FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups 
 ProfileProfile   PreferencesPreferences   Log in to check your private messagesLog in to check your private messages   Log inLog in 
Forum index » Science and Technology » Math » Symbolic
(",) Do You Want To Know For Sure You Are Going To Heaven?
Post new topic   Reply to topic Page 34 of 36 [530 Posts] View previous topic :: View next topic
Goto page:  Previous  1, 2, 3, ..., 32, 33, 34, 35, 36 Next
Author Message
Craig Carey
science forum beginner


Joined: 18 Jun 2005
Posts: 41

PostPosted: Thu Mar 30, 2006 8:58 am    Post subject: Re: The grandest Math Ocean Liner ever Reply with quote

On 23 Mar 2006 21:42:07 -0800, "Vladimir Bondarenko"
<vb@cybertester.com> wrote:

Quote:

The world's largest, longest, tallest, grandest Math Ocean Liner
ever! The most famous Math Ocean Liner in the world! Up to
5,000,000 corporate & individual passengers. Absolutely safe!

Please fill out the form below

http://webstore.maplesoft.com/

to tell us about your math cruise vacation preferences. A Maplesoft
personal consultant will then contact you to assist you with the
details and to help you finalize your math vacation plans. You can
expect a quick response during our regular hours of operation:
Weekdays from 9:00 AM to 5:30 PM EST.

Please find the first 16 Mb of the spectrum of entertainment stuff
we offer you on route to the final point of your voyage

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

Bon voyage!


.................. VM machine: Frame 1 of 1000 ..................

TEST CASE: ............;

EXPECTED: -I


The uninformative text after the "Bon voyage!" seems new.

Once again, one author of sci.math.symbolic who Mr Bondarkenko will
not ever actually name when referring to the origin of the information,
is himself.

It seems that Google is finding adversely against Bondarenko since the
groups.google.com site only lists the last message as being under
sci.math.symbolic, when I search for the title text.

It did get distributed to alt.math.recreational via Usenet data.

He copied from some worthless 21 June 2005 message, that was also
replicated:

/---------------------------------------------------------------------
On 21 Jun 2005 06:08:20 -0700, "Vladimir Bondarenko" wrote to
* sci.math.
* symbolic,
* comp.soft-sys.math.maple,
* sci.math,
* sci.edu,
* alt.math.recreational
Quote:

The world's largest, longest, tallest, grandest Math Ocean Liner
ever! The most famous Math Ocean Liner in the world! Up to
5,000,000 corporate & individual passengers. Absolutely safe!

Please fill out the form below

http://webstore.maplesoft.com/

to tell us about your math cruise vacation preferences. A Maplesoft
personal consultant will then contact you to assist you with the
details and to help you finalize your math vacation plans. You can
expect a quick response during our regular hours of operation:
Weekdays from 9:00 AM to 5:30 PM EST.

Please find the first 16 Mb of the spectrum of entertainment stuff
we offer you on route to the final point of our voyage

http://maple.bug-list.org/MapleCrisis-Review-01.pdf

Bon voyage!
\---------------------------------------------------------------------


Mr Bondarenko is engaged in repetition and concealing that with a
blunt 'unacceptable' technique that cuts hard and that deletes his
own surname and the older headers (ontaining the evidence of the
wrong-doing of junk-mailing Usenet groups).

The excerpt bars (ie. LHS column of ">" (etc.) characters) are
missing.

Also, I note that a query on bugs in Macromedia Dreamweaver:

+macromedia +dreamweaver (crashes OR crashing)

shows that factual credible criticisms of a program can be produced
without all the additional unasked-for detective's evidence from
Mr Bondarenko, allowing prognostications.

Maybe Google could be written to. Google seems to not have any
success matching the Ukranian's own, in replicating his meritless
posts: ie. it varies, but www.google.com prefers sci.edu, and
groups.google.com prefers sci.math.symbolic; at least for
'ocean liner' symbolism.
Back to top
Vladimir Bondarenko
science forum Guru


Joined: 23 Apr 2005
Posts: 601

PostPosted: Mon Apr 03, 2006 6:15 pm    Post subject: Re: Maple bugs are ubiquitous: A bug the long-liver in convert Reply with quote

Dave <solomons_dad.w.marks_and_w...@oracle.com>
writes on Mon, Apr 3 2006 12:51 pm

http://groups.google.com/group/sci.math.symbolic/msg/8fc0cb646a49fccc

D> Vladimir Bondarenko wrote:

VB>> Hello, all Maplesoft's actual and potential customers
VB>> over the world,

D> sigh...
D> Welcome to my killfile Vlad.


Hi Dave,

Do you expect me to get feeling excited or disappointed?

If you would have told me that the melting point of potassium
is about 63.65 degrees Centigrade at the normal pressure, would
you have expected from me a splash of joy, or a wild bout of
the total, utmost disappointment?

You might think, along with some other posters, that I am
looking for some response from my readers. Yes, I am happy
with a response. Also, if a reader selects to keep silence,
it's great, too. But the main goal of our posting is different
from getting of any kind of response.

VB>> Hello, all Maplesoft's actual and potential customers
VB>> over the world,

I am a Maplesoft's customer, and I am shocked with a way this
company pays to quality. In fact, it looks like, for an external
observer, that the Maplesoft's programmers run, to and fro, over
the Maple kernel and hack it with big sharp broadaxes...

And because of this, I have wasted LOTS of my precious time.

Trying to bypass the Maple bugs when possible etc... I've
contacted Maplesoft repeatedly, all in vain!

Please kindly ignore our messages. Or please kindly take them
into account. Or please kindly keep reading some of them while
ignoring the rest. Or a linear combination of the above-said...
or... There are a zillion of choices there ;)

You do not like chemistry? It would be quite funny had everyone
would like chemical facts.

I keep writing about the melting point of the potassium. The
melting point is an invariant w.r.t. my or your desire to know
or do not know about it.

It's not a religion. It's a science of quality assurance. Like
strength of materials or quantum chromodynamics.


Best of luck to you,

Vlad
Back to top
Vladimir Bondarenko
science forum Guru


Joined: 23 Apr 2005
Posts: 601

PostPosted: Tue Apr 04, 2006 3:27 am    Post subject: Re: Maple bugs are ubiquitous: A bug the long-liver in convert Reply with quote

Patrick Hamlyn <p...@multipro.N_OcomSP_AM.au>
writes on Tue, Apr 4 2006 3:00 am

http://groups.google.com/group/sci.math.symbolic/msg/cfda45e230dd2a9d


PH> Perhaps you don't understand the term 'killfile'.

Perhaps you don't understand the direction of our actions.
Perhaps you don't read my resume.

http://www.cas-testing.org/index.php?list=3
http://www.cas-testing.org/SciFace.phtml
http://www.cas-testing.org/index.php?list=7

PH> There's no point responding.

Perhaps you don't understand our goals. Not all of them
have been declared publicly. All in good time.

PH> And welcome to mine, BTW.

Bye-bye, don't worry, stay happy.


'Fraid, whether you like this or not, if you use Maple for
something more serious than 2+3; you almost inevitably
visit our sites again.

By several reasons.
Back to top
Walter Roberson
science forum beginner


Joined: 20 Jan 2006
Posts: 17

PostPosted: Mon Apr 10, 2006 7:09 am    Post subject: Re: A tough wager for a great guy named C. W. (aka Mr Chris) ;) Reply with quote

In article <1144650711.560570.5100@j33g2000cwa.googlegroups.com>,
Vladimir Bondarenko <vb@cybertester.com> wrote:
Quote:
C W <sylvest...@ns.sympatico.ca
writes on Mon, Apr 10 2006 3:30 am

CW> Why do You always malform the code ???

There is an unfixed yet Google Groups bug I reported
to the Google Team several times already, in 2005.

It manifests itself via an automated inserting of an
invisible "-" symbol in any texts if the width of the
line exceeds 67 or so.

By formatting the text in a way that its width never
exceeds 67 characters I try to fight this bug.

Why not just use a real newsreader to post, instead of google?

--
There are some ideas so wrong that only a very intelligent person
could believe in them. -- George Orwell
Back to top
Peter Luschny
science forum beginner


Joined: 13 May 2005
Posts: 36

PostPosted: Mon Apr 10, 2006 12:00 pm    Post subject: Re: An exact 1-D limit challenge - 24 Reply with quote

Vladimir Bondarenko wrote:

Quote:
Yesterday, having seen the terrific MuPAD bug you discovered [...]
I was in a total delight

I think you cannot better describe the difference
between you and a MuPad, Maple, or Mathematica user:

The latter mourn about what thrills you with delight.

Regards Peter
Back to top
Vladimir Bondarenko
science forum Guru


Joined: 23 Apr 2005
Posts: 601

PostPosted: Mon Apr 10, 2006 2:11 pm    Post subject: Re: An exact 1-D limit challenge - 24 Reply with quote

Peter Luschny <spamgr...@luschny.de>
writes on Mon, Apr 10 2006 3:00 pm

http://groups.google.com/group/sci.math.symbolic/msg/0c1ca3b56164343f

PL> I think you cannot better describe the difference...


Hello Peter,

Hurrah!! I am very, very glad to hear your voice again,
even if in a form of characters!!!

http://groups.google.com/group/comp.soft-sys.math.maple/msg/276b5ef697a9e015

PL> I allow a 90 per cent discount :)

No, I am not a politician still, - so 50:50! ;)

http://groups.google.com/group/sci.math.symbolic/msg/0c1ca3b56164343f

PL> I think you cannot better describe the difference
PL> between you and a MuPad, Maple, or Mathematica user:

PL> The latter mourn about what thrills you with delight.

I was wondering could I better describe the happy chime
between me (AXIOM, Derive, MuPAD, Maple, and Mathematica
user) and a MuPAD, Maple, or Mathematica user than this
way?

Bugs which thrill me with delight, have higher chances
to be published quickly, so the chances that they will
be fixed in the next release are presumably higher, so
in fact both me and an AXIOM, Derive, MuPAD, Maple, or
Mathematica user should rather rejoice that there is
a approach to discover bugs quickly ;)

(Now the CAS manufacturers should also rejoice and to
stand in a long line to use this opportunity to the
total delight of their customers - or reinvent our
algorithms - do you personally like this picture? Wink)

"Maple - unsafe at any evaluation!" -- Peter Luschny

;-)

How, how, how much you are right!
We calculated SUCH Turkish delight ;)

I was wondering what could be a right email address
to send you some this ambrosia? I promise you, a person
who knows what Beauty is, a theoretician, you will feel
yum-yum-yum! :-)

Just in case, my most usable address is

v b @ c y b e r t e s t e r . c o m

(I wrote it with blank spaces to fool the Google Groups
truncation...)

http://groups.google.com/group/comp.soft-sys.math.maple/msg/276b5ef697a9e015

PL> And if you succeed and eventually a better Maple is
PL> build, you will become famous like Nader and might
PL> some day be candidate for the presidential election
PL> of the Ukraine :)

Preparing for the elections '08 ! ;-)


Best, Vladimir
Back to top
Peter Luschny
science forum beginner


Joined: 13 May 2005
Posts: 36

PostPosted: Mon Apr 10, 2006 5:58 pm    Post subject: Re: An exact 1-D limit challenge - 24 Reply with quote

Quote:
Vladimir Bondarenko wrote:
Peter Luschny writes

PL> I think you cannot better describe the difference
PL> between you and a MuPad, Maple, or Mathematica user:
PL> The latter mourn about what thrills you with delight.

I was wondering could I better describe the happy chime
between me (AXIOM, Derive, MuPAD, Maple, and Mathematica
user) and a MuPAD, Maple, or Mathematica user than this
way?

I said, no, you could not.

Quote:
(Now the CAS manufacturers should also rejoice and to
stand in a long line to use this opportunity to the
total delight of their customers - or reinvent our
algorithms - do you personally like this picture? Wink)

They really ought to know all about it. And if they
do not, they should invest in your software, sure.

Quote:
"Maple - unsafe at any evaluation!" -- Peter Luschny

Well, this is a little bit unfair, especially if taken
out of context, so let me be more precise:
Every CAS is unsafe at any evaluation.

However, apart from freshmen, every user should know that
and exercise accordingly great caution in using this
kind of software.

Quote:
PL> And if you succeed and eventually a better Maple is
PL> build, you will become famous like Nader

Yes, but I observed that your campaign lost rather
than gained support. You forgot another quotation
from me: 'Vladimir's marketing is not perfect yet.'

Sadly, this is still true.

Quote:
Preparing for the elections '08 ! Wink

You are not the only one:
http://pandagon.net/2006/04/04/let-me-be-your-hillary-clinton/

Regards Peter
Back to top
Vladimir Bondarenko
science forum Guru


Joined: 23 Apr 2005
Posts: 601

PostPosted: Fri Apr 14, 2006 10:45 am    Post subject: Re: The grandest Math Ocean Liner ever Reply with quote

................... VM machine: Frame 1 of 1000 ..................


TEST CASE: I^2147483647;

EXPECTED: -I


-------------------- (2005) Maple 10.00 -------------------------
Error, numeric exception: overflow
-------------------- (2004) Maple 9.5.1 -------------------------
Error, numeric exception: overflow
-------------------- (2004) Maple 9.5 ---------------------------
Error, numeric exception: overflow
-------------------- (2003) Maple 9 -----------------------------
Error, numeric exception: overflow
-------------------- (2002) Maple 8 -----------------------------
Error, numeric exception: overflow
-------------------- (2001) Maple 7 -----------------------------
Error, numeric exception: overflow
-------------------- (2000) Maple 6 -----------------------------
-I
-------------------- (1997) Maple V Rel 5 -----------------------
-I
-------------------- (1995) Maple V Rel 4 -----------------------
-I
-------------------- (1994) Maple V Rel 3 -----------------------
-I
------------------------------------------------------------------


...................................................................
Back to top
Repeating Rifle
science forum Guru Wannabe


Joined: 25 Mar 2005
Posts: 205

PostPosted: Fri Apr 14, 2006 5:31 pm    Post subject: Re: The grandest Math Ocean Liner ever Reply with quote

On 4/14/06 3:45 AM, in article
1145011559.840088.226670@z34g2000cwc.googlegroups.com, "Vladimir Bondarenko"
<vb@cybertester.com> wrote:

Quote:
.................. VM machine: Frame 1 of 1000 ..................


TEST CASE: I^2147483647;

EXPECTED: -I


-------------------- (2005) Maple 10.00 -------------------------
Error, numeric exception: overflow
-------------------- (2004) Maple 9.5.1 -------------------------
Error, numeric exception: overflow
-------------------- (2004) Maple 9.5 ---------------------------
Error, numeric exception: overflow
-------------------- (2003) Maple 9 -----------------------------
Error, numeric exception: overflow
-------------------- (2002) Maple 8 -----------------------------
Error, numeric exception: overflow
-------------------- (2001) Maple 7 -----------------------------
Error, numeric exception: overflow
-------------------- (2000) Maple 6 -----------------------------
-I
-------------------- (1997) Maple V Rel 5 -----------------------
-I
-------------------- (1995) Maple V Rel 4 -----------------------
-I
-------------------- (1994) Maple V Rel 3 -----------------------
-I
------------------------------------------------------------------


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


Of course this is retrograde development, but no one is perfect. At least
you are warned that no answer is supplied. I suppose that is infinitely
better than getting an unflagged wrong answer. Except for showing a
programming flaw, I would never attempt doing this problem.

From a practical point of view, if you must absolutely positively get a
correct answer try

Quote:
I^100000000*I^47483647;
-I


Bill

-- Ferme le Bush
Back to top
Carl Love
science forum beginner


Joined: 15 Apr 2006
Posts: 3

PostPosted: Sat Apr 15, 2006 7:32 pm    Post subject: Re: A tough wager for a great guy named C. W. (aka Mr Chris) ;) Reply with quote

Vladimir Bondarenko wrote:
Quote:
There is an unfixed yet Google Groups bug I reported
to the Google Team several times already, in 2005.
It manifests itself via an automated inserting of an
invisible "-" symbol in any texts if the width of the
line exceeds 67 or so. This makes no damage to regular
text, but might cause real headache with a solid amount
of the code (if its width >= about 67 characters) as,
say, Maple tumbles over the unexpected "-"s.

Then to understand what the author was going to tell us,
one should delete those nasty extra "-"s manually.

Now imagine that being copied into Maple, one cannot
tell by vision the genuine vs the false "-"s... so
one should resort to trial-and-error editing of the
code ;)

(well, at times, some of those "-"s are false obviously
but not always)

Here are some tips to make removing the false "-" a little easier. If
you follow these, you will always be able to distinguish the real minus
signs from the fake minus signs. This assumes that the original code
is syntactically correct and that it's author is not trying to
deliberately make things confusing by names that contain minus signs.

1. Remove any minus signs that are obviously wrong

2. Remove any that cause syntax errors. The parser will usually place
your cursor very close to or directly on the offending character.

3. (The non-obvious step): All of the remaining false minus signs are
manifested as names; they will never be interpretted by Maple as minus
signs. So, after entering the expression, let's say it is called e,
give the command
Quote:
indets(e, name);
If the resulting set has any minus signs, either alone or embedded in

other names, these are the fakes.
Back to top
Robert B. Israel
science forum Guru


Joined: 24 Mar 2005
Posts: 2151

PostPosted: Wed Apr 26, 2006 6:46 pm    Post subject: Re: Yet another Maple bug the long-liver discovered [collect] Reply with quote

In article <1146074700.328474.119130@y43g2000cwc.googlegroups.com>,
Vladimir Bondarenko <vb@cybertester.com> wrote:
Quote:
"Roman" <wfri...@gmx.net> writes on Wed, Apr 26 2006 4:57 pm

http://groups.google.com/group/comp.soft-sys.math.maple/msg/dc5c790c01ddaa36


e:=sqrt(x^3)+a*sqrt(x^3)+b*sqrt(x^3);
collect(e, indets(e, sqrt(anything)));

Error, (in collect) cannot collect (x^3)^(1/2)

Before you get all excited about "another Maple bug", find out whether
this really is a bug. The help page ?collect says:

The indeterminates can be names or unevaluated function calls but not
sums or products or fractional powers. To collect fractional powers,
for example x^(1/2) just collect x.

So: no bug, just a known and documented limitation of functionality.

Robert Israel israel@math.ubc.ca
Department of Mathematics http://www.math.ubc.ca/~israel
University of British Columbia Vancouver, BC, Canada
Back to top
Vladimir Bondarenko
science forum Guru


Joined: 23 Apr 2005
Posts: 601

PostPosted: Sat Apr 29, 2006 2:31 am    Post subject: Re: Noosphere Academy spams our poor Noosphere - Sample # 3 Reply with quote

An anonymous pusher of Noosphere Academy
v...@hotmail.com writes on Fri, Apr 28 2006 5:22 pm

http://groups.google.com/group/sci.math.symbolic/msg/405c408293312b59

Quote:
"translated onto English"


Ah "onto".....

Zero cool!

I hope it is NOT you who are the translator?

Please keep in mind my free words of advice both for you
personally and at least some Full members of Noosphere
Academy, like Mr Aladjev who produces, in packs, gems
like "polinomial", "typies", "simmetrical", "analysys",
"choosen" and many dozens more not forgetting about
variegated stylistic howlers, too.

"At present, full knowledge of this topic does not exist."

(Wow! More fun, say, here

Noosphere Academy and its "typies" (!): Rewritting getailed choosen
spreaded reprentation...
http://groups.google.com/group/sci.math.symbolic/msg/dd8fb71c6e5c8349

and is coming.)

So I'd propose, once more time, both you and the Noosphere
Academy to fight a bit your English illiteracy via

1) hiring a professional translator
2) spell checking his/her output

Have no doubt, I will read this masterpiece coming with
much interest, and produce new comments depending on what
I'd see there... I have still much to add about quality
of the Noosphere Academy's writings.

Now while you are sweating over translation from Russian
into English, (please remember, "into", not "onto") here
comes some info in English, enjoy.

http://groups.google.com/group/sci.math.symbolic/msg/48cbd4c2d2a7e455


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
Back to top
Daniel Lichtblau
science forum beginner


Joined: 12 May 2005
Posts: 48

PostPosted: Mon May 01, 2006 2:29 pm    Post subject: Re: The VM machine presents yet another freshman's level Maple bug [limit/assume] Reply with quote

Vladimir Bondarenko wrote:
Quote:
.................................................................
[irrelevance snipped --dl]


Quote:
DESCRIPTION: Global assume affects a local variable.

TEST CASE: assume(z=0): limit(exp(Pi*z), z= infinity);

EXPECTED: infinity


--------------- (2005) Maple 10.00 ------------------------
1
[more irrelevance snipped --dl]


Quote:
Viva la advanced QA methods,

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

I am not able to respond to the Maple-specific aspects of this example.
I wanted to point out why it is a form of garbage-in, garbage-out.

I think most readers are familiar with the idea that False implies
anything, hence if we accept that the two uses of z, first in assume
and later in limit, refer to the same variable, then pretty much any
result is acceptable since we have both z=0 and z>0.

So the next issue is to understand why a mathematical programming
language might want them to be related rather than have the limit usage
"locally scoped". [Note: I make no claims as to what are the actual or
documented Maple semantics here.] I illustrate below using the
development version of Mathematica why it makes sense to have
assumptions that can apply to the limit variable. Specifically they
will affect the path of approach.

In[14]:= Assuming[z>0, Limit[E^(1/z), z->0]]
Out[14]= Infinity

In[15]:= Assuming[z<0, Limit[E^(1/z), z->0]]
Out[15]= 0

In math languages, generally speaking, there are subtleties to the
issue of how to "scope" entities that mathematicians would regard as
dummy variables. One might have varying ideas of how to handle the
variable for limits, series, indefinite and definite integrals, and
sums. I certainly would not claim that Mathematica has a uniform, let
alone perfect, notion of how to do all these. My point is that there
may be bona fide reasons for not scoping such variables locally to the
function in question.

Daniel Lichtblau
Wolfram Research
Back to top
Richard Fateman
science forum Guru Wannabe


Joined: 03 May 2005
Posts: 181

PostPosted: Mon May 01, 2006 2:49 pm    Post subject: re-inventing the wheel... scope of assumptions. Reply with quote

Below is a description of the handling of assumptions in contexts in
Macsyma, circa 1976, some 30 years ago. As I have observed
previously, "new" computer algebra systems, by which I mean
Mathematica, Maple, Mupad...typically stand on the feet of, rather
than the shoulders of, previous work. By either not understanding
the importance of assumptions in their initial design, or ignoring the
design problems encountered regarding assumptions in Macsyma, they
repeated the mistakes and then had to try ad-hoc techniques when
it became apparent that "assume" etc would really be important.
To a limitedf extent, assumptions were successfully grafted on to
design that excluded them initially. How much better might it have
been if Maple and Mathematica started out with assumptions?
(Should assumptions be scoped along with value bindings?
Scoped according to file boundaries? Scoped globally by name
e.g. as in FORTRAN by first letter of variable name(!), or by a
mechanism specific for storing assumptions similar to
Lisp packages?)


Anyway.... from the Macsyma/Maxima on-line info.....

`contexts' is a list of the contexts which currently exist,
including the currently active context.

The context mechanism makes it possible for a user to bind together
and name a selected portion of his database, called a context.
Once this is done, the user can have Maxima assume or forget large
numbers of facts merely by activating or deactivating their
context.

Any symbolic atom can be a context, and the facts contained in that
context will be retained in storage until destroyed one by one by
calling `forget' or destroyed as a whole by calling `kill' to
destroy the context to which they belong.

Contexts exist in a hierarchy, with the root always being the
context `global', which contains information about Maxima that some
functions need. When in a given context, all the facts in that
context are "active" (meaning that they are used in deductions and
retrievals) as are all the facts in any context which is a
subcontext of the active context.

When a fresh Maxima is started up, the user is in a context called
`initial', which has `global' as a subcontext.

See also `facts', `newcontext', `supcontext', `killcontext',
`activate', `deactivate', `assume', and `forget'.


-- Function: killcontext (<context_1>, ..., <context_n>)
Kills the contexts <context_1>, ..., <context_n>.

If one of the contexts is the current context, the new current
context will become the first available subcontext of the current
context which has not been killed. If the first available unkilled
context is `global' then `initial' is used instead. If the
`initial' context is killed, a new, empty `initial' context is
created.

`killcontext' refuses to kill a context which is currently active,
either because it is a subcontext of the current context, or by
use of the function `activate'.

`killcontext' evaluates its arguments. `killcontext' returns
`done'.

..................................
"Daniel Lichtblau" <danl@wolfram.com> wrote in message
news:1146493741.378443.145910@v46g2000cwv.googlegroups.com...
Quote:

Vladimir Bondarenko wrote:
.................................................................
[irrelevance snipped --dl]


DESCRIPTION: Global assume affects a local variable.

TEST CASE: assume(z=0): limit(exp(Pi*z), z= infinity);

EXPECTED: infinity


--------------- (2005) Maple 10.00 ------------------------
1
[more irrelevance snipped --dl]


Viva la advanced QA methods,

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

I am not able to respond to the Maple-specific aspects of this example.
I wanted to point out why it is a form of garbage-in, garbage-out.

I think most readers are familiar with the idea that False implies
anything, hence if we accept that the two uses of z, first in assume
and later in limit, refer to the same variable, then pretty much any
result is acceptable since we have both z=0 and z>0.

So the next issue is to understand why a mathematical programming
language might want them to be related rather than have the limit usage
"locally scoped". [Note: I make no claims as to what are the actual or
documented Maple semantics here.] I illustrate below using the
development version of Mathematica why it makes sense to have
assumptions that can apply to the limit variable. Specifically they
will affect the path of approach.

In[14]:= Assuming[z>0, Limit[E^(1/z), z->0]]
Out[14]= Infinity

In[15]:= Assuming[z<0, Limit[E^(1/z), z->0]]
Out[15]= 0

In math languages, generally speaking, there are subtleties to the
issue of how to "scope" entities that mathematicians would regard as
dummy variables. One might have varying ideas of how to handle the
variable for limits, series, indefinite and definite integrals, and
sums. I certainly would not claim that Mathematica has a uniform, let
alone perfect, notion of how to do all these. My point is that there
may be bona fide reasons for not scoping such variables locally to the
function in question.

Daniel Lichtblau
Wolfram Research
Back to top
Daniel Lichtblau
science forum beginner


Joined: 12 May 2005
Posts: 48

PostPosted: Mon May 01, 2006 3:35 pm    Post subject: Re: re-inventing the wheel... scope of assumptions. Reply with quote

Richard Fateman wrote:
Quote:
Below is a description of the handling of assumptions in contexts in
Macsyma, circa 1976, some 30 years ago. As I have observed
previously, "new" computer algebra systems, by which I mean
Mathematica, Maple, Mupad...typically stand on the feet of, rather
than the shoulders of, previous work. By either not understanding
the importance of assumptions in their initial design, or ignoring the
design problems encountered regarding assumptions in Macsyma, they
repeated the mistakes and then had to try ad-hoc techniques when
it became apparent that "assume" etc would really be important.
To a limitedf extent, assumptions were successfully grafted on to
design that excluded them initially. How much better might it have
been if Maple and Mathematica started out with assumptions?
(Should assumptions be scoped along with value bindings?
Scoped according to file boundaries? Scoped globally by name
e.g. as in FORTRAN by first letter of variable name(!), or by a
mechanism specific for storing assumptions similar to
Lisp packages?)


Anyway.... from the Macsyma/Maxima on-line info.....

`contexts' is a list of the contexts which currently exist,
including the currently active context.

The context mechanism makes it possible for a user to bind together
and name a selected portion of his database, called a context.
Once this is done, the user can have Maxima assume or forget large
numbers of facts merely by activating or deactivating their
context.

Any symbolic atom can be a context, and the facts contained in that
context will be retained in storage until destroyed one by one by
calling `forget' or destroyed as a whole by calling `kill' to
destroy the context to which they belong.

Contexts exist in a hierarchy, with the root always being the
context `global', which contains information about Maxima that some
functions need. When in a given context, all the facts in that
context are "active" (meaning that they are used in deductions and
retrievals) as are all the facts in any context which is a
subcontext of the active context.

When a fresh Maxima is started up, the user is in a context called
`initial', which has `global' as a subcontext.

See also `facts', `newcontext', `supcontext', `killcontext',
`activate', `deactivate', `assume', and `forget'.


-- Function: killcontext (<context_1>, ..., <context_n>)
Kills the contexts <context_1>, ..., <context_n>.

If one of the contexts is the current context, the new current
context will become the first available subcontext of the current
context which has not been killed. If the first available unkilled
context is `global' then `initial' is used instead. If the
`initial' context is killed, a new, empty `initial' context is
created.

`killcontext' refuses to kill a context which is currently active,
either because it is a subcontext of the current context, or by
use of the function `activate'.

`killcontext' evaluates its arguments. `killcontext' returns
`done'.
[...]

You describe the context(s) in which assumptions may apply in
Macsyma/Maxima. These are important matters but not I believe they are
not terribly germane to the earlier notes in this particular thread.

As I see it, the issue there is not where the assumptions apply but
rather how the dummy variables in various math functions are to be
scoped. My point is that it makes sense to do that scoping in a way
that allows for "outside" assumptions to be visible, if those
assumptions are visible to the math functions when invoked. That does
not really have much to do with how those outside assumptions might
themselves be provided or scoped.

For the record, and since you raise issues that are important in their
own right, let me note ways in which Mathematica assumptions may be
scoped.

(1) Globally, using
$Assumptions = ...
This will affect all functions that take an Assumptions option.

(2) In a local setting using
Assuming[assumptions, expression]
This will affect functions invoked within expression that take an
Assumptions option.

(3) Purely local to a function via
f[args, Assumptions->assumptions]
This applies only to functions that take Assumptions as an option.
Among these are Simplify, Refine, Integrate, and Limit.

I think this is a bit different from the description provided for
Maxima assumption handling but there seems to be some commonality.
Certainly one can next Assuming blocks in Mathematica. As seen below
they will interact with one another.

In[5]:= Assuming[x==4, Refine[x]]
Out[5]= 4

In[6]:= Assuming[x==0, Assuming[x==4, Refine[x]]]
$Assumptions::cas:
Warning: Contradictory assumption(s) x == 4 && x == 0 encountered.
Out[6]= x

This might be a good or bad thing depending on point of view.

One can set up "sets of assumptions" e.g. as
context1 = assumptions1;
context2 = assumptions2;
....
and later use them as Assuming[context1,expression]

I don't know if this addresses all possible issues of points of
similarity or difference between current Mathematica assumptions usage
and that of Maxima. I think it gives some idea, though. I'll get off
your feet now.

Daniel Lichtblau
Wolfram Research
Back to top
Google

Back to top
Display posts from previous:   
Post new topic   Reply to topic Page 34 of 36 [530 Posts] Goto page:  Previous  1, 2, 3, ..., 32, 33, 34, 35, 36 Next
View previous topic :: View next topic
The time now is Thu Jul 27, 2017 6:32 pm | All times are GMT
Forum index » Science and Technology » Math » Symbolic
Jump to:  

Similar Topics
Topic Author Forum Replies Last Post
No new posts TURMEL: H.G. Wells' Open Conspiracy for Heaven John Turmel Engineering 1 Wed Feb 15, 2006 1:19 pm
No new posts TURMEL: No WSF social currency lecture listed; Kennedy's ... John Turmel Engineering 0 Sat Jan 14, 2006 7:42 pm
No new posts Stairway to Heaven- lyrics- literal truth... kutaman Physics 0 Tue Jun 28, 2005 8:37 am
No new posts unified field theory-real stairway to heaven kutaman Physics 0 Wed Jun 22, 2005 11:55 pm
No new posts unified field theory-real stairway to heaven kutaman Physics 0 Wed Jun 22, 2005 7:44 pm

Copyright © 2004-2005 DeniX Solutions SRL
Other DeniX Solutions sites: Electronics forum |  Medicine forum |  Unix/Linux blog |  Unix/Linux documentation |  Unix/Linux forums  |  send newsletters
 


Powered by phpBB © 2001, 2005 phpBB Group
[ Time: 1.4293s ][ Queries: 16 (1.3931s) ][ GZIP on - Debug on ]