Slashdot Mirror


User: Nutria

Nutria's activity in the archive.

Stories
0
Comments
5,954
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 5,954

  1. Re:Psssh. on New 'No Military Use' GPL For GPU · · Score: 1
    "Conflict is a fact of existence"
    well, sure, with that attitude.


    One word: Testosterone.

    Another word: Adrenaline.

    A phrase: fight or flight.

    Google for George Orwell's opinions on British Pacifism in the late 1930s and early 1940s. They might be instructive.

  2. Re:Decimal Arithmetic on The Trouble With Rounding Floats · · Score: 1
    To be practical, a datatype needs to be usable by the 5 base arithmetic operators.
    And so the problem with just using C++ and overloading some operators is...?


    Nothing. Have you seen it done, though?

    A reference to a Java implementaion would also be tres' useful.

  3. Re:Decimal Arithmetic on The Trouble With Rounding Floats · · Score: 1
    SAP uses them too. ABAP, the language it's written in, is sort of a pascalish c0807 with objects nailed on the side.

    Good to hear.

    Unfortunately, ABAP is a pretty specialized language.

  4. Re:BCD isn't the answer on The Trouble With Rounding Floats · · Score: 1
    64-bit computing is reasonably efficient on most 32-bit machines.

    Are you sure? Such a needs 4 or 6 32-bit registers (depending on whether you overwrite one of the operands with the answer), and x86 has so few registers that you use them all in a single calculation, meaning that you must push the existing contents of the registers. Unless of course you leave one of the operands in RAM, which really sucks.

    And $92,233,720,368,547,758.07 is more than

    If I implied that 92 quadrillion isn't enough, sorry. Not my intent.

    Still, most (all?) fixed point implementations use a scale of 2, which is adequate most of the time, but not all of the time.

    And there's still the problem that Fixed Point is not integrated into any "modern" programming languages (except SQL).

  5. Re:BCD isn't the answer on The Trouble With Rounding Floats · · Score: 2, Insightful
    I mean, do you need a cash register than can tally sums > $1000000?

    You do realize that there's more to business than cash registers, right?

    A 32-bit fixed point number maxes out at 21,474,836.47 which is severely limiting for all but small-sized businesses and tiny governments.

    64-bit fixed point number (max 92,233,720,368,547,758.07) are obviously better, but are only efficient on 64-bit machines, which are still a minority of installed machines.

  6. Re:Decimal Arithmetic on The Trouble With Rounding Floats · · Score: 1
    BCD is fine, but so is fixed point if implemented correctly.

    Fixed Point has typically been implemented using 32-bit integers with a scale of 2.

    BCD is much more flexible.

  7. Re:Decimal Arithmetic on The Trouble With Rounding Floats · · Score: 0
    To represent the cents. And possibly to account for whatever sales taxes are relevent in the country.

    Stupid responses like this are caused by elistist CompSci professors thinking that it's not really important to remind language creators that accurate decimal math is important.

    There is a better way, and it's been around for 45+ years: BCD.

  8. Re:Decimal Arithmetic on The Trouble With Rounding Floats · · Score: 2, Insightful
    The RPG language also had an implementation of BCD,

    Argh, "forgot" about RPG.

    and probably any compiler for the IBM S/370 line would at least have a library for it,

    Having a library isn't the issue. Early versions of TurboPascal also had BCD libraries.

    That you used via function calls. Very not useful.

    To be practical, a datatype needs to be usable by the 5 base arithmetic operators.

    as I believe the IBM mini/mainframe architectures had implemented it in hardware.

    The System 3x0 "CPU" is extremely CISC.

  9. Re:Decimal Arithmetic on The Trouble With Rounding Floats · · Score: 1
    I bothered to ask the question of what to use for monitary
    usage at a financial institution in my recent past.


    http://en.wikipedia.org/wiki/Packed_decimal

    All CISC CPUs had opcodes to do the work, but AFAICT only COBOL (being, of course, a Business Oriented Language) implemented BCD as a primary data type.

    Damned shame, too, since it eliminates all the hassle of working with financial software.

  10. Re:WikiPedia on iPod! on Compress Wikipedia and Win AI Prize · · Score: 1
    Well, since it's currently only 1 Gig?

    You didn't RTFA, did you?

  11. Re:This is a horrible idea on Cleaning Uranium Waste with Bacteria · · Score: 1
    To be fair, I think it's yeast that gives us beer, not bacteria.

    Without http://en.wikipedia.org/wiki/Brewer's_Yeast, there is no beer.

  12. Re:IMHO on How Do I Make Sense of Microsoft Access? · · Score: 1
    Spouting blather on Slashdot? Where have you been the past few years?

    I know, I know. Still, shining the light of critical thought onto Wrong Conventional Wisdom is always a Good Thing.

  13. Re:This is a horrible idea on Cleaning Uranium Waste with Bacteria · · Score: 1
    Screw the bacteria! What's it ever done for us? Well, aside for cheese. And beer.

    Bacteria spoils beer, which is a Bad Thing, no?

  14. Re:IMHO on How Do I Make Sense of Microsoft Access? · · Score: 1
    Java which is, IMNSHO, the worst programming language evar (except maybe COBOL)

    Have you actually used COBOL in a production environment, or are you just spouting elitist CompSci blather?

  15. Re:Twelfth of Never on Next Generation Stack Computing · · Score: 1

    In fact, Windows NT did ship for the PowerPC PREP platform

    And the Alpha. We (the company I worked for) actually had an AlphaStation 255(?) running Windown 3.51(?).

  16. Re:No - that's bullshit on How Not To Run a Campaign Website · · Score: 1
    Any system adminstrator worthy of the name should be able to do system-level development as well -- debugging everything from applications to drivers and other kernel-level code as necessary.

    You're right. That is setting the bar pretty high.

  17. Re:Looks like a stomp and a doorslam. on Lead PHP Developer Quits · · Score: 1
    However, one thing that you do learn as you get older is when you feel strong negative emotions towards other people, hiding them is the best first reaction.

    Wow, isn't that a change from the let-it-all-hang-out 1960s and 1970s. When did this change occur?

  18. Re:Beware automatic grading... on Resources for Programming Course TA? · · Score: 1
    Automatic grading apps are, uh, somewhat prone to breakage.

    Programs with bugs. What a fscking novel revelation!

    (fork off a child process that sleeps for a bit then overwrites the grade file, then just output whatever)

    What that means is that their security process sucks eggs.

  19. Re:MS, grrr on PowerPoint 0-Day Points to Corporate Espionage · · Score: 0, Flamebait
    especially home users for whom a patch would pose very little problem even if it was badly written

    That's just about the dumbest thing I've ever read on Slashdot.

  20. Re:ah well, that's all we can muster? on Paul Thurrott Bitten by WGA · · Score: 1
    either way you get exactly what you deserve.

    Either way? Which ways are that?

  21. Re:Useless indeed on Japan Plans 30-Year Supercomputer Forecasts · · Score: 1

    Their effort should be spent on increasing the amount of data collected over the globe to have a significant number of datapoints to analyze.

    That is certainly a very important key.

    A much better understanding water vapor's effect upon weather is also crucial factor.

  22. Re:Question on Debian Server Compromised · · Score: 5, Informative
    I realise that debian stable release has packages that are very old in order to stay stable. Does this mean that they lack patches later versions of programs use? Or are patches typically backported to the stable release packages?

    http://www.debian.org/security/

    Security (not feature) patches are backported if possible, and if the patches are too extensive, an upgraded version goes into Stable.

  23. Re:And this is bad? on State Department Hit With Many More Break-Ins · · Score: 1
    The Pentagon Papers trial created a fine balance that is worth preserving. The government can keep things secret in the interest of security, but at the same time it's not illegal for the press to print whatever is leaked to it.

    That's the difference between the Pentagon Papers and the State Department cracks.

    The Papers were leaked by an insider, in the June incident, foreign nationals probably working for a semi-enemy country cracked into goverment computers.

  24. Re:I'm going to have to use the /. rule of thumb on OSS Web Stacks Outperformed by .Net? · · Score: 1

    To Dell/HP/Etc - You must not sell naked or Linux systems or your the price of OEM Windows gets larger.

    To my knowledge, Dell does sell naked or Linux systems.


    Look harder. In 45 seconds, I found the PowerEdge 850, which comes naked by default, and has options for Win2k3, RHES and NetWare 6.5.

  25. Re:NOT a hard drive alternative on A Magnetic Memory Alternative to Hard Disk · · Score: 1
    On a desktop machine, program load time and swap space access speed are much more important than data access. Almost all desktop applications perform relatively little data I/O compared to that involved in loading & executing them.

    Really? How often do you start up your web browser, email client, word processor, spreadsheet app, bit torrent, etc? Once a day or less, depending on the app. You load them, and they sit there, waiting for you to use them.

    Or does Windows suck so bad that you've got to close apps as soon as possible, to free up RAM for other apps?