Maple bugs: asympt quality degradation sample

From: Vladimir Bondarenko (vb_at_cybertester.com)
Date: 12/27/04 

Date: 27 Dec 2004 08:33:56 -0800

Hello the symbolic computation community,

Let's for a while leave Maplesoft's SAND-SOLID numerics alone
to have great fun with them soon properly.

Let's consider the quality degradation in symbolic mode. Let's
take something exquisitely simple.

Visiting the blog by Laurent Bernardin, Maplesoft Chief Scientist
and VP ...

http://bernardin.com/maple/index.php

(has Laurent already again forbidden our placid entering his
writings? - there, there, don't cry! if this is the case, then
please do not sorrow over your hard fate, we'll fix this soon,
Daddy has a cure ;)

... we read the following, with sincere pleasure.

LB> More recently, with Maple 7, 8, 9 and now with Maple 9.5
LB> we have taken the issue of backwards compatibility very
LB> seriously.

Great! Hurrah!

LB> We have not stopped making language changes but we are
LB> making them in such a way that we provide new functionality
LB> without breaking existing code.

Excellent! What about this?

Here is yet another type of Maple quality degradation - an
abrupt discontinuity in quality. All was well for many years,
now - bang! - and the things went awry.

What GEMM proposes to our attention?

> asympt(arctan(ln(z)), z);

-------------------------Maple 9.5.1------------------------------
-1/2*Pi+1/ln(z) # Ohhh!
-------------------------Maple 9.5--------------------------------
-1/2*Pi+1/ln(z) # Ahhh!
-------------------------Maple 9----------------------------------
1/2*Pi-1/ln(z)
-------------------------Maple 8----------------------------------
1/2*Pi-1/ln(z)
-------------------------Maple 7----------------------------------
1/2*Pi-1/ln(z)
-------------------------Maple 6----------------------------------
1/2*Pi-1/ln(z)
-------------------------Maple V Rel 5----------------------------
1/2*Pi-1/ln(z)
-------------------------Maple V Rel 4----------------------------
1/2*Pi-1/ln(z)
-------------------------Maple V Rel 3----------------------------
Error, (in asympt) unable to compute series # Zilch!
------------------------------------------------------------------

VB> Ask yourself, do you do your morning exercises regularly to
VB> reach an age this bug (and at least 10,000 of other distinct
VB> Maple bugs) will be fixed? Additionally, remember, if you
VB> still smoke, do quit as quickly as you can!

Is quitting smoking enough? Shouldn't be extra measures to be
taken? The doctors say, There is nothing to touch sea air for
bracing you up!

Just an idea. Maybe Maplesoft will pay to each Maple customer
a short trip to seashore?

Best wishes,

Vladimir Bondarenko

GEMM architect
Co-founder, CEO, Mathematical Director
Cyber Tester, LLC
13 Dekabristov Str, Simferopol
Crimea 95000, Ukraine
tel: +38-(0652)-447325
tel: +38-(0652)-230243
tel: +38-(0652)-523144
fax: +38-(0652)-510700

http://www.cybertester.com/
http://maple.bug-list.org/
http://www.CAS-testing.org/