Date: May 18, 2013 2:38 AM Author: Andrzej Kozlowski Subject: Re: Work on Basic Mathematica Stephen! I agree with almost all of this with a couple of important exceptions.

Firstly, Mathematica 9 has lots of, what to me, is tremendously

important mathematical functionality. I am referring to functions such

as RandomFunction and numerous functions related to stochastic

processes, including Levy processes etc. For example, you can now

generate a path of a Wiener process in Mathematica 9 like this:

ListLinePlot[RandomFunction[WienerProcess[], {0, 1000, 0.1}], AxesOrigin -> {0, 0}]

Try programming this yourself in Mathematica 8 (not too hard but not

trivial either) and then compare performance. You can also do this just

as easily with much more complicated stochastic processes, where even

achieving any reasonably efficient result with Mathematica 8 will be a

tough job. This is enormously important for applications in all kinds of

areas of science and in mathematical finance. I don't know of any other

program at this price level that can match this.

I mention this fact in order to make a point that nobody has been

making: these sort of things are important for many Mathematical users,

in fact infinitely more so that the kind basic language issues that

Richard Fateman has been carping about for over 20 years. These kind of

functionality is why many people use Mathematica. I therefore absolutely

disagree with the statement: "I am very happy with the functions

available in Mathematica 8". Or rather, Mathematica 8 was fine (although

it could not, for example, fit the Meixner prosess to the S&P500 stock

price without lots of extra work) but Mathematica 9 is a marvel (and, of

course, it can do it easily).

There are of course other areas of mathematics and science where the

same can be said. While I agree that we probably do not need many more

basic language functions, there is actually no limit to how many

mathematical functions Mathematica can accommodate. User created

packages are not really suitable for any work that is computation

intensive.

Now, of course, basic stability is the most fundamental issue: there is

not point having a program with the most magnificent computational

abilities if it is going to crash all the time. I also agree that the

FrontEnd has fallen far behind the Kernel in stability - this really

does need to be given the highest order of priority. But don't forget

that these two things do not really compete with one another. People who

work on mathematical functions are not the same as the ones who work on

the front end - these are two different specialities and different

teams. It's pointless to say: we have enough mathematical functions,

let's now concentrate on eliminating the Front End bugs. It doesn't work

this way.

Andrzej Kozlowski

On 17 May 2013, at 10:35, Fred Simons <f.h.simons@tue.nl> wrote:

> Like David Park and Raul Martinez, I am a user of Mathematica since

> version 1. Since I am (was) really very enthousiastic about Mathematica,

> in my courses on Mathematica that I regularly present to professionals

> who are going to use Mathematica or want to deepen their knowledge of

> this package, I had no problem in convincing them how powerfull and

> strong Mathematica is (was).

>

> Now we have Mathematica 9. My feeling about this version becomes more

> and more negative. We have the problem with the sudden change of the

> default style between version 8 and version 9. Going back in version 9

> to the style of Mathematica 8 is not bug free, so I, and the

> participants of my courses, have problems with the course material.

>

> Mathematica 9 is very unstable. For me it is the poorest version that I

> have used since the frontend was added. In many of my notebooks, I

> cannot use simple commands like selecting a name and pressing shift-F1

> for going to the help, or ctrl( for converting a selection to a

> mathematica formula. The result is that I get the message Mathematica 9

> stopped working...

>

> Last Tuesday I gave a course on interactivity in Mathematica.

> Mathematica 9 was running on 7 platforms. On at least 5 of them

> Mathematica had to be restarted during the course because of it started

> showing incorrect output. When things like this happen, I cannot

> recommend Mathematica to a potential new user ....

>

> But what worries me most is my feeling that WRI seems not to be

> interested in producing a stable, reasonable bugfree release.

>

> I read some examples in this thread. My personal example is that there

> is a bug in Mathematica with respect to the MINGW-compiler. That bug

> forces me to keep using Mathematica 8 for my applications that use

> compilation. The bug is found by WRI (good work!) and I was informed

> about it. But the bug is not repaired in the latest releases ...

>

> Like many of you, I hope that WRI will lay their priority on the quality

> of the software, instead of on the number of included functions.

> Personally, I am very happy with the functions available in Mathematica 8.

>

> Fred Simons

> Eindhoven University of Technology

>

> Op 13-5-2013 9:49, R Martinez schreef:

>> All,

>>

>> Count me as one who has been using Mathematica since Version 1, and who uses Mathematica for serious Mathematics and for teaching. Mathematica enables me to do things expeditiously that would take a tremendous amount of effort to do otherwise. So I love Mathematica. I wish I'd had it when I was in college.

>>

>> Having said that, I agree with many of David Park's comments, especially about the Front End and stability. I would like to see Wolfram Research pay attention to David's constructive criticisms.

>>

>> Adding my bit to David's comments about Front End stability, I'm one who complained about the new default notebook style that surfaced in Mathematica 9. The fonts and text color used in the new default style are ugly. Moreover, opening an old Mathematica 8 default notebook in Mathematica 9 changed all the fonts and colors to the new undesirable ones. I complained to Wolfram (apparently I was not alone) and a workaround was found by including the Mathematica 8 version notebook as a style that one has to select (as opposed to the default style). So Wolfram was responsive, but the lack of foresight remains: Why not reverse roles and include the new default notebook style as an option and keep the old one as default? This would address one aspect of stability that David eloquently describes.

>>

>> So yes, Wolfram, please pay attention to Mathematica and the effect changes on users, particularly the stability issue. Change is good, but it has to be done well.

>>

>> Raul Martinez

>>

>>

>

>