RE: Re: Maybe it's me but the integrator at wolfram doesn't work very well
- To: mathgroup at smc.vnet.net
- Subject: [mg45012] RE: [mg45003] Re: Maybe it's me but the integrator at wolfram doesn't work very well
- From: "Owen, HL (Hywel)" <H.L.Owen at dl.ac.uk>
- Date: Sat, 13 Dec 2003 06:05:55 -0500 (EST)
- Sender: owner-wri-mathgroup at wolfram.com
The integrator tool is at: http://integrals.wolfram.com/ The integrator does NOT botch the integral: if you enter 'Log[x]' into the form on the page, it gives the correct answer. I'm guessing you entered 'ln x'. If you read the help at: http://integrals.wolfram.com/about/input/ (by clicking on the link 'How to enter your input', it explains how to input functions. The basic confusion here is because traditional mathematical notation is poor, and ambiguous. Brackets () are used for both functions and collecting terms. Mathematica does it right, by using () for collecting terms, and [] for functions. And of course, you have to use the right syntax for the function name, like in any other program. As always, read the instructions! Hywel > -----Original Message----- > From: drbob at bigfoot.com [mailto:drbob at bigfoot.com] To: mathgroup at smc.vnet.net > Sent: 12 December 2003 09:42 > To: mathgroup at smc.vnet.net > Subject: [mg45012] [mg45003] Re: Maybe it's me but the integrator at wolfram > doesn't work very well > > > Having never heard of the "integrator tool" (a link might be helpful), > I can only give the right answers: > > Integrate[Log[x]/x, x] > > Log[x]^2/2 > > Integrate[Log[x], x] > > -x + x*Log[x] > > Bobby > > geoman <geogeo9090 at yahoo.com> wrote in message > news:<br9i4v$d22$1 at smc.vnet.net>... > > Firstly the "integrator" tool at wolfram evaluates the > indefinite integral. Fine > > > > > > Here are some examples of incorrect answers of the "integrator" > > > > INT means the integral sign > > > > INT 1 / x ln x > > > > Don't feel bad though I found another site at > mss.vanderbilt.edu that got it wrong as well > > > > Now this is quite a fundamental integral but is it > hopelessly botched by the "integrator." > > > > > > Here is another > > > > INT ln x > > > > the "integrator" got that wrong as well. I did find a site > that got it right though. > > > > Where the error lies I don't know but do in fact think it > may be somewhere in the original mathematica code. > > > > BTW can anyone on this list give me the correct answers to the above > > thanks geo. >