• Open Menu Close Menu
  • Apple
  • Shopping Bag
  • Apple
  • Mac
  • iPad
  • iPhone
  • Watch
  • TV
  • Music
  • Support
  • Search apple.com
  • Shopping Bag

Lists

Open Menu Close Menu
  • Terms and Conditions
  • Lists hosted on this site
  • Email the Postmaster
  • Tips for posting to public mailing lists
Re: a numeric bug.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: a numeric bug.


  • Subject: Re: a numeric bug.
  • From: Helmut Fuchs <email@hidden>
  • Date: Fri, 16 Sep 2005 10:01:43 +0200

At 17:55 Uhr -0400 15.09.2005, deivy petrescu wrote:
Please document it.

For once I want to defend the dated and incomplete AppleScript documentation: You will run into that problem in EVERY language using floating point.


Most computer languages employ the floating point capabilities of the underlying hardware and operating systems. It's documented THERE.

(If you look at Apple's Intel transition docs you'll see what incredible subtleties there are in how floating point is done on each platform.)

At 17:55 Uhr -0400 15.09.2005, deivy petrescu wrote:
Somehow I feel that if a program offers the tool and makes no point on the limits of the tool, something is really wrong.

Even every "scientific" pocket calculator I know of has those limitations, though they may be set a little higher. Try it! And then look up the reason for getting a wrong result in its manual...


I guess most people doing serious calculations with floating point have been bitten by its limitations at least once. But then they've learned a lesson, which they won't forget and wouldn't have learned by reading about them in the docs, because only READING about these limitations is simply far too abstract for most.

Just recently a colleague had a problem with some software that computed the cumulated taxes for something or other. And the results were simply wrong. He, a computer scientist with a university degree, simply didn't realize that even "rounding in far away" places might quickly do visible harm to numbers. He had to learn by experience that you need to keep your eye on floating points and better limit/cease their application in financial math.

If you want to do math with large numbers or very high precision one ALWAYS has to you use some specialised tool (like "bc", a special library, or a do it all wonder like Mathematica).

Regards,

Helmut
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Applescript-users mailing list      (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden


References: 
 >Re: a numeric bug. (From: has <email@hidden>)
 >Re: a numeric bug. (From: Emmanuel <email@hidden>)
 >Re: a numeric bug. (From: Christopher Nebel <email@hidden>)
 >Re: a numeric bug. (From: deivy petrescu <email@hidden>)

  • Prev by Date: Re: convert AS syntax to CLI
  • Next by Date: Re: iTunes 5 scripting - thoughts?
  • Previous by thread: Re: a numeric bug.
  • Next by thread: Re: a numeric bug.
  • Index(es):
    • Date
    • Thread