Slashdot Mirror


User: dgatwood

dgatwood's activity in the archive.

Stories
0
Comments
14,277
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 14,277

  1. Re:Still overdue on Russian Meteor Largest In a Century · · Score: 5, Funny

    It still won't protect against rouges but at least they can map ones that cross our orbit.

    At first blush, that would seem to reduce the usefulness significantly....

  2. Re:Writing LaTeX directly is often unnecessary on Collaborative LaTeX Editor With Preview In Your Web Browser · · Score: 2

    I suggest you look up what "write-only" means before you spout bullshit. Hint: it means "hard to read, and thus hard to maintain", with Perl often being given as an example .

    I maintain my statement. It is a fundamentally hard-to-read language by its very nature because of the lack of any easy-to-parse delineation between markup and content. It is possible—easy even—to write TeX code that would take weeks to unravel enough to fully understand what's going on or debug. Just take a look at any complex macro package and you'll quickly see what I'm talking about.

    So Spanish is a useless language too, as almost nobody can understand it but other Spanish-speakers.

    That's really not a fair analogy. Because the TeX language is so malleable, TeX requires a turing machine just to parse it correctly. As a result, AFAIK, all the TeX interpreters out there (except for one obscure variant from 20+ years ago) are built on top of the actual TeX source code. For this reason, TeX is like PostScript in that it makes a decent language for print-based typesetting, but a remarkably bad language to use as input to any non-print-based-typesetting tools.

    In contrast, Spanish can be easily translated into other languages, and is not a monoculture. Most human languages that are spoken only by the descendants of a single person have already died out.

    "A boat does only one thing particularly well -- travelling on water". Producing fixed-layout .pdfs is pretty bloody useful for very, very many applications. And produces those so that they actually look good, and the author can focus on content and structure without having to worry (much) about layout.

    Ah, but you missed my whole point. I never said that PDFs aren't useful, or that TeX isn't useful. What I said was that both are lousy source formats, because they are easy to convert to, and hard to convert away from. Good source formats are easy to convert away from, because that's what they're designed for.

    If you author your content in a language that was actually designed to be read by a wide range of apps, such as DocBook, you get the exact same benefits as authoring in TeX, because it is trivial to translate that content into TeX and use a TeX interpreter to produce those fixed-layout PDFs. The only difference is that by starting with XML, you don't have the huge impedance mismatch when translating that content into an eBook format like EPUB or MOBI.

  3. Re:I can say, after having upgraded to mountain li on WebKit As Broken As Older IE Versions? · · Score: 2

    It has nothing to do with the retina display. I'm seeing it on the non-retina current-generation MacBook Pro. I think it is limited to a single model of Intel GPU, though, as I don't see this behavior on any other machines, and it goes away if I lock my machine to use only the NVIDIA GPU.

  4. Re:Problem with egos really on CNN Replicates John Broder's Drive In the Tesla Model S · · Score: 1

    This sounds like a hardware defect. Most automobiles have a relay or bank of relays controlled by the engine control unit that disables power to the rest of the vehicle while the car is turned off. If those relays stick on, whether because the ECU is bad or the relays are bad, you'll get a significant drain on your battery while the vehicle is turned off. The exact amount of drain varies from vehicle to vehicle, but in a car with electric heat, I'd expect the drain to be very high.

    I know this because my rear body module failed on my Ford Windstar. The turn signal started acting like there was a blown bulb (but still worked) and the car would run down its battery in a week instead of a few months. Eventually, the turn signal failed, I replaced the body module, and the battery drain problem went away, too.

    Now as for why a 5% drop in charge corresponds with a 78% drop in range, I'd have to see the actual numbers in order to even make a halfway decent guess. It could be a simple threshold below which it isn't safe to discharge the cells (but which isn't reflected in the percentage numbers for some reason), or it could be that the cells in a series-parallel pack were discharged non-uniformly or that one or more of the cells is weak, resulting in a battery pack that could provide the voltage but not the amperage, or vice versa.

    Eventually, assuming the batteries are wired together, AFAIK, the charges should balance out among the batteries as energy migrates from one cell to the next, but you're basically charging up the weaker battery from the other batteries, which is no more instantaneous than any other charging operation.

    But this is all pure speculation without the actual data, and in truth, we might not be able to go much beyond speculation even if we had the data, because I doubt anybody but Tesla fully understands the way their hardware is wired.

  5. Re:It's called the key on Driver Trapped In Speeding Car At 125 Mph · · Score: 1

    Won't help in electric or hybrid cars, which is what we're talking about here. For gasoline-powered cars, turning off the key cuts off power to the spark plugs and the fuel pump, which effectively does just what you're asking for.

  6. Re:It's called the key on Driver Trapped In Speeding Car At 125 Mph · · Score: 1

    You mean somebody still builds a car with a clutch pedal? Maybe in Europe.... :-D

  7. Re:It's called the key on Driver Trapped In Speeding Car At 125 Mph · · Score: 1

    After further digging, it turns out that you're right; the Prius actually uses a modified brake-by-wire design. It is brake-by-wire for the first part of the pedal's throw (for the regenerative braking), then becomes a normal hydraulic mechanical brake below a certain amount of pressure. So never mind. It's staggering how misleading the reporting has been on that issue.

  8. Re:It's called the key on Driver Trapped In Speeding Car At 125 Mph · · Score: 1

    Ideally, you'd want the relay on the power line to the drive motors, somewhere after the last normal control relay/transistor/* that controls whether power is going to the wheels. Otherwise, that component providing power to the drive motors could fail closed, and rebooting the computer wouldn't help. Any failsafe that doesn't absolutely guarantee a failure is not actually a failsafe. :-)

  9. Re:It's called the key on Driver Trapped In Speeding Car At 125 Mph · · Score: 1

    All non-electric cars, yes, but most of the cars with these brain-damaged designs are hybrids. You can cut off the fuel all you want, but if there's power in the cells, they'll keep going for miles. For hybrids, you would have to add a switch to control a relay that's electrically inline with the drive motor power supply.

  10. Re:It's called the key on Driver Trapped In Speeding Car At 125 Mph · · Score: 1

    Although that is often true for the accidents where someone claims that they pressed the brake but actually pressed the throttle, you can't honestly tell me that you believe that none of the people whose vehicles were runaway for minutes at a time never hit the brake pedal. If you actually believe that, I have a bridge to sell you.

    See, the thing is, these cars don't have real brake pedals. They have little stepper motors and a spring. The computer decides whether it sees the brake pedal being pressed. If that wire breaks or the computer wedges, you can press the pedal to the floor, and it isn't going to slow the vehicle down at all, because you are not in control over the brakes at all. The computer is.

    Thus, you have a single electronic point of failure for your throttle and your brakes, and no way to disable or hard reboot the single computer that is running both of those systems. Without a mechanical key switch, you are no longer in control. At all.

  11. Re:It's called the key on Driver Trapped In Speeding Car At 125 Mph · · Score: 4, Insightful

    I think it is way past time for us (as in every country on Earth) to pass laws that ban any further manufacture of vehicles that lack a mechanical key switch and mandate that any existing vehicles be modified by the manufacturer to comply with that law within six months, or else those vehicles are no longer allowed to be driven on public roads.

    This isn't the first or even the second time this has happened. This is at least the third time I've seen a story about such a runaway vehicle. It is just not acceptable for such a severe safety problem to occur that many times without the manufacturers being forced to design an actual, provable fix, and by provable, I do not mean "We fixed the software bug that caused it to happen in this particular instance".

    As long as you have a computer in complete control over the operation of a vehicle, from the electronic transmission and brakes to the throttle control, a failsafe kill switch within easy reach of the driver should be mandatory, by law. Without the ability to kill the computer if it malfunctions, your vehicle is fundamentally unsafe, period, and should never have been allowed on the road in the first place.

    Unfortunately, knowing our lawmakers in the U.S. and how badly they're in the pockets of industry, such regulations won't happen until the first time somebody dies in one of these situations, and maybe not even then. Perhaps France will do better.

  12. Re:Writing LaTeX directly is often unnecessary on Collaborative LaTeX Editor With Preview In Your Web Browser · · Score: 4, Informative

    LaTeX is basically a write-only language. Almost nothing else can read it except for other TeX variants. And TeX is pretty much a one-trick pony. It does only one thing particularly well: produce fixed-layout PostScript/PDF pages. Other output formats are bolted-on hacks. The problem is that PDF and PostScript are terrible for electronic publishing because of wide variations in screen size and resolution. All the fancy typesetting that looks great on an 8.5"x11" printed page looks lousy when you shrink the PDF down to fit on a seven inch Kindle or Nook screen. For this reason, most electronic publishing is done using HTML so that the reading devices can reflow the content freely to fit the screen. (This is arguably less true for textbooks, mind you.)

    Although I'm told that the LaTeX path to HTML has improved a lot since I last tried to use it, you're still starting from source material that was designed for fixed-layout publishing, complete with formatting instructions, and trying to cram that into a non-fixed-layout publishing scheme. Such an inherently lossy transformation can never feasibly produce results that are as good as you would get if you started out with a proper separation between the formatting information and the content, e.g. authoring in DocBook XML and transforming it to HTML and LaTeX as post-processing steps.

  13. Re:It's the future... on Surface Pro: 'Virtually Unrepairable' · · Score: 1

    We all understand that Apple has convinced people they always have to have the most shiny new thing, and so batteries don't have time to wear out. For the rest of us, though, many parts of devices need replaced (batteries, cables, cases, etc.) long before the useful life of the device is up.

    Maybe for Wi-Fi-only devices, but my experience has been that cellular service providers start moving capacity away from older cell services long before the batteries give out. My original iPhone's battery is still pretty close to its original capacity, but AT&T ripped out much of their 2G capacity in favor of 3G a couple of years ago, and ever since then, it had been dropping calls several times per day. I'm now on Sprint with an iPhone 5.

  14. Re:Which "isle"? Manhattan? on Interviews: Ask Derek Khanna About Government Regulations and Technology · · Score: 1

    Maybe it was covered in asphalt. You know, a heat island.

  15. Re:What happens when the machine dies? on Retail Copies of Office 2013 Are Tied To a Single Computer Forever · · Score: 1

    Continue using the old software whose license doesn't expire, switch to OpenOffice or other competing office suites, use the 365 program, or buy your computer with Office preinstalled so that the risk falls on the hardware vendor.

    Of course, the one Microsoft would like you to choose is #3 (the 365 program). The one most people should choose, unless they absolutely have to maintain perfect compatibility, is #2 (a competing Office suite).

  16. Re:What happens when the machine dies? on Retail Copies of Office 2013 Are Tied To a Single Computer Forever · · Score: 1, Interesting

    With that option removed, Office 2013 effectively becomes a much more expensive proposition for many.

    Not this.

    But frankly, if they are going to tie it to the hardware, then they need to price it as a product that you are going to have to renew every 2 to 3 years. So make it $25, and I'll be happy to buy a new one when I get a new PC.

    This.

    No right to resell and no right to continue using it after hardware problems or major upgrades, etc. means that buying a copy represents a significant risk to the consumer. Based on that, the perceived value of a copy of Office for most people just fell through the floor. At most, the retail copies are now worth no more than an OEM copy.

    I say "at most" because with a retail copy, the consumer takes on that liability for the computer's warranty period instead of the manufacturer, so in practice, the retail copy is worth considerably less than a preinstalled copy of Office.

    But for me, the value is way, way less. As a serious computer user that moves between hardware regularly, if I can't transfer your app (and, for that matter, use it on at least two machines), your app is worth $10. That's an upper bound. If your app costs more than $10 and I can't transfer it freely, I will not even consider purchasing your app. It could cure cancer, triple the size of my genitalia, and make my computer absolutely crash-proof, and I still wouldn't care. Ten bucks.

    Full disclosure: I do not now own, nor have I ever owned, any version of Microsoft Office.

  17. Re:Personally on Ask Slashdot: What Features Belong In a 'Smartwatch'? · · Score: 1

    I can't see myself actively doing anything with a device that small, but I wouldn't mind a watch that acts as a passive display for info from my phone (caller, text messages, meeting name/time/location, etc.) so that I don't have to get it out of my pocket every time it buzzes to find out why it is buzzing.

  18. Re:eInk on Ask Slashdot: What Features Belong In a 'Smartwatch'? · · Score: 1

    I think an e-Ink screen is an absolute must. You'll be looking at your watch often in broad sunlight, and with e-Ink, the screen could be on all the time and not take much power when it's idling.

    Not sure why you'd need e-Ink to do that. I used to use black-and-white passive-matrix LCD panels outside all the time with the backlight disabled. It has only been since the turn of the century that we've forgotten that not all LCDs are bright, full-color things that consume tons of power.... :-)

  19. Re:Testing blood though the skin viable? on Ask Slashdot: What Features Belong In a 'Smartwatch'? · · Score: 1

    Yes, at least in theory, but you'd have to build it into a pair of glasses.

  20. Re:Time? on Ask Slashdot: What Features Belong In a 'Smartwatch'? · · Score: 4, Informative

    The time stamp used in CDMA packets doesn't have anything to do with the OS-level system clock. You certainly could synchronize the system clock to a time value provided by the cellular chipset, but there's no inherent reason that you must do so. The time being shown on the phone's screen could say 3:00 last year and it still wouldn't affect telephony. :-)

  21. Re:Public schooling is a bad idea. on Missouri Legislation Redefines Science, Pushes Intelligent Design · · Score: 1

    Student unions and gyms are arguably a natural outgrowth of the increasing number of students. The larger the college, the less it can realistically rely upon the local community to adequately provide services for the students. At some point, the college basically becomes a town, with all the problems that entails, and must provide comparable facilities. Yes, growth of such services could ostensibly be slowed, but doing so results in fewer students and poorer economies of scale in other areas (including paying off the cost of all the facilities that aren't optional).

    That said, whether it is a required facility or a perk, there are ways to budget intelligently and ways to budget poorly. Cutting the wrong corners during design or construction can often lead to much larger expenses fixing the resulting problems down the road. Cutting corners when buying equipment can result in hardware that becomes useless just a couple of years later. And so on. Unfortunately, avoiding that corner-cutting requires the ability to budget flexibly, spending more money less often, and this is something that the budget system used by public universities does not handle very well, from what I've seen.

  22. Re:Public schooling is a bad idea. on Missouri Legislation Redefines Science, Pushes Intelligent Design · · Score: 1

    That went into effect in 1965, right? College tuition for public 4-year universities increased by about 68% in the decade after that. College tuition between 1910 and 1920 also increased by about 66% at UPenn (the only school for which I could find statistics from that far back).

    In fact, if you take the difference from 1910 through 1920 and keep applying the same multiplication, the 2010 tuition should have been about $24,807. Actual UPenn tuition in 2010: $34,868. So it's not exactly on target, but it is only off by a mere five or six years worth of increase.

    If instead you took the tuition increase from the decade before the loan guarantees went into effect—say between 1950 and 1960—and used 1910 tuition as a starting point, the 2010 tuition would have been be a whopping $231,000 annually.

    In fact, the college loan guarantees were in part a reaction to a huge tuition inflation problem in the late 1940s and early 1950s that was caused by all the WWII veterans coming home and wanting an education under the G.I. bill. That huge influx of students resulted in a huge increase in costs because A. the schools had to increase their physical facilities far more rapidly than normal, and B. it was a short-term bump in enrollment, after which enrollment went back down. Thus, a much smaller number of students had to pay down the resulting debt from those construction costs, and tuition skyrocketed.

    Those folks returning from the war then had kids (a.k.a. the baby boom), resulting in a second bubble that extended from about 1964 to about the mid-1980s. Then, there was a decade of low birth rate, followed by another bubble beginning in 1995. That boom isn't likely to end for a few more years.

    But ignoring seasonal variation, the rate of college tuition increase has been hovering somewhere around 70% growth every ten years since the early 1900s. There was no huge spike from student loan guarantees. The numbers I'm seeing simply don't back up that hypothesis. Unless state spending on education per student has increased significantly faster than the tuition increases, percentage-wise (and I'm pretty sure the reverse is true), loan guarantees have had a negligible effect on tuition.

  23. Re:Look I know God is real, but this isn't the bat on Missouri Legislation Redefines Science, Pushes Intelligent Design · · Score: 2

    Depends on how you define "literal creation". If you mean that as "God literally created the universe," then there's no conflict. If you mean that as "God literally created the Earth and all life upon it in six days, there's a fairly fundamental conflict. A literal interpretation of the book of Genesis simply cannot be reconciled with acceptance of evolution. A figurative or metaphorical view of Genesis is readily reconciled with belief in evolution.

  24. Re:Public schooling is a bad idea. on Missouri Legislation Redefines Science, Pushes Intelligent Design · · Score: 2

    I don't buy that argument. Market forces can't usefully hold the cost of education down. Whenever reductions in education cost occur, the people who invariably take it on the chin are teachers, administrators, and other staff, both in salary cuts and staffing cuts, but mostly in salary cuts. If you keep the salaries down, those teachers will get jobs in industry that actually pay the bills. It's hard enough to convince people to teach at the college level as it is. The last thing we need to do is make changes that further jeopardize colleges' ability to attract talent.

    Don't get me wrong, there are lots of other places where budgets can be cut, but all of those cuts affect things that attract students, and most of those schools get their funding based on the number of students they attract. Thus, the administration is always wary of making those cuts.

    And even if they did make those cuts, they would never be efficient about it. The whole system is quite literally designed to spend more money each year, whether it is needed or not. Instead of allowing excess resources to carry over from one fiscal year to the next like a responsible organization would, in higher ed, every penny that isn't spent goes back to the campus general fund. Worse, they usually reduce your department's budget by that amount going forward. This results in a strong incentive to spend any excess money before the end of the fiscal year. Then, when you actually need money for some unusual expense, you end up with a huge shortfall.

    Until those fundamental structural problems are solved at every level of higher education, the efficiency problems are inevitable and unavoidable. If market forces "held costs down", they would do so by cutting teachers and reducing salaries. But that's not where the bleeding is actually occurring; it is the sum total of all the tiny expenses that causes problems, and those expenses are harder to control. So the quality of education would simply decline on a continuous basis, year after year. Indeed, this is basically what has happened to public K-12 education, precisely because they can't make up the difference through tuition and other external funding sources.

    Of course, that problem is mostly limited to government-run schools. The reason that private institutions' costs have gone up is that they have to be more expensive than public schools, or else you don't have the prestige factor. Yes, to a limited degree, it would not occur if there were not adequate scholarships and other outside sources of funding available, if only because the cost and quality of public education would be reduced, but reducing the cost of an education at Yale isn't a particularly good argument for removing the government loans that enable the poor to attend state schools.

  25. Re:It's a race... on Missouri Legislation Redefines Science, Pushes Intelligent Design · · Score: 5, Interesting

    Oh, no. They got the text of the law exactly right. They said that it had to be taught, then said that you cannot teach who the creator is unless you can prove it scientifically. In order to comply with the law, schools in Missouri will have to teach intelligent design in a way that clearly casts it as an unprovable philosophical discussion rather than science. If anything, this will help disabuse those students of any notion that ID is a true scientific theory, which will actually lead to folks in that state having a better grasp of science in the long run.

    Don't get me wrong, it ain't science, and it really doesn't belong in a science classroom, but since we don't have philosophy classes in American high schools, at least Missouri's students will get to hear the science side of the issue instead of just an ultraconservative preacher's views.