MathGroup Archive 2009

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

Search the Archive

Re: HoldAll for Integrate

  • To: mathgroup at smc.vnet.net
  • Subject: [mg98108] Re: HoldAll for Integrate
  • From: Szabolcs Horvát <szhorvat at gmail.com>
  • Date: Mon, 30 Mar 2009 04:44:37 -0500 (EST)
  • References: <gqfknb$kht$1@smc.vnet.net> <gqia1o$ofe$1@smc.vnet.net>

dh wrote:
> Hi Tammo,
> 
> what Integrate really does can only somebody from Wolfram explain. But 
> 
> it completely eludes me why the integration variable in "Integrate" is 
> 
> not localized. Maybe some can give a reason for this.
> 

Hello Daniel,

It does not make any sense to localize the integration variable inside 
Integrate because the result may contain the integration variable (in 
case of indefinite integrals).

However, it might be argued that it is a bad idea to use the same syntax 
colouring for the integration variable that is used for localized 
variables (e.g. in Table), as this will make it impossible to tell from 
the colouring whether the variable has a value.


  • Prev by Date: Re: Two versions on the same system
  • Next by Date: Re: simpler way to get a particular banded matrix?
  • Previous by thread: Re: HoldAll for Integrate
  • Next by thread: animation in version 7.0 looks bad