MathGroup Archive 2011

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Series[log[x], {x, 0, 3}]

  • To: mathgroup at smc.vnet.net
  • Subject: [mg119201] Re: Series[log[x], {x, 0, 3}]
  • From: Murray Eisenberg <murray at math.umass.edu>
  • Date: Wed, 25 May 2011 05:54:24 -0400 (EDT)

I agree that in many instances, using something of the form f@x may seem 
at odds with traditional mathematical notation. (Although it's still 
useful in avoiding the eye-nuisance of nested brackets in a construction 
such as g[f[x]].)

But something of the form g@f[x] is very natural from the viewpoint of 
traditional mathematical notation: the "@" is reminiscent of the circle 
operator denoting functional composition.

Usually, using @ seems to be a matter of either stressing a particular 
meaning or else making an expression easier to read. (Making an 
expression easier to type is hardly ever the reason I, at least, would 
use @.)

On 5/24/2011 5:59 AM, Helen Read wrote:
> On 5/23/2011 6:24 AM, Bill Rowe wrote:
>> On 5/22/11 at 6:55 AM, hszhao.cn at gmail.com (Hongsheng Zhao) wrote:
>>
>> While I cannot speak for DrMajorBob, my reason for using the
>> notation f@x rather than f[x] is primarily readability. Constrast
>>
>> Sqrt[Abs[Sin[x]]]
>>
>> with Sqrt[Abs@Sin@x]
>>
>> Both do the same, but for me, it is easier to see what the
>> second form does than the first. Deeply nested brackets are more
>> difficult for me to read. And there is the additional factor of
>> less typing required for the second form.
>
> Each to his/her own. Personally I far prefer the nested brackets, which
> to me is more readable -- it's closer to familiar written mathematical
> notation, and it's clear where each function ends.
>
> And I don't see how the @ sign results in any less typing.
>
> @  requires pressing two keys simultaneously, Shift+2
>
> [ ] requires two keys (one at a time), the [ and ]
> or two keys simultaneously Alt+]  to get matched brackets
>
>

-- 
Murray Eisenberg                     murray at math.umass.edu
Mathematics & Statistics Dept.
Lederle Graduate Research Tower      phone 413 549-1020 (H)
University of Massachusetts                413 545-2859 (W)
710 North Pleasant Street            fax   413 545-1801
Amherst, MA 01003-9305


  • Prev by Date: Re: Part isn't recursive?
  • Next by Date: NDSolve issues with initial and boundary conditions
  • Previous by thread: Re: Series[log[x], {x, 0, 3}]
  • Next by thread: Re: Series[log[x], {x, 0, 3}]