Slashdot Mirror


User: romons

romons's activity in the archive.

Stories
0
Comments
378
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 378

  1. Re:Dangerous recursion! on Stack Overflow Could Explain Toyota Vehicles' Unintended Acceleration · · Score: 1

    Many processors will use alternate stacks to service interrupts. All the user threads use a different stack pointer. Some processors allow you to set up different stacks for each level of interrupt.

    Also, you don't need a disk to do guard pages, just an MMU. I did it for a MIPS OS I wrote, mapping stack pages out into the boonies of the address space. Overflow or underflow caused a page fault, which vectored to a recovery routine.

  2. Re:This is a case of manual override on Stack Overflow Could Explain Toyota Vehicles' Unintended Acceleration · · Score: 1

    I see this as critical in a driverless car. There needs to be a way for people to pull the plug and there needs to be a way for people to phone in an emergency. So if someone is lying in a pothole being run over by car after car, or the bridge is failing, there needs to be a way for 911 to say that a stretch of road is now cut off. The key is that this cannot be ab abusable by officials. I do not want my car grinding to a halt because the police are looking for some runaway or a bank was robbed.

    Or the Governor doesn't like the Mayor of a particular city.

  3. Re:Go Amish? on Stack Overflow Could Explain Toyota Vehicles' Unintended Acceleration · · Score: 1

    Yes. I had a 2005 prius, one of the floor mat hooks broke almost immediately, and the remaining one would unhook regularly when I pushed on the mat to get out of the vehicle. When the mat was unhooked, it got pushed up under the pedals unless I was very careful. The dealership gave out free hooks if you asked for them.

  4. Re:Go Amish? on Stack Overflow Could Explain Toyota Vehicles' Unintended Acceleration · · Score: 1

    If someone comes running out from behind a tree or van or corner or doorway, there will almost certainly be an accident.

    Yes, if someone tries to throw themself beneath your wheels it will still be possible to have an accident. But a self-driving car will obey posted speed limits and avoid excessive acceleration (see prior point) and thus this will be less of an issue. Also, the car can have thermal vision and be much better at sussing those situations than are you.

    Sure. The car will be able to predict, with reasonable certainty, whether it is safe to zoom by a van at 30 mph. I'm the sort of driver who, when I can't see around something, will slow down to a speed where I can stop if somebody runs out, since this happens all the time. I'm sure autonomous cars will be able to do this.

    I saw an ad suggesting that the new braking software on high end cars predicts when the guy in front of the guy in front of you will slam on his brakes. Not sure what the limits of that are, but if they are really doing that, they are already better than I am.

    The real win with autonomous cars, however, will be with older drivers, who are unwilling or unable to give up driving, and with drunks, who cause something like 1/2 of all fatal accidents. Compromised or sleepy drivers. If your car can get you home when you are shit faced, the technology will have won, and soon it will be like seatbelts, required on all vehicles. I can't wait.

  5. Re:Survey results != Real world on Psychologists: Internet Trolls Are Narcissistic, Psychopathic, and Sadistic · · Score: 1

    The really shocking thing is that nearly 10% of the respondents who comment self-identified as trolls. If your facts are accurate, that would mean that perhaps twice that many are trolls. So, 20%? Hardly seems possible that any discourse could go on with that percentage of people who are committed to hurting others.

    On the other hand, everybody has a mean streak. Some hide it better than others. If everybody who has noticed that at one time or another they have been very mean to someone online responds in the positive, well, that might explain the numbers. Otherwise, I don't buy it.

  6. Re: Cost on Ugly Trends Threaten Aviation Industry · · Score: 1

    Yes, this. Also, fewer people out of the military who want to fly. Better airline routes. More security. Fewer interesting destinations. I got rid of my Cessna in 2003. I miss it occasionally. I don't miss annuals, tie down fees, or increasing airspace restrictions.

  7. Professors are in a conflict of interest on Adjusting GPAs: A Statistician's Effort To Tackle Grade Inflation · · Score: 1

    The conflict is between academic standards and promotions. They want to do the right thing, they simply can't if they are going to be rewarded by the system.

    So, fix the problem by having two professors for each course. One of them gives the lectures. The other one creates homework, writes tests, and grades them. You could even do the latter with a committee. The students need to know that the professor giving the lecture is not the one responsible for homework or testing, which may or may not be standardized. They also need to NOT know who is writing and grading their tests for a particular class, at least before the class is given. Then, student recommendations will focus on what is important, mainly, how well the professor communicates the subject. The committee (of which the teacher may be a part) can thus give cover to the act of suppressing grade inflation. "Not my fault you got a D, that was the committee! I'm only one small cog in a big machine!". This will also create a bit of feedback on who is actually teaching, and who is mumbling to the blackboard.

    As part of this, the homework/grader professors should not know who the students are that they are grading. That information should be kept by the lecturer alone. That might keep them honest. If there was a professional competitive situation between the professors, that might also be a problem, but using a committee should fix that. It would also cut down on the work required.

    All this assumes that the point of school is to learn the subjects being taught. I'm not sure that assumption is valid, however. It may in fact be a form of test, to judge whether one is worth to be admitted to the priesthood. If so, grades don't matter at all. What matters is recommendations and personal contacts.

  8. Re: "Not Reproduclibe" on GOP Bill To Outlaw EPA 'Secret Science' That Is Not Transparent, Reproducible · · Score: 1

    there has never been a single reproducible experiment in macroeconomics, ever. In fact there's been precious little *scientific* study of the matter at all, it's all philosophers spinning stories compatible with their biases.

    There have been lots of reproducible experiments in macroeconomics. They happen all the time. They are called legislation, and are analyzed constantly by economists. They may not be well designed experiments, but they are experiments. One form of this is predictions about how a policy would affect certain key variables. Krugman and others predicted that interest rates would not rise due to the expanding money supply (QE). Others predicted a dire calamity, where interest rates would return to those of the early 80s. Krugman and Bernanke and Yellen were right, the others were wrong. So, that theoretical model better fits the current facts, and so is probably true. If you bet on gold because you believed those freshwater guys, you got screwed.

    The FED has been managing the economy with monetary policy since forever. They've done pretty well at keeping inflation in check. Not so well at keeping unemployment low. They also couldn't prevent huge corrections like the stock bubble in the 90s, or the housing bubble in 2008, but look at interest rates! That has been their real mandate, at least since the 80s. They aren't in the business of saving speculators from themselves. (They are in the business of propping up banks. That is what they do, right? What is the interest rate on your savings account?)

    Similarly, we are currently running a long, dangerous experiment in climate science. The predictions are there, and preliminary results suggest that the hypothesis (that we are screwed, and millions will die) is correct. However, we won't be sure until the experiment is complete. The Koch brothers just want us to complete the experiment, right? They are pro science. They support NOVA!

  9. Re:The more simple you make it the less complex it on Ask Slashdot: Why Are We Still Writing Text-Based Code? · · Score: 1

    The reason programming languages are still as they are is for a simple reason, because you can't produce something complex with something simple, I.E. the more you simplify something the less control you have of it. Can a programming language be made that is not text based? Sure, but I highly doubt you are going to get the flexibility to do a lot of things. Even assembly is still required sometimes.

    The assertion that complexity cannot come from simplicity is patently false. There are far too many examples to list.

    Graphical tools for hardware and chip design are mainly for replicating patterns over and over, and testing their connections.

    The real reason is that programming these days is a sequential process (mostly). As such, you need one dimension, rather than 2 or 3. So, visual paradigms make little sense. Nobody uses flow charts anymore for good reason. Most visual coding paradigms are ways to input flow charts.

    Programs are a linear narrative, because that is how the CPU sees it. So, it is treated like that. Once we have thousands or tens of thousands of interconnected CPUs working together, 2d programming paradigms will make more sense.

  10. Sad and Stupid on How Adobe Got Rid of Traditional Stack-Ranking Performance Reviews · · Score: 1

    The real way to do assessments is peer ranking. Managers often don't even know what their reports are doing, at least in any detail. However, everybody on the team knows that Bill in the cube near the window has been spending more time reading slashdot beta than working, or that every time Sanjeev tries to do anything, things get fucked up beyond belief.

    This opens up a whole set of other problems, but it allows managers do what they should be doing, which is to set goals, and then to acquire resources needed to accomplish them on, so said reports can do whatever the company actually needs. If the team is then judged on results as a whole, the incentives are in the right place to get things done.

    Having a manager offer mentoring assumes the manager actually has something useful to say. This is nearly always false. Line managers are typically either brown noses who got promoted because they failed miserably in their primary job, or technical people who were in the wrong place at the wrong time, want to continue to do technical work, got sucked into management, and have no idea how to fulfill their real job (setting goals and acquiring team resources.) The first ones can get resources, but usually let their teams spin out of control while they dream of moving up to middle management, and the second ones argue with their managers, preventing them from getting resources, while they micromanage to the point of stifling any creativity, causing anyone with talent (this manager's competitors) leave the team in search of a team managed by the first type. This leads to teams where the talented are out of control, wasting resources on stupid projects that aren't in line with the real goals of the company, and other teams where the untalented are being dragged along by a former superstar who can't get any resources, and further can't understand why her lazy team members aren't willing to work 80 hours a week doing things (badly) that they could do much better during a long lunch.

    All of this goes away with peer ranking, which enables teams to self-assemble and repair themselves when required. If you then fire or reward entire teams based on how well they accomplish their goals, you may actually get something done. I don't define releasing buggy adobe shitware as 'getting something done'.

  11. Re: Should Everybody Learn Calculus? on Should Everybody Learn To Code? · · Score: 2

    As a math and CS major (circa 1983), I can verify what you say. Math classes were designed for learning the rigorous definitions and proofs. They weren't designed to help CS or science majors. On the other hand, neither were the CS courses. They were concerned with analysis of the order of algorithms. There were a few lab classes, but the main thrust was theory, to get ready for graduate school. My first job was a huge shock, since I was prepared for the wrong job. Classes in debugging would have served me better than analysis of sort algorithms, which was interesting but useless.

  12. Re:Just catering to their demographics on David Pogue and Yahoo's "Normals" Problem · · Score: 1

    I don't think I've ever run across a person with extreme-left-wing politics in the U.S., oddly enough. I've met a few communists, but they're all Europeans.

    The power of indoctrination. We in the US were all shown the movies of the 'red menace', taking over the map of the world in school. We are thus terrified of communism, while accepting as a fact social security, welfare, medicare, obamacare, labor unions, unemployment insurance, strict government control and oversight (and sometimes government takeover) of corporations, etc. These are all overtly socialistic programs that most Americans agree with. We just can't call it socialism, because then we've fallen victim to the imaginary red menace.

  13. Re:Just catering to their demographics on David Pogue and Yahoo's "Normals" Problem · · Score: 1

    That's funny, I've noticed the same except with my left-leaning friends (and extreme left-wing politics of course). I'm not trying to be facetious, it does seem that way to me. Perhaps it says more about us than them?

    The fact that anybody can find evidence to support any position they want to take makes this sort of thing possible. There used to be a supposed 'neutral' font of news in the networks, but they are basically gone. Everybody gets to exist in their own political bubble. When our bubble is challenged, we get angry and defensive.

  14. Re:Just catering to their demographics on David Pogue and Yahoo's "Normals" Problem · · Score: 1

    One thing that has really stood out for me in the last 5 or 6 years is just how conservative their readers tend to skew. It's where the Fox News crowd goes. Just read the comments section of any random news story and you'll see what I mean.

    So, the right thing to do is to use big data to determine the political slant of whomever you are serving pages to, and to prioritize stories that you think they will like. This is what google seems to do.

    This Pogueish nonsense is a stupid strategy. They aren't FOX news. FOX became FOX because they were the first network to pander to an audience for ratings in a long while (since Hearst papers were making up stories, I suppose). There had been a taboo on that sort of thing, probably as a response to overt propaganda during WWII. FOX broke with that, and used it to their advantage. They have basically destroyed the ability of the country to come to consensus politically, but they are making money, and now couldn't back off if they wanted to.

    Yahoo, on the other hand, can't really dominate in this way. Nobody uses the web the way they use television. Links go everywhere, and there is no exclusivity. There are also lots of alternatives who have already tried it. They are trying to differentiate from Google, but it is stupid.

    Google wins by using big data to serve up a user's own biases. This also destroys consensus and community in general, but what the hell, they are making money. If you define evil as "not making money", they they are holding true to their stated purpose.

  15. Oddly on Searching the Internet For Evidence of Time Travelers · · Score: 1

    I was thinking about this this morning, before I read the article. Spooky!

    However, the clear answer is that this is the reason we never see aliens. They invent time machines, and then some alien schmuck goes back in time and prevents life from existing on their planet.

    We are almost there. Watch for strange gravity waves from Neptune. Oops, spoilers.

  16. Re:This is the problem with religious people. on US Justice Blocks Implementation of ACA Contraceptive Mandate · · Score: 1

    For some strange reason, the ACA did not fix this problem. We need to decouple health care and employers by eliminating the tax break that employers get.

    McCain wanted to do that in 2008

    Didn't happen because politics.

    The only reason McCain could say he wanted to do that is because he knew it would NEVER pass in congress. Republicans have a history of offering solutions to health care during elections that seem to dissolve if they are elected. Not too surprising.

    Obamacare (Actually, Reid/Pelosi care) sucks. That is because it was the only thing they could pass, given the politics (Democrats also take money from insurance companies). However, THEY DID PASS IT. We have been trying to do this since Teddy Roosevelt. It is going to help millions of people, will provide sorely needed economic stimulus, and allow more entrepreneurs to start businesses without fear of losing their healthcare. There are a whole host of clauses in it that attempt to limit the insane inflation of healthcare costs, and appear to be working. It will prevent many of the the bankruptcies that affected 2 million people a year in the US.

    It isn't what I wanted, but it does pretty well.

  17. Re:NSA leaking on The Biggest Tech Mishap of 2013? · · Score: 1

    There were two sides to the CIA when it was formed; the I guys, and the spy guys. OSS was the spy guys, wasn't it? Weren't they doing things like overthrowing democracies to set up puppet governments?

    Gore Vidal pointed out that the CIA was never really a legal organization; it was always a bit outside the law. They could do whatever they thought would be 'best' for the country. If that meant spying on US citizens, I'm sure they would have done it (and probably did do it) without a thought. Spying on citizens is certainly less onerous than experimenting on citizens using LSD, without the subject's knowledge or consent.

    What the NSA is doing was instigated by Cheney and Bush after 9/11. They had astonishing resources and little oversight, so they ratcheted up. Why not? Bureaucracies exist to expand themselves. I'm disappointed that Obama and his gang didn't do any oversight when they came in. I guess they were too busy saving the economy and helping the poor to set themselves up as culpable if a terrorist attack happened.

  18. Tribal Lore on New Study Shows One-Third of Americans Don't Believe In Evolution · · Score: 1

    I don't believe it for a second. The fact is that if you thought your group believed that the earth was flat, you would say it was flat, even if you knew it to be false. So, the media frenzy on this makes it more likely that people will SAY that they disbelieve evolution, because of their need to be part of the tribe.

  19. Re:CCTV Link on Space Junk or a Meteor? Fireball Lit Up Midwestern Skies · · Score: 1

    Looks like the start of "War of the Worlds". Anybody notice any three legged aliens in the area?

  20. Re: Easy solution on E-Books That Read You · · Score: 1

    My kindle is always connected using their 'whispernet', which is cell phone tech. They could, if they wanted to, use that to snoop on reading habits.

    I really am not concerned about this. At this point, they can track me in the supermarket using my cell phone when it is in my pocket. Tracking how I read a book I've purchased isn't such a big deal compared to that. If Amazon isn't using its sales information to its advantage, it isn't doing its job for its shareholders.

  21. Re:Incentives. on Netflix: Non-'A' Players Unworthy of Jobs · · Score: 1

    The 'invisible hand' market based world would simply fire people who didn't directly contribute to the bottom line. This makes it possible (although not likely) for resources (that is YOU) to be used more efficiently in the economy. It is similar to the strategy that a farmer would use when pulling weeds and using them for mulch. You don't feel sorry for them, do you?

    The government, on the other hand, has a constitutional mandate to 'promote the general welfare'. This means they are not allowed to allow people to starve to death, or be mulched.

    So, the way it should work is that government should tax the shit out of folks like netflix to provide a safety net for the 90% of people they cast aside when they've lost their usefulness. This would have the added advantage that people could go and work someplace where they could contribute without having their kids starve when they lose their healthcare and end up in bankruptcy due to a preventable illness.

    I mean, we have all had coworkers that have been shit at their jobs, knew they were shit, but were afraid to go someplace they could be useful. Some of us have actually been these people. The crappy safety net in the US prevented these people from leaving. That in itself is a net drain on the economy. People should either be employed in something they can do properly, or be put safely in front of a TV where they can't do any harm. We have enough resources in the economy for everybody to eat, have shelter, and have healthcare without doing 'make work'.

  22. Re:Sometimes those warnings are muted on 2013: an Ominous Year For Warnings and Predictions · · Score: 1

    It wouldn't matter if 'climate change' were true or not. They'd do 'deny' it, because it's good business to protect your interests, particularly against the ever changing political winds.

    More money was also probably spent on beer advertising. Why? It pays dividends. I don't see what your point is. People (and the companies they run) make choices in the interest of self preservation and self-interest.

    Maybe science should stop doing warnings and studies and let things happens with no preparations from our side.

    That'd be nice - 'science' could just stick to doing sciencey things, then, instead of creating contrived and falsifiable histrionic reports about things which, almost invariably, will not prove out to be true.

    You mean non-falsifiable, I think...

    The problem is that scientists are mostly trying to figure out what is happening, and present it in a logical way, and the deniers are mostly generating propaganda to further their personal financial interests. I find it hard to put these two groups into the same category.

    The true stroke of genius by the 'Koch' guys was to somehow graft the denial of global warming onto political affiliation (using Al Gore's selfless position as a starting point). That way, blind belief becomes a badge of membership for conservatives.

  23. Re:What about the Little Ice Age? on Sun Not a Significant Driver of Climate Change · · Score: 1

    The above got 5, insightful? Sheesh. I'm going to reddit. You guys are pathetic.

  24. Bioluminescent light bulbs are the answer! on US Light Bulb Phase-Out's Next Step Begins Next Month · · Score: 1
  25. Re:News for Nerds? on Oregon Signs Up Just 44 People For Obamacare Despite Spending $300 Million · · Score: 1

    Sorry, you are simply wrong about all republican proposals including exchanges. Regarding your medical bankruptcy information, the study most commonly used do make that claim treated every bankruptcy which included a medical bill as a medical bankruptcy, no matter how small a portion of their total debts was tied to medical bills.

    So, 3 out of 4 quoted in your washingtonexaminer article seem to include exchanges, and the fourth is widely considered nonsense.

    Regarding bankruptcy numbers, are you seriously suggesting that medical costs are not contributory to a huge number of bankruptcies in the US every year?