on record for the CC, and require it to approve the purchase.
This requires that I carry a tracking device with me everywhere I go, and it requires that I only make purchases in cellular service areas. This usually sounds like a great idea to Europeans, who are going to be on camera everywhere they go anyway, and whose urban density is such that, everywhere they go, they have cell service. So while that may count as "simply" in Europe, it doesn't in Montana, and it doesn't in large parts of California.
In fact, AT&T is taking down 2G towers because they legally have to grandfather the unlimited data plans which they only offered on 2G service, and in order to force people off those plans, they are pulling down the 2G service, even though it's not like they are replacing it with 5G, and it's not like the same tower electronics and antenna weren't capable of 2G and 3G anyway. They've specifically discontinued the "booster" (actually a CISCO Systems box that is a cellular/VOIP bridge, also known as a microcell) for 2G, even though the CISCO documentation says 2G is supported. They do this by refusing to put the 2G IMEI's into their database, even though it would work perfectly fine, were they to do so.
So relying on cell service for two factor -- even long established cell service, which would allow you to roam to other carriers like T-Mobile or MetroPCS -- is not really an option for about 70% of the U.S..
A similar process could be required for applying for a new CC. Reshipping wouldn't be as much of an issue if CC fraud was made more difficult.
Good luck with that, given the known problems that already exist with chip-and-PIN systems. Credit card fraud is here to stay; there's only damage control after the fact, and that's going to become more difficult for defrauded consumers, due to the pretend safety of chip-and-PIN systems.
Personally, I'm fine with how things are currently, but then I'm not a bank or a credit card company. Not that I have a great deal of sympathy with them at all, given that it's pretty common for card companies like Capitol One to offer credit to students, and the credit card debt is no longer easily discharged through even the heavy handed option of bankruptcy. That law changed under Clinton, and they're about as hard to get rid of as a student loan.
Slashdot Mirror
What if for half the cost of a single person, you could instead have 500 martian probes distributed across the planet?
Besides, NASA scientists have made statements contradicting what you just said. Robots are slower, but can operate nonstop indefinitely where humans need long rest periods.
Humans advocate humans; roboticists advocate robots. Film at 11.
You'd need 1,500 people on Earth working 8 hour shifts, for a much longer period of time, analyzing visual data, and saying "Hmmm... that rock that the probe passed 4 hours of analysis, and 24 minutes ago looked interesting; let's send a command now to the probe, which it will receive in 24 minutes, to go back and look at it. The elapsed time to deciding if it was 'just a shadow' should only be another time to compose command + 24 minutes + 5 hours travel + 24 minutes of transmitting back new pictures of the rock. Then we can decide what to do next, which will only take decision time + compose message time + 24 minutes transmit time + ??? doing time + 24 minutes transmitting back the result."
Do you kind of see the problem?
Low Specific Impulse requirements make the Martian moons a good place for permanent basing, but not so good for staging a Mars mission, and not so great an idea if we are just going to go there to go there, rather than go there to stay there.
Mostly, they are a great staging area for asteroid missions, given that the escape velocity is generally achievable with spring-loaded catapults and electric winches, rather than the more expensive and hard to construct mass drivers that you'd have to build to get mass off Earth's moon.
Can't or won't? I would have thought that it would be possible to create a habitat in either that would require nothing incoming.
So far we cannot. We've tried several times and haven't cracked the problem yet.
We can and we did. It was euphemistically called "The U.S. Government Relocation Facility", but it's code name was "Project Greek Island", and it was capable of sustaining a fairly large population and support staff for 30 years, in the event of a nuclear war.
"The Raven Rock Mountain Complex" was built as a similar "relocation facility" intended for the Pentagon.
"The Cheyenne Mountain Complex" was another facility, for SAC/NORAD.
"The Mount Weather Emergency Operations Center" was intended for use by FEMA; it's sometimes called "The High Point Special Facility".
There are a bunch of U.S. COG facilities (Continuation Of Government) besides that, and most major countries have their own equivalents.
Generally most of them have nuclear power plants, of the type used on U.S. nuclear submarines. Several of the facilities have more than one of them.
We really cannot build a "sustainable habitat" anywhere, "biosphere 2" has the longest record of about 2 years, the experiment ended when they ran out of oxygen, food, and patience with each other.
Biosphere2 failed for many reasons, including the fact that they didn't take into account the fact that curing concrete sequestered CO2 at a high enough rate that the plants were unable to survive, and that in not surviving, they failed to produce O2 from the CO2 they were getting, which made it harder on the animals (including humans).
In the second series of experiments, it was also very common for the door alarms to be bypassed, and they would order out for pizza. The pizza delivery guy who used to go out there lived in the apartment building next to mine at 2000E Roger Road in Tucson. So it was seriously not a sealed environment, and their ambitions exceeded their technical capability, since the thing was more or less funded with the eventual goal of it being a tourist attraction anyway.
I personally could live in a small group, or even on my own, for quite a long period of time, and there are a lot of other people who could do the same; not everyone has to sleep in a "puppy pile" to maintain their sanity or their patience with other people of the same bent.
P.S.: In case you care: the pizza guy's apartment was in the building on the left in the picture located here:
https://media.apts247.info/63/...
Mod it to -1 if you like, people will never *be* on Mars. The closest they'll get is to see it through a visor or a monitor.
And if that's the case, a monitor on a different planet is more convenient.
You are obviously not a geologist. A person, even in a suit, and wielding a rock hammer, and equipped with a rather small lab can do more geology in one day than all of the Mars probes ever sent have done, combined.
Not to mention the fricking communications latency of using RPVs, or depending on the cleverness of remotely targeted semi-autonomous robots.
No, they only mentioned the Linux Foundation code, the 115M lines. And they did it to emphasize what they perceived as the value of the code.
My Point here was that this code neeeds to be broken down into different categories of complexity. I fear they only used a "Default category"
Complexity distinctions are rather specious; let me explain.
Yes, a coding error in an OS can crash the whole machine. But at the time the lines of code stats were written, most computers were not running a protected mode OS, and therefore, a coding error in *any* program could potentially crash the entire machine.
By that token, they've overestimated the value, by treating everything as if it could crash the machine. Unless we relax the criteria to "could crash the machine and/or result in a security vulnerability", in which case we are back to all code having the same level of impact on the machine as all other code. Security was pretty much not an issue back then, because there were no such things as "bad actors who did not just crash their individual machine, or, on a multiuser system, had their account suspended until they talked to someone with a scowl on their face, and received a stern warning".
Also, I will claim that programmers who learned to code when anything they did could crash a machine (and yes, I've crashed mainframes dozens of times with user space code, back in the day, taking everyone's programs with mine -- and had mine shot down the same way) learned to be a hell of a lot more careful coders than the generation who grew up with "if it crashes, it'll only kill itself, and we can just look at the core dump". In other words, they were better coders because they had to be.
Let's turn that back on you - is Uber violating the law where you live but no action has been taken against them? What does that make where you live by your own (very stupid) definition?
No, Uber, the B2C *ride sharing* negotiation service, which takes a cut from the *contractor* offering to *share* their car for $$$ via the Uber *negotiation service* is /NOT/ breaking the law in San Francisco.
Or San Francisco would be collecting $100 a pop for any Uber logo'ed car with passengers other than the driver, because that the fine for operating a gypsy cab in San Francisco.
Therefore, the city of San Francisco is /tacitly stating/ that Uber is /NOT/ a Taxi service.
And believe me, the traffic and parking enforcement people in San Francisco are greedy bastards; if they felt they could legally collect those fines -- they'd do it. An expired parking meter costs you $76 + fees (which are not cheap): https://www.sfmta.com/sites/de...
Perhaps you should write a book about your ideas how software is written.
When COCOMO was developed we had ideas about productivity in software development like this,
per day a developer writes about:
25 - 100 lines of code in user applications
5 - 10 in service software, like specialized editors, TCP/IP stacks etc.
< 1 line of code in system software, especially kernels.
And this is independent of programming language used
These stats are largely inaccurate for modern coders, who are much more productive than if they were writing their code in IBM BAL. If you have a modern coder writing at this rate (on average), then you should likely fire them, and hire someone who can code, instead. The "on average" is because you should spend 90% of your time planning and 10% of your time coding.
On a project (The Whistle InterJet), I wrote a Fetchmail replacement to work around a number of issues that the author felt were unnecessary to fix, and refused to incorporate as bug fixes. As gravy on the side, it got us out from under the GPL on the code, so we didn't have to carry around local patches.
The major specific issue was not reverse time-ordering of the "Received" timestamp line being used to identify an addressee -- Fetchmail overrode earlier "for" values as it moved forward through the message, which frequently lost information, and allowed faked headers to override a destination, in order to abuse Fetchmail as a remailer for SPAM.
The resulting production code was a little over 22,000 lines of code, and was written in a period of 2 weeks of normal 5 day work weeks. That was 1,100 lines of code a day. And yes, I ended up with tendonitis, and had to wear wrist braces off and on for a while after that, after that much typing.
Anyway, your idea how to burn down the costs is just ridiculous.
And yet... Mac OS X is UNIX certified, and Linux is not.
And yet... I was tech lead on the team of 6 full time people that did it, and, in aggregate, wrote more lines of code for Mac OS X Tiger than the rest of the Core OS Kernel team and the Core OS user space team wrote for that same OS release, combined.
Yes, we cut through an metric assload of red tape in order to do it,, and we had the dictatorial hammer of a $200M lawsuit to pound people on the head to enable us to do it. In other words, in a lot of cases, with the aid of Apple Legal,, and the backing of upper management: we were dictatorial asses. If we had to change an API, and it broke your project as a consequence of you using -Werror: you got to fix your project, because the consequences of us not making the change were $200M out of Apple's wallet, and so our hammer was bigger than yours.
I have a number of very, very good friends who were hit by this hammer; but when you work in a corporate environment, one of the things you are paid for is to put the company's good above your personal sensibilities. The other major thing you are paid for is to get along amicably with people you would never work with, were it not for the fact that you were being paid to work with them.
In other words, unlike in volunteer Open Source, where you bike-shed, bicker, and kingdom-build to the point of process ossification, you have your part, you do what you are paid to do, and the product ships, as the product was intended to ship.
The point is: there are assloads of red tape between doing a mediocre job, and doing great things, and you merely need to enable people to do them.
And I bet my ass and my balls that the original estimate is off by a factor of ten, minimum. Because they likely did not judge the projects into the required "category" mentioned above.
No, they only mentioned the Linux Foundation code, the 115M lines. And they did it to emphasize what they perceived as the value of the code.
The "required" ancillary code doesn't matter, because you could just
Hospital emergency rooms go offline during strikes?
How to recognise a dumb shill - he will use an utterly ridiculous hyperbole.
How do you *get to* the hospital emergency room, when your only transportation is public transportation?
How do you *get to* the hospital emergency room *in time* when the French Taxi drivers go "on strike" by way of intentional traffic jams for everyone, because they are pissed that no one wants to ride in French Taxis, if there is any reasonable alternative, whatsoever?
How to recognize a dumb shill: they answer only your direct argument, out of context, rather than taking the consequences of the situation into account.
P.S.: During the 1979 nurses strike in Montreal, Canada, several hospitals had to close their emergency rooms to incoming patients. The Baton Rouge General Medical Center’s mid city emergency room was closed at 7AM on 31 March of 2015 due to a strike by doctors and nurses.
P.P.S.: Why don't you do a google search before you call something "hyperbole" next time?
A freedom to do something is no protection from social enforcement of the consequences of you exercising that freedom.
Ridiculous. By that argument, everyone has the freedom to do absolutely anything they want—they might just get thrown in jail if it happens to violate a law. This makes a mockery of the very word "freedom".
Nevertheless.
Daniel Ellsberg with conspiracy, espionage and theft of government property for his disclosure of the Pentagon Papers. Ellsberg later got off on a procedural technicality resulting from disclosures by the "Plumbers" group members during the investigation of the Watergate scandal.
If you want similar cases where journalists have been jailed for what they've published, look no further than Josh Wolf, Judith Miller, Jim Taricani, Vanessa Leggett, Timothy Crews, James Risen, Glenn Greenwald, Janine Gibson, Eric Lichtlau, and so on (some were merely threatened with censure; others were actually jailed -- all of them since 2000.
The thing that's currently protecting Glenn Greenwald is the backlash that the government would suffer now that he's won the Pultizer.
There's a big difference between "can lead to negative consequences" and "will lead to negative consequeces", or even "would a reasonable person *expect* it to lead to negative consequences -- which is the actual measure taken by the courts.
Yeah, there should be shield laws. There are some at the state level, but for the most part, journalistic speech is *not* protected from government suppression.
Writing hardware drivers / kernel code is a bit more time consuming than writing a database application. Did you take that into account?
You mean like when I personally wrote nearly a million lines in the Mac OS X kernel, over a period of several years, in order to get it UNIX certified?
Yeah, I took that into account.
Perhaps not enough? I should probably reduce that cost estimate, given that pretty much any dink can write user space code and get it to work, and we are mostly talking about user space code...
How dare working people have any rights.
How dare publicly funded services, such as transportation, or hospital emergency rooms, go offline for mere economic enrichment of a few individuals?
It is being fixed where I live and Uber is getting fined and forced to play by the rules or get out so your insult is misplaced.
What was the point of your insult anyway? If you are just doing it to try to prove you are better than some stranger on the internet then that is a very pitiful state of affairs.
No, the point was to correctly place "bribery works" as a measure of "shithole-ness" in context relative to the designations "first world vs. second world vs. third world". Being in the "third world" doesn't make some place a "shithole"; bribery working there as a normal part of the process *does*, most assuredly, make some place a "shithole".
Intrinsic bias exists in gifted programs. Period.
They are biased towards gifted students, and biased against students that are not gifted.
We may argue about of the definition of the word "gifted", but the most commonly accepted definition throughout the history of gifted programs has been "of most future economic or strategic benefit to society".
They are not aimed at producing more Pablo Picassos, Walt Whitmanns, or Maya Angelous, they are aimed at producing more John Glenns, Albert Einsteins, Robert Goddards, and so on. They are aimed at providing adults who would be of most benefit to winning the Cold War, and any similar future conflicts which may arise, in order to ensure the continued existence of the larger society.
What constitutes "gifted" is "more naturally inclined to further the goals of society as a whole". In other words, what society values is what we then term as "gifted", when it is evinced at an earlier age.
If you want to change the definition of "gifted", then first change what society values. You can get a rough idea of current valuations by comparing a two axis graph of "economic reward" vs. "job category", and then look at the top earners in each category. Thus, you get that society values functional plumbing (plumbers), but values substantially less correct grammar and spelling (english majors).
[...]the H1-B system is totally broken and is being used to help decimate the American middle class.
Dec.i.mate: kill one in every ten of (a group of soldiers or others) as a punishment for the whole group.
As long as it's only one in ten, I'm kind of OK with this. Also, I'm kind of OK with the idea that such punishment is actually deserved, since it implies 90% "good apples" and 10% "bad apples", which, if you've ever worked a middle class job, is very easy to credit as an underestimation.
It's only 115M lines of code.
My calculation on that comes out closer to $1.3B for a 5 year project to replace all of it.
With a much smaller number of highly dedicated people who are 3X as expensive as the average software engineer in Silicon Valley, I think it'd be possible to drive the number down closer to $790M and 2 years.
The people would need to be dedicated, and the project would need to be driven by (in effect) a dictatorial ass whom everyone has agreed to follow to the ends of the Earth. In other words, a rather strict hierarchy.
Oh yeah... the project would have to be operated Cathedral style, and not use most "agile" techniques.
a nice upgrade to LLVM.
And all of the architectures that are not supported by LLVM are then screwed...
That's part of the upgrade.
Freedom of the press is not the same as civil disobedience which is not the same as a company ignoring laws.
Point of order: there are no federal shield laws for journalists in the U.S.. Just because there is freedom of the press written into the bill of rights, does not mean that you can not be held legally accountable for what you print.
The common argument is "yelling 'fire!' in a crowded theater", which is commonly misinterpreted as stating that the yelling itself is illegal; it is not, it is protected by "freedom of speech"; the consequences *may* however be something you can have your ass thrown into jail over.
A freedom to do something is no protection from social enforcement of the consequences of you exercising that freedom. That is pretty much the very definition of "civil disobedience".
Rosa Parks was, in fact, arrested for her act of civil disobedience on 1 December 1955.
It's a bit much seeing them act like everywhere is a third world shithole with easily bribed officials, and even worse when it seems to be working.
If bribery works for anyone... then they are right, and your country is a third world shithole.
Instead of arguing about whether or not you live in a third world shithole, fix it.
Err, why do you think that Uber is superior? Surge pricing during a Tube strike is a real bitch, as is the difficulty in arranging for a guaranteed 5am pick-up for the airport arranged the night before.
Quit having 'tube strikes'?
Problem solved.
Wont work unless the HDMI matrix has EDID management so the computers do not see the displays appearing and disappearing.
You need a good quality one, not the junk from monoprice. About $1100 for a good 4X4 with full edid management.
THIS, x1000... with caveats...
The caveats are:
(1) Many monitors won't negotiate an EDID on an inactive channel (i.e. one not selected as the primary channel at the time the computer attempts to negotiate). This is mostly because the firmware in these monitors is not running a finite state automaton and/or is not multithreaded, and so ignores channels which are not currently selected when negotiating. A *lot* of monitors and television sets being used as monitors, particularly Samsung models, have this problem.
(2) When using left/right or top/bottom negotiation, your KVM needs to present EDID information for the effective aggregate monitor size for the negotiation. And most KVMs, even those with EDID management, can't do that. In other words, the KVM has to present one *or more* monitors to the computer as if they were a single, higher resolution monitor. Right there, you've blown the price of the KVM hardware out of most price ranges. Even then, you may find that the KVM suffers from the same "inactive channel negotiation" problem described as a monitor problem in #1.
(3) Most operating systems fail to implement active resolution renegotiation. This is a problem in two ways:
(3)(a) For the proposed use model, when doing mode switching, i.e. when switching between which PC "owns" a given display, the EDID management that negotiated with the OS over the display resolution is no longer valid... the virtual resolutions may have changed out from under them, based on what the EDID management presented as possible display resolutions, vs. what the monitors that ignored the KVM because it wasn't currently active present as the possible resolutions, vs. the difference between presenting one monitor, or more than one monitor, as the virtual monitor.
(3)(b) The OS renegotiation problem may prevent it from working anyway. Specifically, a lot of "media center" PCs are built on top of Linux, and Linux is really poor at renegotiating an EDID with a device when it goes from the "device not connected" to the device connected" state. This problem tends to find itself exacerbated by the fact that a lot of (stupid) monitors will not switch to a non-default input channel, unless they see that there is active input on the port.
Apple products (*not* Hackint0sh!) tend to have much fewer problems in this regard; Windows, it really depends on the hardware (Toshiba tends to be bad; Sony tends to be better at it), the video driver (which depends on the card in use), and which version of Windows.
P.S. Linux really needs to fix its EDID negotiation on these axises -- but FreeBSD is generally worse, even though both these platforms inherit their EDID negotiation from their windowing systems.
If we really wanted to stop CC fraud, we could almost eliminate it. It's pretty simple, but we've abandoned this in favor of convenience.
100% guaranteed that there would not be any credit card fraud, if there were no credit cards. In other news, people who have their appendix removed don't get appendicitis, which is why it was SOP for the Mercury, Gemini, and Apollo astronauts.
I visited USA last year, and was horrified when my transaction went through when the merchant swiped the mag strip on my Australian chip-and-PIN card, without requiring my PIN or signature.
Signatures are not required on charges of $25 or less, since the store is indemnified against a loss up to that amount, when they fail to collect a signature. Over that amount, it usually requires a signature, and then your signature is floating around as a digital copy for a forger to use.
Chip-and-PIN has reduced brick and mortar fraud, but online fraud is alive and well, as is ATM fraud. Just expect that, as in Europe, the U.S. incidence of card skimming, card trapping, and cash trapping to go through the roof. Expect also to see a rise in chip-and-PIN "preplay" attacks. Expect also that the "foreign currency loophole" will be scammed on contactless cards. Martin Emms of Newcastle University in the UK demonstrated this attack at a conference at the end of last year.
NB: Your NZ card would still be susceptible to the "foreign currency loophole", even if there is wide scale deployment of chip-and-PIN everywhere -- anywhere the transaction isn't taking place in the local currency... no PIN is required.
PS: Doug Johnson of the American Bankers Association has stated that banks prefer that we move to one time token systems, like ApplePay and SamsungPay, in any case.
But the simplest way is to use the stolen card to buy gift cards, use these to purchase merchandise, and fence that via reship or whatever, even eBay.
Once the gift card is used, the link to the original cardholder is lost, AVS is useless. In fact, use out of town mules to use the gift cards, bus them in and out, and even the video of them at the register is useless.
Have you tried buying a gift card with a credit card? There are a few mall locations which allow you to do this, on camera, but if you try it at a grocery store, they'll deny the purchase. They'll let you buy it with a debit card, but not a credit card (I got to watch an insistent lady in front of me in line try very very hard to throw a hissy fit until they let her get away with it; it was like watching a 19 year old trying to buy alcohol).
The more straightforward solution seems to be to simply make CC fraud much more difficult. We have the technology to do so, but seem unwilling to implement it. The new CCs in the US are better in that they have chips, but inexplicably still use signatures rather than PINs.
I personally know of 9 methods of scamming a chip-and-PIN system. The only real value they have are to the credit card company, and the merchant, both of which get to blame you, instead of being blamed themselves, for when one of these scams is run. The intent is clearly to offload the cost of fraud onto the consumer, rather than keeping it the problem of the large financial markets that have some hope of being able to curb the abuses, by virtue of economy of scale approaches to the problem.
The typical Russian Mobster who gets scammed simply has the scammer killed.
We could implement two factor authentication for large purchases and when the CC isn't swiped. Simply send an authorization code to the phone on record for the CC, and require it to approve the purchase.
This requires that I carry a tracking device with me everywhere I go, and it requires that I only make purchases in cellular service areas. This usually sounds like a great idea to Europeans, who are going to be on camera everywhere they go anyway, and whose urban density is such that, everywhere they go, they have cell service. So while that may count as "simply" in Europe, it doesn't in Montana, and it doesn't in large parts of California.
In fact, AT&T is taking down 2G towers because they legally have to grandfather the unlimited data plans which they only offered on 2G service, and in order to force people off those plans, they are pulling down the 2G service, even though it's not like they are replacing it with 5G, and it's not like the same tower electronics and antenna weren't capable of 2G and 3G anyway. They've specifically discontinued the "booster" (actually a CISCO Systems box that is a cellular/VOIP bridge, also known as a microcell) for 2G, even though the CISCO documentation says 2G is supported. They do this by refusing to put the 2G IMEI's into their database, even though it would work perfectly fine, were they to do so.
So relying on cell service for two factor -- even long established cell service, which would allow you to roam to other carriers like T-Mobile or MetroPCS -- is not really an option for about 70% of the U.S..
A similar process could be required for applying for a new CC. Reshipping wouldn't be as much of an issue if CC fraud was made more difficult.
Good luck with that, given the known problems that already exist with chip-and-PIN systems. Credit card fraud is here to stay; there's only damage control after the fact, and that's going to become more difficult for defrauded consumers, due to the pretend safety of chip-and-PIN systems.
Personally, I'm fine with how things are currently, but then I'm not a bank or a credit card company. Not that I have a great deal of sympathy with them at all, given that it's pretty common for card companies like Capitol One to offer credit to students, and the credit card debt is no longer easily discharged through even the heavy handed option of bankruptcy. That law changed under Clinton, and they're about as hard to get rid of as a student loan.