Re: New to MuPad


On Mon, 27 Jun 2005 11:15:03 +0200, Christopher Creutzig wrote:

How many MuPad programmers are there?, or were there in 2005 ?.
I was guessing: about 1 to 4.

It seems that MuPad is run-time-ishy (like BASIC).


> Actually, there is no syntax error and MuPAD behaves exactly as
>documented. It is just highly unlikely that the user realle meant what
....
I suppose I can't check since not having read your company manual on
syntax.
>
>> That MuPad ":=" is assigning a pointer to a code routine, it seems.
>
> I'm not sure what you mean by pointers. Functions are first-level
>objects in MuPAD [...]

It sounds like ":=" copies the code.

....
>> The last line would be evaluating "F" after "F" ceased to exist.
>
> In MuPAD, F would not have ceased to exist, since there is no static
>lifetime bound to scopes, but reference counting augmented with a
>garbage collector.
>

Reference counting can be slower, unnecessary, and buggy in
multitasking scenarios.
Other problems include:
* no weighting numbers inside of predicate-names.
* use of C or C++ or whatever the lagnuage was. It sounded like exactly
100.0% of the source code is available since the website said MuPad
was an "open" system, but inconsistently the MuPad Pro version was not
free.

....
>> Seems to be a mistake with "begin" being present in one and absent in
>> another.
>
> No, it's not. Read the documentation.
>
The documentation is hard to read, e.g. due to the red colour and spaces.

E.g. here is a bit from this MuPAD wepage:
http://research.mupad.de/doc/31/eng/tutorium_se32.html#x64-570006

------------------------------------------------------------------------
| The user can neither control nor extend the automatic simplifier.
|
| In most cases, however, MuPAD does not automatically simplify
| expressions. The reason is that the system generally cannot decide
| which is the most reasonable way of simplification. For example,
| consider the following expression which is not simplified:
|
| >> y := (-4*x + x^2 + x^3 - 4)*(7*x - 5*x^2 + x^3 - 3)
|
|
|
|
| </span></span></span></span></span></span></span></span></span>
| ( 3 2 ) -x - x + 4x +
| 4</span></span></span></span></span></span></span></span></span>
| ( 3 2 ) -x + 5 x - 7 x+ 3
|
| Naturally, you can expand this expression, which may be reasonable,
| for example, before computing its symbolic integral:
------------------------------------------------------------------------

There are lots of blank lines, red instead of dark blue, and some
"</span>" and "( 3 2 )" text. I used the [Mozilla and Firefox browsers].


....
>> replacing "::" with "." (imporove beauty)
>
> I'm not going to change a language that has been in use for ten years
>based on this argument. Besides, the . operator already has a clear
>meaning in MuPAD.

Maybe I'll discover the MuPad meaning for "." later.

Suppose you did want to change the syntax of MuPad: is an improvement
possible in that sort of circumstnace ?.

>> Ie. has MuPad been borrowed features of syntax, from C++?.
>
> The only syntax feature (apart from the things C++ obviously got from
>established notation in both mathematics and other languages, such as an
>infix + for addition or round parentheses for function application) is
>the :: operator. That may have been taken from C++ or something
>completely different ? I wasn't there back then.

One example of copying may suffice to show excessive copying from C++.

> I beg to differ. It's absolutely logical. Any number of operands can
>be enclosed in (). Any, including zero.

Whatever. I'd have to return to the original topic.

....

> Again, this has nothing to do with ?inherits.? There is a function
>named ?new? defined here, which would usually look something like
>
> new := (x, y) -> new(dom, [x, y], 0);
>
....
I shall avoid writing about the both important and not actually used,
syntactically pleasing, "new" name/token (not 'keyword').

> I am not going to discuss C++, but please learn about the meaning of
>MuPAD language features before arguing against them.

A MuPad download page is give below for me.

....
> Again, I do not understand what you are trying to say. Working at
>MuPAD is definitely not the best paid position I have been offered.
>Make of that whatever you like, I don't really care.

I can't make about anything worthwhile from that talk on cash flow rates
in Germany.

>
>> What was "->" doing in MuPAD?: [...]
>
> Just read the documentation, for goodness' sake!

>All you need for that is a web browser, please point it at
>www.mupad.de/doc/31/eng/stdlib.html.
>
>... you are definitely not qualified to judge its syntax.


> I do not ask you to start using MuPAD, so I have no obligation to show
>you why you should. All I am asking is to stop personal insults and to
>get your facts straight before criticising a system.

I criticised the syntax.

....
> I did not say that. (And something 100% available at our web site,
>included with the free versions and not sold as a separate product
>anywhere does not fill my understanding of the word ?pricey.?

So I made a mistake by saying MuPad was not free using guessing.

>be confusing MuPAD with something completely different.) What I said is
>that the doc should have more examples. That is a plain and simple way
>of acknowledging a bug, in my opinion.

A bug in what? software or documentation. The message started out with a
suggestion of no bug in the software (so I shall assume it was a bug in
the software, since reading documentation

....
> I do find the square brackets in Mathematica annoying. I have no idea
>whom you refer to who had problems with his parser.

Mr ancow65@xxxxxxxxx who I sent a message to on 26 June 2005. He didn't
reply so maybe the hihgly anonymous Mathematica programmer who
appeared in a 2003 thread between me and the student of LISP, has
lost interest in the e-mail address.

> f[1] has a meaning
>in MuPAD; if I remove that meaning, I can *of course* easily make it a
>function call, even without touching the parser at all. That's just
>definitely not the way I want to write a function call.
>
> Oh, and by the way: Sending copies to your posts as email is worse

I fixed that by giving an invalid SMTP server address in my Agent 2.0
Usenet client for Windows. It refuses to post news if no SMTP server,
and it automatically fills in the e-mail field with there seeming to be
no way to stop that.



Craig Carey.





.

Relevant Pages

  • Re: plotting data in MuPAD
    ... >> I find in the documentation that there is supposed to be something ... >> called Listplot, but when I try to use it, MuPAD says the command does ... plot::Listplot was introduced in MuPAD 3.1. ...
    (sci.math.symbolic)
  • Re: New to MuPad
    ... >> domain constructor in MuPad, and I can not get the 'new' proc to work. ... Ie. has MuPad been borrowed features of syntax, ... should kept (for run-time dereferencing) or simply eliminated (or maybe ...
    (sci.math.symbolic)
  • Re: mupad
    ... The fate of MuPAD is not quite obvious now. ... To the best of my knowledge, SciFace GmbH has at least 1 ... MuPAD please do not pay much attention by their public bug ...
    (sci.math.symbolic)
  • Re: MuPAD: TeX output bug?
    ... Zdenek Hurak wrote: ... > arithmetical expression. ... Why does MuPAD return wrong result and ... I have fixed the bug ...
    (sci.math.symbolic)
  • MuPAD bugs: solve -> Error: Domain attribute "_concat" missing;
    ... during evaluation of 'removeIdentsFromSolution' ... This bug has been introduced in MuPAD 3.0. ... It is absent in MuPAD 2.5.2. ...
    (sci.math.symbolic)