Heaviside simplify bug

Walter Roberson

unread,

Jul 6, 2006, 5:18:29 PM7/6/06

to

> eval(subs(x=2,y=7,Heaviside(x^2-y^2)));
0
> eval(subs(x=2,y=7,simplify(Heaviside(x^2-y^2))));
2
> simplify(Heaviside(x^2-y^2));
Heaviside(-x + y) + Heaviside(x + y)

The correct value (at least over reals) would be
(Heaviside(x+y) - Heaviside(x-y))^2
which can be simplified to
abs(Heaviside(x+y) - Heaviside(x-y))
--
All is vanity. -- Ecclesiastes

Message has been deleted

Vladimir Bondarenko

unread,

Jul 7, 2006, 1:11:53 AM7/7/06

to

................................................................

http://www.maplesoft.com/support/

"Maplesoft is committed to providing the highest
level of support for the products it sells"

................................................................

http://www.maplesoft.com/academic/students/

"using Maple in your studies is a terrific way
to get a head start in your career."

................................................................

http://groups.google.com/group/sci.math.symbolic/msg/dd5e6a02c428dad9?hl=en&

Maple bugs: A dangerous, misleading ad
statement at the main Maplesoft's site

................................................................

Walter Roberson writes on Fri, Jul 7 2006 12:18 am

http://groups.google.com/group/comp.soft-sys.math.maple/msg/2bd0c157c72f1d1a

................................................................

DESCRIPTION: Yet another Maple regression bug, one of many 1000s
regression bugs unfixed in many years.

Heaviside function is the step function

{ 0 x < 0
{ undefined x = 0
{ 1 0 < x

TEST CASE: eval(subs(x=0, y=1, Heaviside(x^2-y^2)) );
eval(subs(x=0, y=1, simplify(Heaviside(x^2-y^2))));

EXPECTED: 0
0

-------------------- (2005) Maple 10.00 -----------------------
0
2 <-------------------------------------------------------- BUG
-------------------- (2004) Maple 9.5.1 -----------------------
0
2 <-------------------------------------------------------- BUG
-------------------- (2004) Maple 9.5 -------------------------
0
2 <-------------------------------------------------------- BUG
-------------------- (2003) Maple 9 ---------------------------
0
2 <-------------------------------------------------------- BUG
-------------------- (2002) Maple 8 ---------------------------
0
2 <-------------------------------------------------------- BUG
-------------------- (2001) Maple 7 ---------------------------
0
2 <-------------------------------------------------------- BUG
-------------------- (2000) Maple 6 ---------------------------
0
2 <-------------------------------------------------------- BUG
-------------------- (1997) Maple V Rel 5 ---------------------
0
0
-------------------- (1995) Maple V Rel 4 ---------------------
0
0
-------------------- (1994) Maple V Rel 3 ---------------------
0
0
----------------------------------------------------------------

................................................................

Buy Maple, The World's Most Screwed Computer Algebra System

http://webstore.maplesoft.com/

Shell out!

................................................................

Best wishes,

Vladimir Bondarenko

VM and GEMM architect
Co-founder, CEO, Mathematical Director

http://www.cybertester.com/ Cyber Tester, LLC
http://maple.bug-list.org/ Maple Bugs Encyclopaedia
http://www.CAS-testing.org/ CAS Testing

................................................................

Walter Roberson writes on Fri, Jul 7 2006 12:18 am

http://groups.google.com/group/comp.soft-sys.math.maple/msg/2bd0c157c72f1d1a

WR> All is vanity. -- Ecclesiastes

http://groups.google.com/group/sci.math.symbolic/msg/9aabd559309afffb

................................................................

Preben Alsholm

unread,

Jul 7, 2006, 3:44:46 AM7/7/06

to

The value should be corrected to
(Heaviside(x+y) - Heaviside(-x+y))^2
or
abs(Heaviside(x+y) - Heaviside(-x+y))

Preben Alsholm

Phil Carmody

unread,

Jul 7, 2006, 4:04:42 AM7/7/06

to

"Emil" <Emil...@bulgaria.com> writes:
...
> Vladimir Bondarenko

Being killfiled by huge proportions of the intelligent
readership of a group was probably the best indication
you could have received that you were in fact a usenet
kook. Morphing because of that killfiling is certainly
proof of fact that you are as much a troll as a usenet
kook.

*_plonk_*, again.

Phil
--
The man who is always worrying about whether or not his soul would be
damned generally has a soul that isn't worth a damn.
-- Oliver Wendell Holmes, Sr. (1809-1894), American physician and writer

Preben Alsholm

unread,

Jul 7, 2006, 4:31:21 AM7/7/06

to

`simplify/Heaviside` does the following two steps

`piecewise/red`(Heaviside(x^2-y^2),x);
1 - Heaviside(x - y) + Heaviside(x + y)
`piecewise/red`(%,y);

Heaviside(-x + y) + Heaviside(x + y)

The first result is correct, the second is not.

Preben Alsholm

unread,

Jul 7, 2006, 5:01:58 AM7/7/06

to

Preben Alsholm wrote:
> `simplify/Heaviside` does the following two steps
>
> `piecewise/red`(Heaviside(x^2-y^2),x);
> 1 - Heaviside(x - y) + Heaviside(x + y)
> `piecewise/red`(%,y);
> Heaviside(-x + y) + Heaviside(x + y)
>
> The first result is correct, the second is not.
>

I take that back. The error occurs in the first step.

Preben Alsholm

Peter Luschny

unread,

Jul 7, 2006, 10:57:09 AM7/7/06

to

Vladimir Bondarenko writes:

TEST CASE: eval(subs(x=0, y=1, Heaviside(x^2-y^2)) );
eval(subs(x=0, y=1, simplify(Heaviside(x^2-y^2))));

-------------------- (2005) Maple 10.00 -----------------------BUG
-------------------- (2004) Maple 9.5.1 -----------------------BUG
-------------------- (2004) Maple 9.5 -------------------------BUG
-------------------- (2003) Maple 9 ---------------------------BUG
-------------------- (2002) Maple 8 ---------------------------BUG
-------------------- (2001) Maple 7 ---------------------------BUG
-------------------- (2000) Maple 6 ---------------------------BUG

-------------------- (1997) Maple V Rel 5 ---------------------

-------------------- (1995) Maple V Rel 4 ---------------------

-------------------- (1994) Maple V Rel 3 ---------------------

>Phil Carmody writes:

> Being killfiled by huge proportions of the intelligent
> readership of a group was probably the best indication
> you could have received that you were in fact a usenet
> kook. Morphing because of that killfiling is certainly
> proof of fact that you are as much a troll as a usenet
> kook.

However, for me Vladimir provides valuable information.

Being a Maple V Rel 5 user, I get informed, that I do not
have to worry about this error. Additionally, I see that
it might be of no good value to update to a later version.

Even for Maple Vladimirs list might be of value, since
it indicates for the umpteen times, that between Rel 5
and Rel 6 something terrible happened to the Maple core,
from which large parts of it still have not recovered.

These informations have some value. Other circumstances,
which you choosed to characterize by 'kook', 'troll',
'killfiling' etc., are best ignored in my opinion.

Peter

Vladimir Bondarenko

unread,

Jul 7, 2006, 12:55:06 PM7/7/06

to

DESCRIPTION: Yet another Maple regression bug, one of many 1000s

Heaviside function is the step function

{ 0 x < 0
{ undefined x = 0
{ 1 0 < x

TEST CASE: eval(subs(x=0, y=1, Heaviside(x^2+y^2)) );
eval(subs(x=0, y=1, simplify(Heaviside(x^2+y^2))));

EXPECTED: 1
1

--------------- (2005) Maple 10.00 ------------------------------
1
Error, (in Heaviside) not defined for non-real values
--------------- (2004) Maple 9.5.1 ------------------------------
1
Error, (in Heaviside) not defined for non-real values
--------------- (2004) Maple 9.5 --------------------------------
1
Error, (in Heaviside) not defined for non-real values
--------------- (2003) Maple 9 ----------------------------------
1
Error, (in Heaviside) not defined for non-real values
--------------- (2002) Maple 8 ----------------------------------
1
Error, (in Heaviside) not defined for non-real values
--------------- (2001) Maple 7 ----------------------------------
1
Error, (in Heaviside) not defined for non-real values
--------------- (2000) Maple 6 ----------------------------------
1
Error, (in Heaviside) not defined for non-real values
--------------- (1997) Maple V Rel 5 ----------------------------
1
Error, (in Heaviside) not defined for non-real values
--------------- (1995) Maple V Rel 4 ----------------------------
1
1
--------------- (1994) Maple V Rel 3 ----------------------------
1
1
------------------------------------------------------------------

Message has been deleted

Peter Luschny

unread,

Jul 7, 2006, 2:12:42 PM7/7/06

to

tim....@gmail.com wrote:

> Guess what, I just tried both of your test cases in Maple 10.04
> and they both returned 1.

So what? This does not contradict what Vladimir writes.

And how many users do you think work with 10.04?
And how many with older versions?

> So, this makes me wonder how many more of your "bugs" are really there.

Don't wonder, find it out! The case I could check works exactly
as Vladimir writes:

--------------- (1997) Maple V Rel 5 ----------------------------
> 1
> Error, (in Heaviside) not defined for non-real values

> I wish Google Groups had a killfile.

Why? Why always kill the person who brings the bad news?

Peter

Vladimir Bondarenko

unread,

Jul 7, 2006, 2:24:35 PM7/7/06

to

Phil Carmody, Nokia http://fatphil.org wrote:

http://groups.google.com/group/sci.math.symbolic/msg/35267924148c3021

Yet another beautiful theory killed by an ugly fact...

Had you were right, and we really wanted to sign in under
a new account, consciously, how much trouble would take to
us to generate, say, 1000 new accounts?

PC> intelligent readership

It's great to hear your opinion about your own IQ.

PC> *_plonk_*, again.

Could I propose you a bit more expressive version like

*___ppppllllloooooonnnnnkkkk!___*, aggggggain boom boom boom

?

PC> *_plonk_*, again.

Last but not least, are you, personally, a Maplesoft's customer?
Or your Maple (if you use it) is paid by Nokia, not out of your
own pocket?

Is your experience in programming and prime numbers a substitute
for quality assurance items?

If your goal was to shock my QA soul with your amazing software
testing achievements, please feel free to show them.

C W

unread,

Jul 7, 2006, 4:00:25 PM7/7/06

to

Shall we write a Vlad replacement software ?

Chris

Tim Lahey

unread,

Jul 7, 2006, 7:31:25 PM7/7/06

to

Peter Luschny wrote:
> tim....@gmail.com wrote:
>
> > Guess what, I just tried both of your test cases in Maple 10.04
> > and they both returned 1.
>
> So what? This does not contradict what Vladimir writes.
>

But he very conveniently left out the latest version and given that
it is a simple download for those who have licensed Maple 10,
I doubt that many people still use 10.0.0 which is the version he
compared to.

> And how many users do you think work with 10.04?
> And how many with older versions?

I guess that most people are either using Maple 9, 9.5 or 10.

>
> > So, this makes me wonder how many more of your "bugs" are really there.
>
> Don't wonder, find it out! The case I could check works exactly
> as Vladimir writes:

Most of the "bugs" he continues to post are for cases I don't care
about at all. Hell, even this one I don't care about, I just happened
to test it.
I just wondered because if he neglected 10.04 in this case, how many
other cases does he ignore particular versions.

>
> > I wish Google Groups had a killfile.
>
> Why? Why always kill the person who brings the bad news?
>

BTW, a killfile is a file that usenet reader software uses to
automatically hide posts by particular people so you don't have
to read their comments.

I would like a killfile because of his inflammatory comments he
likes to make. Witness the title of this post which has changed
from the original which was an honest query about a bug someone
found. I don't doubt that Maple has bugs, I've run across a few
myself, that doesn't make them liars. Why not report the bugs to
Maplesoft support like the rest of us rather than coming to the
newsgroup and making inflammatory comments. I read this group
to get advice, help, and new insights. None of his posts provide
that for me and they seem to clog up this and many other groups.

Walter Roberson

unread,

Jul 7, 2006, 9:21:33 PM7/7/06

to

In article <1152315085.0...@h48g2000cwc.googlegroups.com>,
Tim Lahey <tim....@gmail.com> wrote:

>But he very conveniently left out the latest version and given that
>it is a simple download for those who have licensed Maple 10,
>I doubt that many people still use 10.0.0 which is the version he
>compared to.

> interface(version);
Command-line Interface, Maple 10.03, X86 64 LINUX, Mar 24 2006 Build ID 222\
373

> simplify(Heaviside(x^2-y^2));

Heaviside(-x + y) + Heaviside(x + y)

So if it is fixed in 10.0.4 then it is quite a recent fix that people
might well not have installed yet.
--
I was very young in those days, but I was also rather dim.
-- Christopher Priest

Tim Lahey

unread,

Jul 7, 2006, 10:35:14 PM7/7/06

to

Walter,

I was actually replying to Vladimir's test. I thought I was running
10.04
but it turns out it is only 10.03. In fact, the same build as yours
just for
OS X.

Just so you know, for your problem. I got the following the first time
I tried
it on a just opened version of Maple 10.
> interface(version);
Standard Worksheet Interface, Maple 10.03, Mac OS X, March 25 2006
Build ID 222373
>eval(subs(x=2,y=7,Heaviside(x^2-y^2)));
0
>eval(subs(x=2,y=7,simplify(Heaviside(x^2-y^2))));
0
>simplify(Heaviside(x^2-y^2));
1 - Heaviside(x - y) - Heaviside(y + x)

However, with either your problem or Vladimir's if I issue a restart,
I get the same errors as reported before. Very curious.
Have you reported this to Maplesoft?
What happens if you just open Maple and run the commands?
I suggest you report it so it can be fixed in a future update.

Vladimir Bondarenko

unread,

Jul 8, 2006, 3:25:43 AM7/8/06

to

C W wrote:

CW> Shall we write a Vlad replacement software ?

Gee! ;) Your ideas are invariably fresh...

What are supposed to be the main functionalities?

Vlad

Vladimir Bondarenko

unread,

Jul 8, 2006, 4:09:13 AM7/8/06

to

Tim Lahey writes on Sat, Jul 8 2006 2:31 am:

http://groups.google.com/group/comp.soft-sys.math.maple/msg/97e0df05808e4c5b

TL> "Why not report the bugs to Maplesoft support?"

Yes, right you are!

Generally, this is a NECESSARY condition to get those DOZENS
THOUSANDS distinct Maple bugs fixed. But over-a-decade-long live
practice of (at least MANY DOZENS) Waterloo Maple/Maplesoft
customers shows with a shocking, dismaying clearness there are
it is NOT SUFFICIENT.

In a single sentence, "Maplesoft support" being a Maplesoft's
service, is a highly inefficient for Maple customers mechanism
directed often to quality assurance problem concealment, and
persuasion of the customers that Maple is far much better than
it is actually (read: it worthwhile to be acquired).

Not so infrequently, Maplesoft support provides the customers
(me, too, several times already, despite I do not write to the
Maplesoft too often) with formal answers reading which one thinks
involuntarily that the answering person simply either did not
read your letter at all, or was too busy (or unable?) to
understand it...

Maybe reading this message to the end would help you to see
why reporting the Maple bugs publicly is ESSENTIAL.

................................................................

Thomas D. Dean writes on Mon, Nov 10 2003 6:51 pm

http://groups.google.com/group/comp.soft-sys.math.maple/msg/3d1cc919649da7b6

Subj: Silly Answer from Maplesoft

I received a silly answer from Maplesoft in response
to a simple problem. Seems like they did not read my
problem report. ...

................................................................

Carl DeVore, a Maple Guru, writes on Sat, Oct 5 2002 2:57 pm

http://groups.google.com/group/comp.soft-sys.math.maple/msg/afe3c2f7efd63f62

Subj: explicit solution of polynomial and bug (?)

These are good in the sense that the information is thus put
in the public record. MUG and newsgroups are archived at
several sites around the world and are easily searchable with
standard search engines. ...

> sup...@maplesoft.com

Send your bug reports there AND post them here.

................................................................

http://groups.google.com/group/sci.math.symbolic/msg/6002b726f9f2e669?hl=en
Subj: Maple bugs: Big festal day

...

We celebrate 15 years since a bug in Maple was reported *publicly* !

...

................................................................

John Harper writes on Thu, Jul 22 2004 1:47 am

http://groups.google.com/group/comp.soft-sys.math.maple/msg/677e7afc81e36314

Subj: asympt(hypergeom(...)) problem

Five years ago I complained to Waterloo Maple Inc.
that Maple V.5 couldn't do asymptotic expansions of
hypergeometric functions, though V.4 could if they
reduced to more elementary functions. This problem
has still not been fixed! ...

................................................................

Carl DeVore writes on Tue Oct 1, 2002 4:43 pm

http://groups.yahoo.com/group/maple8/message/111
Subj: Bug tracking system

Some users are getting very frustrated with the lack of public
information about bugs.

I believe that if nothing is done to address this issue, then
Maple will quickly become a system which is only useful for
teaching calculus and other low-level courses. It will be useless
for research.

................................................................

Vladimir Bondarenko writes on Wed, Jun 15 2005 12:28 am

http://groups.google.com/group/sci.math.symbolic/msg/9340ac393e83c4d8

In a nutshell, one of the final goals is making all the Maple's
consumers' life easier via making Maple faster, more stable and
so to say, more intelligent.

So, one of the goals is to HELP Maplesoft to improve Maple if
the company wants this, and to FORCE Maplesoft to improve Maple
if the company does not wants this. The maximal public Maple bugs
exposure is a tool of our users' influence on the wistful case.

................................................................

Vladimir Bondarenko

unread,

Jul 8, 2006, 11:26:50 PM7/8/06

to

................................................................

http://www.maplesoft.com/students/index.aspx

"using Maple in your studies is a terrific way
to get a head start in your career."

http://www.maplesoft.com/company/reporter/archive.aspx?d=120

"the most powerful and trusted computation engine available."

http://www.maplesoft.com/products/maple/pastversions/pastversions_maple95.aspx?D=11

"Maple engine, the most trusted software for exploring mathematics"

................................................................

DESCRIPTION: The value of the limit depends on the point
where the limit is taken.

TEST CASE: limit(product(sin(k/n), k=1..n), n = 1);

EXPECTED: sin(1) # best
limit(product(sin(k/n),k=1..n),n=1) # tolerable

-------------------- (2005) Maple 10.00 -----------------------

exp(sum(ln(sin(k)),k = 1 .. n))

-------------------- (2004) Maple 9.5.1 -----------------------

exp(sum(ln(sin(k)),k = 1 .. n))

-------------------- (2004) Maple 9.5 -------------------------

exp(sum(ln(sin(k)),k = 1 .. n))

-------------------- (2003) Maple 9 ---------------------------

exp(sum(ln(sin(k)),k = 1 .. n))

-------------------- (2002) Maple 8 ---------------------------

exp(sum(ln(sin(k)),k = 1 .. n))

-------------------- (2001) Maple 7 ---------------------------

limit(product(sin(k/n),k = 1 .. n),n = 1)

-------------------- (2000) Maple 6 ---------------------------

limit(product(sin(k/n),k = 1 .. n),n = 1)

-------------------- (1997) Maple V Rel 5 ---------------------

limit(product(sin(k/n),k = 1 .. n),n = 1)

-------------------- (1995) Maple V Rel 4 ---------------------

limit(product(sin(k/n),k = 1 .. n),n = 1)

-------------------- (1994) Maple V Rel 3 ---------------------

sin(1)
----------------------------------------------------------------

COMPARE: limit(product(sqrt(k/n), k=1..n), n = 1);

-------------------- (2005) Maple 10.00 -----------------------

1

-------------------- (2004) Maple 9.5.1 -----------------------

1

-------------------- (2004) Maple 9.5 -------------------------

1

-------------------- (2003) Maple 9 ---------------------------

1

-------------------- (2002) Maple 8 ---------------------------

1

-------------------- (2001) Maple 7 ---------------------------

1

-------------------- (2000) Maple 6 ---------------------------

1

-------------------- (1997) Maple V Rel 5 ---------------------

1

-------------------- (1995) Maple V Rel 4 ---------------------

1

-------------------- (1994) Maple V Rel 3 ---------------------

1
----------------------------------------------------------------

................................................................

The time is not come yet to show the next layer of our results.

Robert Israel

unread,

Jul 9, 2006, 4:09:11 AM7/9/06

to

In article <e8n1at$lv9$1...@canopus.cc.umanitoba.ca>,

Walter Roberson <robe...@ibd.nrc-cnrc.gc.ca> wrote:
>In article <1152315085.0...@h48g2000cwc.googlegroups.com>,
>Tim Lahey <tim....@gmail.com> wrote:
>
>>But he very conveniently left out the latest version and given that
>>it is a simple download for those who have licensed Maple 10,
>>I doubt that many people still use 10.0.0 which is the version he
>>compared to.
>
>> interface(version);
>Command-line Interface, Maple 10.03, X86 64 LINUX, Mar 24 2006 Build ID 222\
> 373
>
>> simplify(Heaviside(x^2-y^2));
> Heaviside(-x + y) + Heaviside(x + y)
>
>So if it is fixed in 10.0.4 then it is quite a recent fix that people
>might well not have installed yet.

This is still present in 10.04.

Robert Israel isr...@math.ubc.ca
Department of Mathematics http://www.math.ubc.ca/~israel
University of British Columbia Vancouver, BC, Canada

Mate

unread,

Jul 10, 2006, 8:29:05 AM7/10/06

to

Vladimir Bondarenko wrote:
> ................................................................
>
> DESCRIPTION: The value of the limit depends on the point
> where the limit is taken.
>
> TEST CASE: limit(product(sin(k/n), k=1..n), n = 1);
>
> EXPECTED: sin(1) # best
> limit(product(sin(k/n),k=1..n),n=1) # tolerable

Your expected value is not correct, because the side limits for n --> 1
are not equal:
limit to the right = sin(1)
limit to the left = 1.
You shoult "limit" yourself to the real Maple bugs,
or have you discovered all of them?

Mate

Vladimir Bondarenko

unread,

Jul 10, 2006, 1:06:38 PM7/10/06

to

Mate writes on Mon, Jul 10 2006 3:29 pm

http://groups.google.com/group/comp.soft-sys.math.maple/msg/08cceec73e4aaf4d

Thanks for your comment.

You mean this?

product(sin(k/1), k=1..0);
product(sin(k/1), k=1..1);

1
sin(1)

OK. Also, the answer

limit(product(sin(k/n), k=1..n), n = 1)

would not be too bad, as I already pointed out. But in fact,
in the given case, I do not feel it's really material what
is the expected answer.

The main point is that Maple 8 - Maple 9 - Maple 10 return
a piece of total bunk, exp(sum(ln(sin(k)),k = 1 .. n)) .

M> You shoult "limit" yourself to the real Maple bugs,

Take it or leave it, but right away I am the # 1 world's
QA engineer in this field.

http://groups.google.com/group/sci.math.symbolic/msg/5123d38af7c24ad3

This said, please realize that the showed publicly stuff
is a far cry from my last word in the field.

M> You shoult "limit" yourself to the real Maple bugs,

Jocularly speaking, with the course of time, I decided that
I was too lazy to discover Maple bugs.

This is why I invented the VM machine we keep implementing.

But of course, this explanation shows only one of the
aspects of why I decided to do it; now it's still ahead
of time to discuss this.

M> or have you discovered all of them?

The VM machine has already discovered far many more
distinct Maple bugs than we the human being have done
and posted to public domain, altogether.

However, the shower of distinct Maple bug manifestations
is not getting weaker...

One fine day, not tomorrow, we will come back to this
point.

Mate

unread,

Jul 10, 2006, 1:30:45 PM7/10/06

to

Vladimir Bondarenko wrote:

> Take it or leave it, but right away I am the # 1 world's
> QA engineer in this field.
>

What about # 0 ?

Vladimir Bondarenko

unread,

Jul 10, 2006, 8:21:42 PM7/10/06

to

Mate, I am really sorry to clog the groups even with such a tiny
remark about you, but over my career, I have met about a dozen
suckers like you who had no qualification in QA whatsoever, and
never measured their IQ.

But where those guys are now...

Dave (from the UK)

unread,

Jul 11, 2006, 6:49:28 AM7/11/06

to

Vladimir Bondarenko wrote:

> Buy Maple, The World's Most Screwed Computer Algebra System

As someone who lurks around here sometimes could you explain to me what
your interest in Maple is? If you feel its the worlds most screwed
computer algebra system why don't use just ignore it

Or perhaps use my CAS below which is bug free - always computing the
result exactly with no bugs. (Assuming your compiler or operating system
are not broken of course).

Here is the source code - feel free to copy and redistribute if you wish.

#include <stdio.h>
int main(int argc, char **argv)
{
printf("1 + 1 = 2\n");
return(0);
}

I just downloaded a scientific calculator for my Pocket PC and found it
was very poor indeed. But I don't feel the need to make a life-long
hobby out of pointing out how bad it is to others. I just delete it and
move on.

--
Dave K MCSE.

MCSE = Minefield Consultant and Solitaire Expert.

Please note my email address changes periodically to avoid spam.
It is always of the form: month-year@domain. Hitting reply will work
for a couple of months only. Later set it manually.

http://witm.sourceforge.net/ (Web based Mathematica front end)

Leo

unread,

Jul 14, 2006, 3:58:42 AM7/14/06

to

Vladimir Bondarenko wrote:
> Mate, I am really sorry to clog the groups even with such a tiny
> remark about you, but over my career, I have met about a dozen
> suckers like you who had no qualification in QA whatsoever, and
> never measured their IQ.
>
> But where those guys are now...
>

Getting laid?

Vladimir Bondarenko

unread,

Jul 15, 2006, 10:48:11 PM7/15/06

to

............................................................

http://www.maplesoft.com/

"The Maple 10 interactive environment is the easiest way to
harness the full power of mathematics"

............................................................

TEST CASE: sum(n!, n= 10^5..infinity);

EXPECTED: infinity # in 0 second

CHECKUP: evalf(Int(n!, n= 10^5..infinity));

Float(infinity)

-------------------- (2005) Maple 10.00 -------------------
infinity 1351.874 seconds
-------------------- (2004) Maple 9.5.1 -------------------
infinity 1156.162 seconds
-------------------- (2004) Maple 9.5 ---------------------
infinity 1095.195 seconds
-------------------- (2003) Maple 9 -----------------------
infinity 1018.174 seconds
-------------------- (2002) Maple 8 -----------------------
infinity 1180.948 seconds
-------------------- (2001) Maple 7 -----------------------
infinity 738.441 seconds
-------------------- (2000) Maple 6 -----------------------
GAMMA(100001)*hypergeom([1, 100001],[],1) 737.079 seconds
-------------------- (1997) Maple V Rel 5 -----------------
Error, (in expand/factorial) object too large 6.900 second
-------------------- (1995) Maple V Rel 4 -----------------
Error, (in expand/factorial) object too large 10.491 seconds
-------------------- (1994) Maple V Rel 3 -----------------
sum(n!,n = 100000 .. infinity) 0 second
------------------------------------------------------------

COMPARE: Timing[Sum[n!, {n, 10^5, Infinity}]]

Sum::div: Sum does not converge

{0.01 Second, Sum[n!, {n,100000,Infinity}]}

............................................................

http://www.wolfram.com/technology/guide/

Wolfram Technology Guide explains important, often unique
Mathematica technologies and relates them to practical
advantages for users

............................................................

http://www.wolfram.com/products/mathematica/trial.cgi

Get a fully functional, 15-day, save-disabled trial version
of Mathematica 5.2.

Macintosh
Windows
Linux x86/x86-64bit
HP Tru64 Unix
HP-UX
IBM AIX
Linux-Itanium
SGI IRIX
Solaris UltraSparc
Solaris x86

............................................................

Vladimir Bondarenko

unread,

Jul 17, 2006, 3:58:36 AM7/17/06

to

Mr Leo Lazauskas, masters student in applied mathematics,
The University of Adelaide, Australia

http://www.maths.adelaide.edu.au/applied/graduates/RGradStudents/llazausk.html

wrote on Fri, Jul 14 2006 10:58 am

http://groups.google.com/group/sci.math.symbolic/msg/a3b5778365c087be

My answer is

http://web.tickle.com/tests/uiq/index-pop.jsp?sid=&supp=&z=

Vladimir Bondarenko

unread,

Jul 17, 2006, 4:58:40 AM7/17/06

to

............................................................

http://www.maplesoft.com/

Being a testing amateur, we

"were totally blown away by Maplesoft's Maple 10
mathematical package"

because we have not enough time or care to read
the postings by the Cyber Tester's engineers, and
thus we are not able to realize the simple truth
that there are many dozens of thousands distinct
Maple bugs including but not limited to many 1000s
of (multiple) regression bugs.

Really, why care, we are a writer, not a reader.

............................................................

TEST CASE: dsolve({diff(y(z),z$2)= y(z)/z,
y(0) = 0, D(y)(0)=1},y(z));

EXPECTED: y(z) = z^(1/2)*BesselI(1,2*z^(1/2))

CHECKUP: ode := diff(y(z),z$2)= y(z)/z:
ics := y(0)=1, D(y)(0)=1:
sol := y(z) = z^(1/2)*BesselI(1,2*z^(1/2)):
odetest(sol, [ode, ics]);

[0, 1, 0]

-------------------- (2005) Maple 10.00 -------------------

NULL

-------------------- (2004) Maple 9.5.1 -------------------

y(z) = z^(1/2)*BesselI(1,2*z^(1/2))

-------------------- (2004) Maple 9.5 ---------------------

y(z) = z^(1/2)*BesselI(1,2*z^(1/2))

-------------------- (2003) Maple 9 -----------------------

y(z) = z^(1/2)*BesselI(1,2*z^(1/2))

-------------------- (2002) Maple 8 -----------------------

y(z) = z^(1/2)*BesselI(1,2*z^(1/2))

-------------------- (2001) Maple 7 -----------------------

y(z) = z^(1/2)*BesselI(1,2*z^(1/2))

-------------------- (2000) Maple 6 -----------------------

y(z) = z^(1/2)*BesselI(1,2*z^(1/2))

-------------------- (1997) Maple V Rel 5 -----------------

y(z) = z^(1/2)*BesselI(1,2*z^(1/2))

-------------------- (1995) Maple V Rel 4 -----------------

y(z) = z^(1/2)*BesselI(1,2*z^(1/2))

-------------------- (1994) Maple V Rel 3 -----------------

NULL
------------------------------------------------------------

COMPARE: Mathematica 5.2

DSolve[{D[y[z],{z,2}]==y[z]/z, y[0]==0, y'[0]==1}, y[z],z]

MuPAD 4.0

solve(ode({diff(y(z),z$2)= y(z)/z, y(0)=0, y'(0)=1},y(z)));

{z^(1/2)*besselI(1, 2*z^(1/2))}

............................................................

Man + Machine Review Of Maple Crisis

http://maple.bug-list.org/maple-crisis.php

............................................................

Thomas Richard

unread,

Jul 17, 2006, 8:16:25 AM7/17/06

to

"Vladimir Bondarenko" <v...@cybertester.com> wrote:

> Being a testing amateur, we
>

> TEST CASE: dsolve({diff(y(z),z$2)= y(z)/z,
> y(0) = 0, D(y)(0)=1},y(z));
>
> EXPECTED: y(z) = z^(1/2)*BesselI(1,2*z^(1/2))
>
> CHECKUP: ode := diff(y(z),z$2)= y(z)/z:
> ics := y(0)=1, D(y)(0)=1:
> sol := y(z) = z^(1/2)*BesselI(1,2*z^(1/2)):
> odetest(sol, [ode, ics]);
>
> [0, 1, 0]

There is a typo in your first IC. See odetest output.

> -------------------- (2005) Maple 10.00 -------------------
> NULL

Works correctly in Maple 10.04.

[off topic newsgroups removed from list]

--
Thomas Richard
Maple Support
Scientific Computers GmbH
http://www.scientific.de

Vladimir Bondarenko

unread,

Jul 17, 2006, 11:05:55 AM7/17/06

to

............................................................

Vladimir Bondarenko writes on Mon, Jul 17 2006 11:58 am
http://groups.google.com/group/comp.soft-sys.math.maple/msg/0555e1f2818779e2

Thomas Richard writes on Mon, Jul 17 2006 3:16 pm
http://groups.google.com/group/comp.soft-sys.math.maple/msg/4b77668b6b992847

............................................................

A typo is corrected and the data is extended due to
Herr Thomas Richard. We observe here one of several
typical kinds of Maple quality destruction we called
"spike".

............................................................

TEST CASE: dsolve({diff(y(z),z$2)= y(z)/z,
y(0) = 0, D(y)(0)=1},y(z));

EXPECTED: y(z) = z^(1/2)*BesselI(1,2*z^(1/2))

CHECKUP: ode := diff(y(z),z$2)= y(z)/z:

ics := y(0)=0, D(y)(0)=1:

sol := y(z) = z^(1/2)*BesselI(1,2*z^(1/2)):
odetest(sol, [ode, ics]);

[0, 0, 0]

-------------------- (2006) Maple 10.04 -------------------
y(z) = z^(1/2)*BesselI(1,2*z^(1/2))
-------------------- (2005) Maple 10.01 -------------------
NULL <------------------------------------------------ BUG

Vladimir Bondarenko

unread,

Jul 17, 2006, 11:22:19 AM7/17/06

to

............................................................

http://www.maplesoft.com/

"were totally blown away by Maplesoft's Maple 10
mathematical package"

............................................................

Reads:

http://www.maplesoft.com/

[Being a testing amateur, we]

"were totally blown away by Maplesoft's Maple 10
mathematical package"

[because we have not enough time or care to read
the postings by the Cyber Tester's engineers, and
thus we are not able to realize the simple truth
that there are many dozens of thousands distinct
Maple bugs including but not limited to many 1000s
of (multiple) regression bugs.

Really, why care, we are a writer, not a reader.

So, dear Editor, pls send us our fee asap.]

............................................................

DESCRIPTION: The solution to a boundary value problem
for an ODE of the second order depends
on a free parameter _C2.

Only Maple 6 of 2000 can solve this BVP
correctly.

TEST CASE: dsolve({diff(y(z),z$2)= y(z)/z^3,
y(0)= 0, y(1)= 1}, y(z));

EXPECTED: y(z) = 1/BesselY(1,-2*I)*z^(1/2)*
BesselY(1,-2*I/z^(1/2))

CHECKUP: ode := diff(y(z),z$2)= y(z)/z^3:
ics := y(0)= 0, y(1)= 1:
sol := y(z) = 1/BesselY(1,-2*I)*z^(1/2)*
BesselY(1,-2*I/z^(1/2)):
odetest(sol, [ode, ics]);

[0, 0, 0]

-------------------- (2005) Maple 10.00 -------------------
y(z) = -(_C2*BesselY(1,-2*I)-1)/BesselI(1,2)*z^(1/2)*BesselI
(1,2/z^(1/2))+_C2*z^(1/2)*BesselY(1,-2*I/z^(1/2))

-------------------- (2004) Maple 9.5.1 -------------------

y(z) = -(_C2*BesselY(1,-2*I)-1)/BesselI(1,2)*z^(1/2)*BesselI
(1,2/z^(1/2))+_C2*z^(1/2)*BesselY(1,-2*I/z^(1/2))

-------------------- (2004) Maple 9.5 ---------------------

y(z) = -(_C2*BesselY(1,-2*I)-1)/BesselI(1,2)*z^(1/2)*BesselI
(1,2/z^(1/2))+_C2*z^(1/2)*BesselY(1,-2*I/z^(1/2))

-------------------- (2003) Maple 9 -----------------------

y(z) = -(_C2*BesselY(1,-2*I)-1)/BesselI(1,2)*z^(1/2)*BesselI
(1,2/z^(1/2))+_C2*z^(1/2)*BesselY(1,-2*I/z^(1/2))

-------------------- (2002) Maple 8 -----------------------

y(z) = -(_C2*BesselY(1,-2*I)-1)/BesselI(1,2)*z^(1/2)*BesselI
(1,2/z^(1/2))+_C2*z^(1/2)*BesselY(1,-2*I/z^(1/2))

-------------------- (2001) Maple 7 -----------------------

y(z) = -(_C2*BesselY(1,-2*I)-1)/BesselI(1,2)*z^(1/2)*BesselI
(1,2/z^(1/2))+_C2*z^(1/2)*BesselY(1,-2*I/z^(1/2))

-------------------- (2000) Maple 6 -----------------------

y(z) = 1/BesselY(1,-2*I)*z^(1/2)*BesselY(1,-2*I/z^(1/2))

-------------------- (1997) Maple V Rel 5 -----------------

y(z) = undefined

-------------------- (1995) Maple V Rel 4 -----------------

NULL

-------------------- (1994) Maple V Rel 3 -----------------
NULL
------------------------------------------------------------

............................................................

Thomas Richard

unread,

Jul 17, 2006, 1:09:01 PM7/17/06

to

"Vladimir Bondarenko" <v...@cybertester.com> wrote:

> DESCRIPTION: The solution to a boundary value problem
> for an ODE of the second order depends
> on a free parameter _C2.

That should not happen, of course.

> Only Maple 6 of 2000 can solve this BVP
> correctly.
>
> TEST CASE: dsolve({diff(y(z),z$2)= y(z)/z^3,
> y(0)= 0, y(1)= 1}, y(z));
>
> EXPECTED: y(z) = 1/BesselY(1,-2*I)*z^(1/2)*
> BesselY(1,-2*I/z^(1/2))

I don't think that's what the user expects. This y(z) will be complex
for 0<z<1, which odetest does not check for, AFAIK.
Call the above expression csol, then plot(Im(csol),z=0..1,y=-10..10);

I'd be interested in a symbolic solution that is real-valued. Compare:
rsol:=dsolve({diff(y(z),z$2)=y(z)/z^3,y(0)=0,y(1)=1},y(z),numeric,method=bvp[midrich]);
plots[odeplot](rsol,z=0..1);

What do other systems return?

Robert Israel

unread,

Jul 17, 2006, 6:47:09 PM7/17/06

to

In article <1153149739.1...@b28g2000cwb.googlegroups.com>,
Vladimir Bondarenko <v...@cybertester.com> wrote:

>DESCRIPTION: The solution to a boundary value problem
> for an ODE of the second order depends
> on a free parameter _C2.

> Only Maple 6 of 2000 can solve this BVP
> correctly.

>TEST CASE: dsolve({diff(y(z),z$2)= y(z)/z^3,
> y(0)= 0, y(1)= 1}, y(z));

The DE has an irregular singular point at z=0, so specifying a boundary
condition at z=0 is a bad idea in general. dsolve is not sophisticated
enough to handle something like this.

>EXPECTED: y(z) = 1/BesselY(1,-2*I)*z^(1/2)*
> BesselY(1,-2*I/z^(1/2))

This does not go to 0 as z -> 0+. The real part
goes to +infinity and the imaginary part to -infinity.
Try this:

> y(z) = limit(rhs(dsolve({de,y(p)=0,y(1)=1})),p=0,right);

y(z) = (BesselI(1,2/z^(1/2))+BesselY(1,-2*I/z^(1/2)))/
(BesselI(1,2)+BesselY(1,-2*I))*z^(1/2)

Vladimir Bondarenko

unread,

Jul 18, 2006, 7:26:47 AM7/18/06

to

................................................................

http://www.maplesoft.com/support/

"Maplesoft is committed to providing the highest level
of support for the products it sells"

................................................................

Walter Roberson writes on Mon, Jul 17 2006 11:40 pm

http://groups.google.com/group/comp.soft-sys.math.maple/msg/91e587e4f51ba18f

[ ... ]

WR> First, notice that the 'local k' has disappeared, leaving
WR> k as a reference to the global variable k.

WR> Second notice that the expression has been moved out of
WR> add() and assigned to the new local variable t4. Local
WR> variables have first-evaluation semantics, so the variable
WR> is not re-evaluated for each of add()'s iteration over k.

[ ... ]

................................................................

DESCRIPTION: None of Maple versions since 1997 on can
generate the required trivial code correctly.

TEST CASE: foo := proc (p,q) local k; add((p[k]-q[k])^2,
k=1..2) end:
fooo := codegen[optimize](foo):
fooo(a,b);

EXPECTED: (a[1]-b[1])^2+(a[2]-b[2])^2

-------------------- (2005) Maple 10.00 -----------------------

2*(a[k]-b[k])^2

-------------------- (2004) Maple 9.5.1 -----------------------

2*(a[k]-b[k])^2

-------------------- (2004) Maple 9.5 -------------------------

2*(a[k]-b[k])^2

-------------------- (2003) Maple 9 ---------------------------

2*(a[k]-b[k])^2

-------------------- (2002) Maple 8 ---------------------------

2*(a[k]-b[k])^2

-------------------- (2001) Maple 7 ---------------------------

2*(a[k]-b[k])^2

-------------------- (2000) Maple 6 ---------------------------

2*(a[k]-b[k])^2
-------------------- (1997) Maple V Rel 5 ---------------------
Warning, `k` in call to `add` is not local
2*(a[k]-b[k])^2
-------------------- (1995) Maple V Rel 4 ---------------------
N/A
-------------------- (1994) Maple V Rel 3 ---------------------
N/A
----------------------------------------------------------------

................................................................

Other Maple codegen bugs

http://groups.google.com/groups/search?q=Maple+codegen+bug&qt_s=Search

................................................................

Vladimir Bondarenko

unread,

Jul 18, 2006, 11:20:22 PM7/18/06

to

................................................................

http://www.maplesoft.com/support/

"Maplesoft is committed to providing the highest level
of support for the products it sells"

................................................................

Sez you!

A Maplesoft customer, enjoy yet another bug the long-liver.

................................................................

Walter Roberson writes on Tue, Jul 18 2006 8:22 pm

http://groups.google.com/group/comp.soft-sys.math.maple/msg/aac55b559c9b178e

[ ... ]

WR> Notice that the saved text version of foo has lost the
WR> delayed evaluation quotes, so when it is read back, the
WR> procedure is evaluated -once- and the single numeric
WR> result is assigned to foo, instead of the entire delay
WR> expression being assigned to foo.

WR> This problem does not happen when one saves in internal
WR> format (e.g., save foo, "/tmp/foo.m" ). Unfortunately,
WR> internal format is not so easy for a human to read...

[Oh yes. -- VB]

[ ... ]

................................................................

DESCRIPTION: None of Maple versions since 1994 on can save
the data correctly.

TEST CASE: _FOO := 0:
foo := 'proc() global _FOO; _FOO:=_FOO+1 end()':
foo; foo;
save foo, `/tmp/foo.maple`;
read `/tmp/foo.maple`;
foo; foo;

EXPECTED: 1 2 3 4

-------------------- (2005) Maple 10.00 -----------------------

1 2 foo := 3 3 3

-------------------- (2004) Maple 9.5.1 -----------------------

1 2 foo := 3 3 3

-------------------- (2004) Maple 9.5 -------------------------

1 2 foo := 3 3 3

-------------------- (2003) Maple 9 ---------------------------

1 2 foo := 3 3 3

-------------------- (2002) Maple 8 ---------------------------

1 2 foo := 3 3 3

-------------------- (2001) Maple 7 ---------------------------

1 2 foo := 3 3 3

-------------------- (2000) Maple 6 ---------------------------

1 2 foo := 3 3 3

-------------------- (1997) Maple V Rel 5 ---------------------

1 2 foo := 3 3 3

-------------------- (1995) Maple V Rel 4 ---------------------

1 2 foo := 3 3 3

-------------------- (1994) Maple V Rel 3 ---------------------

1 2 foo := 3 3 3
----------------------------------------------------------------

................................................................

Maple bugs: An unknown elegant bug discovered
by Jean-Michel Collard

http://groups.google.com/group/sci.math.symbolic/msg/b4fb82dd12304dd6

................................................................

Vladimir Bondarenko

unread,

Jul 21, 2006, 1:32:41 AM7/21/06

to

..............................................................

http://www.maplesoft.com/support/

"Maplesoft is committed to providing the highest
level of support for the products it sells"

..............................................................

Dear Maple customer, please eat yet another
Maple regression bug, one of many 1000s.

Bon appetite!

..............................................................

DESCRIPTION: Only Maple V Rel 4 of 1995 and Maple V Rel 5
of 1997 can calculate this limit correctly.

TEST CASE: limit(1/sqrt(1+z^(z^z)), z= -1);

EXPECTED: undefined

CHECKUP: limit(1/sqrt(1+z^(z^z)), z= -1.001);
limit(1/sqrt(1+z^(z^z)), z= -0.999);

9.335771067 + 1.323435694*I
1.312891021 - 9.280839335*I

-------------------- (2005) Maple 10.00 ---------------------
limit(1/(1+z^(z^z))^(1/2),z = -1)
-------------------- (2004) Maple 9.5.1 ---------------------
limit(1/(1+z^(z^z))^(1/2),z = -1)
-------------------- (2004) Maple 9.5 -----------------------
limit(1/(1+z^(z^z))^(1/2),z = -1)
-------------------- (2003) Maple 9 -------------------------
limit(1/(1+z^(z^z))^(1/2),z = -1)
-------------------- (2002) Maple 8 -------------------------
limit(1/(1+z^(z^z))^(1/2),z = -1)
-------------------- (2001) Maple 7 -------------------------
limit(1/(1+z^(z^z))^(1/2),z = -1)
-------------------- (2000) Maple 6 -------------------------
limit(1/(1+z^(z^z))^(1/2),z = -1)

-------------------- (1997) Maple V Rel 5 -------------------

undefined

-------------------- (1995) Maple V Rel 4 -------------------

undefined

-------------------- (1994) Maple V Rel 3 -------------------

limit(1/(1+z^(z^z))^(1/2),z = -1)
--------------------------------------------------------------

..............................................................

First world's man+machine based Maple Crisis review (beta 0.1)

http://maple.bug-list.org/maple-crisis.php

(... beta 0.2 is already of 400+ pages ...)

..............................................................

Vladimir Bondarenko

unread,

Jul 21, 2006, 7:17:51 PM7/21/06

to

................................................................

http://www.maplesoft.com/support/

"Maplesoft is committed to providing the highest
level of support for the products it sells"

................................................................

Sez you!

................................................................

DESCRIPTION: Only Maple V Release 5 of 1997 can approximate
this integral correctly.

TEST CASE: int(1/(exp(1.0*z^2)+1), z= 0..1);

EXPECTED: .4194664800

CHECKUP: evalf(Int(1/(exp(1.0*z^2)+1), z= 0..1));

.4194664800

--------------- (2005) Maple 10.00 ----------------------------
int(1/(exp(1.0*z^2)+1),z = 0 .. 1)
--------------- (2004) Maple 9.5.1 ----------------------------
int(1/(exp(1.0*z^2)+1),z = 0 .. 1)
--------------- (2004) Maple 9.5 ------------------------------
int(1/(exp(1.0*z^2)+1),z = 0 .. 1)
--------------- (2003) Maple 9 --------------------------------
int(1/(exp(1.0*z^2)+1),z = 0 .. 1)
--------------- (2002) Maple 8 --------------------------------
int(1/(exp(1.0*z^2)+1),z = 0 .. 1)
--------------- (2001) Maple 7 --------------------------------
int(1/(exp(1.0*z^2)+1),z = 0 .. 1)
--------------- (2000) Maple 6 --------------------------------
int(1/(exp(1.0*z^2)+1),z = 0 .. 1)
--------------- (1997) Maple V Rel 5 --------------------------
.4194664800
--------------- (1995) Maple V Rel 4 --------------------------
int(1/(exp(z^2)+1.),z = 0 .. 1.)
--------------- (1994) Maple V Rel 3 --------------------------
int(1/(exp(z^2)+1.),z = 0 .. 1.)
----------------------------------------------------------------

................................................................

First world's Man+Machine Based Maple Crisis Review (beta 0.1)

http://maple.bug-list.org/maple-crisis.php

(... beta 0.2 is already of 400+ pages ...)

................................................................

Robert Israel

unread,

Jul 21, 2006, 8:21:39 PM7/21/06

to

In article <1153523871.3...@75g2000cwc.googlegroups.com>,
Vladimir Bondarenko <v...@cybertester.com> wrote:

>
>DESCRIPTION: Only Maple V Release 5 of 1997 can approximate
> this integral correctly.
>
>TEST CASE: int(1/(exp(1.0*z^2)+1), z= 0..1);
>
>EXPECTED: .4194664800
>
>CHECKUP: evalf(Int(1/(exp(1.0*z^2)+1), z= 0..1));
>
> .4194664800
>
>
>--------------- (2005) Maple 10.00 ----------------------------
>int(1/(exp(1.0*z^2)+1),z = 0 .. 1)

What nonsense is this? "int" is for symbolic integration, not
numerical approximation. Not surprisingly, Maple does not have
a closed form for this integral, so it leaves it unevaluated.
If you want an approximation, use evalf.

BTW, this integral can also be expressed as a series

sqrt(pi)/2 sum_{n=1}^infty (-1)^(n-1) erf(sqrt(n))/sqrt(n)

Vladimir Bondarenko

unread,

Jul 22, 2006, 2:05:38 AM7/22/06

to

................................................................

Robert Israel writes on Sat, Jul 22 2006 3:21 am

http://groups.google.com/group/sci.math.symbolic/msg/3ebe0ce3f2f561c6

You are a Maple Guru, and this is why your misunderstanding
deserves a special topic which I compose and post soon; as for
the time being, if you want to see the things under the light
you had expressed above, I rewrite the DESCRIPTION of the bug.

At this please note: a system which just returns intact ANYTHING
being typed in can be successfully written by a young macaque.

When I (and most probably, any other Maplesoft's individual or
corporative customer) pay money Maplesoft, we have an implicit
hypothesis that a system can give a non-trivial solution which
enrich our viewpoint, and not just echo stupidly our input
as a young-macaque-written one.

................................................................

DESCRIPTION: Only Maple V Release 5 of 1997 can calculate
this integral correctly.

TEST CASE: int(1/(exp(z^2)+1), z= 0..1);

EXPECTED: 1/2*Pi^(1/2)*Sum((-1)^_k1/(_k1+1)^(1/2)*erf((_k1
+1)^(1/2)), _k1 = 0 .. infinity)

.4194664800

CHECKUP: evalf(1/2*Pi^(1/2)*Sum((-1)^_k1/(_k1+1)^(1/2)*
erf((_k1+1)^(1/2)), _k1 = 0 .. infinity));

.4194664800

--------------- (2005) Maple 10.00 ----------------------------