Re: Loss of Mars Climate Orbiter due to units of measurment conflicts

Re: Loss of Mars Climate Orbiter due to units of measurment conflicts

[Larry Kilgallen]

In article <7tvpbe$45t@dfw-ixnews4.ix.netcom.com>, Richard D Riehle <laoXhai@ix.netcom.com> writes:
> 
> I am not sure why this topic is in the Ada Usenet group.  Was the
> Mars Climate Orbiter programmed in Ada?  

The discussion started with people suggesting the use of the Ada
typing system to avoid mismatches of units.  Consensus is that
if it was done in Ada it was not done right, which is also true
if it was done in Algol or any other language.

Larry Kilgallen

Re: Loss of Mars Climate Orbiter due to units of measurment conflicts

[Richard D Riehle]

I am not sure why this topic is in the Ada Usenet group.  Was the
Mars Climate Orbiter programmed in Ada?  

Richard Riehle
http://www.adaworks.com

Re: Loss of Mars Climate Orbiter due to units of measurment conflicts

[Wes Groleau]

> > What makes the kilometer a better measure for space travel ?
> 
> The fact that the fundamental constants are rendered in this
> set of units?  For example, G, the universal Gravitation constant.
> And the most accurate earth model terms are also in kilometers.

While I am personally in favor of the SI, I must insist that the
"fundamental constants" are rendered in whatever units the renderer 
chooses.  I was taught in high school that G = 32.2 ft / sec**2

Re: Loss of Mars Climate Orbiter due to units of measurment conflicts

[Fraser Wilson]

paene lacrimavi postquam Wes Groleau <wwgrol@ftw.rsc.raytheon.com> scripsit:

>While I am personally in favor of the SI, I must insist that the
>"fundamental constants" are rendered in whatever units the renderer 
>chooses.  I was taught in high school that G = 32.2 ft / sec**2

That's not G, that's g!  I bet this wouldn't have happened if you'd
been using SI!  :)

ObAda: if only physical constants were case-insensitive.

Fraser.

Re: Loss of Mars Climate Orbiter due to units of measurment conflicts

[Ehud Lamm]

On Thu, 14 Oct 1999, Wes Groleau wrote:

|While I am personally in favor of the SI, I must insist that the
|"fundamental constants" are rendered in whatever units the renderer 
|chooses.  I was taught in high school that G = 32.2 ft / sec**2
|

Of course you can use whatever units you choose (well measuring distance
in kilograms is not really what I have in mind). But if you choose
constants based on units that don't agree with your measurements you might
crash a space vehicle....
I my last physics exam, which was a lab exam, I measured all distances in
cm. (we had to measure distance of movement after impact) - since all
movement was on a desk top. The constant was of course MKS, so my graph
looked kind of strange. I was flabbergasted for 55 minutes out of the
total 60, and only had time to add on the exam paper "the units!!! always
check the units!!! I used cm. and MKS"
Apperantly the tester liked it, since I got 90 something (I don't remember
exactly). It was really funny - beacuse I was the last guy out of there,
and beacuse all my claculations were wrong.
Well at least it didn't cost anyone $125 million... :-)

Ehud Lamm mslamm@mscc.huji.ac.il
http://purl.oclc.org/NET/ehudlamm <== My home on the web 
Check it out and subscribe to the E-List- for interesting essays and more!

 

Re: Loss of Mars Climate Orbiter due to units of measurment conflicts

[Robert Dewar]

In article
<Pine.A41.3.96-heb-2.07.991021164856.126066B-100000@pluto.mscc.h
uji.ac.il>,
  Ehud Lamm <mslamm@mscc.huji.ac.il> wrote:
> It seems we are sometimes too eager to find arguments where
none really
> exists (like variable names vs. comments, which I really
assume we all
> agree are both useful.


Of course we all agree that both are useful, but that is not
what the discussion is about, and there is indeed some
disagreement here, the disussion is real, reread carefully :-)

The issue is whether names can stand in for comments....


Sent via Deja.com http://www.deja.com/
Before you buy.

Re: Loss of Mars Climate Orbiter due to units of measurment conflicts

[Wes Groleau]

> |While I am personally in favor of the SI, I must insist that the
> |"fundamental constants" are rendered in whatever units the renderer
> |chooses.  I was taught in high school that G = 32.2 ft / sec**2

> Of course you can use whatever units you choose (well measuring distance
> in kilograms is not really what I have in mind). But if you choose
> constants based on units that don't agree with your measurements you might
> crash a space vehicle....

Yes, but I was rebutting a post that claimed the reason for using SI is 
because "that's what the fundamental constants are rendered in."

Re: Loss of Mars Climate Orbiter due to units of measurment conflicts

[Ehud Lamm]

Wes, 
I was not arguing with you. Simply adding a bit to the discussion.

It seems we are sometimes too eager to find arguments where none really
exists (like variable names vs. comments, which I really assume we all
agree are both useful. We then argue on drawing some verey fine line,
which is usally not something anyone can really define precisely).

Even so, as far as I can tell, this is one of the more eduacted and
informative groups.

Ehud Lamm mslamm@mscc.huji.ac.il
http://purl.oclc.org/NET/ehudlamm <== My home on the web 
Check it out and subscribe to the E-List- for interesting essays and more!

Re: Loss of Mars Climate Orbiter due to units of measurment conflicts

[Ehud Lamm]

Let's not argue about whether the arguments are real... :-)

I read the discussions, I just wanted to point out that sometimes
arguments cause exagerations and taking the extreme. 

By the way (to make this a little programming related), I think that there
are cases that naming can reduce the need for comments. This is seen with
people who choose very bad names, which must be accompanined by comments -
since they are otherwise meaningless or misleading.
I had a student who called a routine Input - but the routine simply
returned an initial value. Now - a commnet really helped me understand
what he wanted the routine to do. Naming it better would have been a
better solution.

But after stating this rather banal cliam, I think we can agree that with
normally acceptable names, we still need comments since names do not
provided enough information. 

My approach is sually that if someone can misread you, or not understand
your design reasoning - it is much better to explain (using comments, and
some times, god forbid, external documentation), than to solve the
problems that will arise later. 
Usually the someone misreading you, will be yourself a week later...

Ehud Lamm mslamm@mscc.huji.ac.il
http://purl.oclc.org/NET/ehudlamm <== My home on the web 
Check it out and subscribe to the E-List- for interesting essays and more!

Re: Loss of Mars Climate Orbiter due to units of measurment conflicts

[Robert B. Love]

In <380F272A.988B7524@ftw.rsc.raytheon.com> Wes Groleau wrote:
> 
> > |While I am personally in favor of the SI, I must insist that the
> > |"fundamental constants" are rendered in whatever units the renderer
> > |chooses.  I was taught in high school that G = 32.2 ft / sec**2
> 
> 
> Yes, but I was rebutting a post that claimed the reason for using SI is 
> because "that's what the fundamental constants are rendered in."

Actually, you can't "choose" what units the fundamental constants
are in.  They are decided on by an international body and the method
of measurement, the units used, the accuracy, etc are all part of
the published value.  You wouldn't expect to have "Wes' Own Ada"
would you? What made Ada was decided upon by a large body after
much consultation.  Of course you can convert from SI to Imperial
but I'll stand by my original statement.  Fundamental constants
are published in SI units, not some other units of convenience.


--
----------------------------------------------------------------
 Bob Love                                   
 rlove@neosoft.com                            
----------------------------------------------------------------

Re: Loss of Mars Climate Orbiter due to units of measurment conf

[Wilhelm Spickermann]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset=us-ascii, Size: 911 bytes --]


On 24-Oct-99 Robert B. Love wrote:
> In <380F272A.988B7524@ftw.rsc.raytheon.com> Wes Groleau wrote:
>> 
>> > |While I am personally in favor of the SI, I must insist that the
>> > |"fundamental constants" are rendered in whatever units the
>> > |renderer
>> > |chooses.  I was taught in high school that G = 32.2 ft / sec**2
>> 
>> 
>> Yes, but I was rebutting a post that claimed the reason for using SI
>> is 
>> because "that's what the fundamental constants are rendered in."
> 
> Actually, you can't "choose" what units the fundamental constants
> are in.  They are decided on by an international body and the method
> of measurement, the units used, the accuracy, etc are all part of
> the published value. 
..

I wouldn�t quite say that. Fundamental constants like the gravitational
constant or the speed of light cannot be decided on and no one may
decide, how they are to be measured. 

Wilhelm

Re: Loss of Mars Climate Orbiter due to units of measurment conflicts

[Wes Groleau]

> > > |While I am personally in favor of the SI, I must insist that the
> > > |"fundamental constants" are rendered in whatever units the renderer
> > > |chooses.  I was taught in high school that G = 32.2 ft / sec**2
> >
> > Yes, but I was rebutting a post that claimed the reason for using SI is
> > because "that's what the fundamental constants are rendered in."
> 
> Actually, you can't "choose" what units the fundamental constants
> are in.  They are decided on by an international body and the method
> of measurement, the units used, the accuracy, etc are all part of
> the published value.  You wouldn't expect to have "Wes' Own Ada"
> would you? What made Ada was decided upon by a large body after
> much consultation.  Of course you can convert from SI to Imperial
> but I'll stand by my original statement.  Fundamental constants
> are published in SI units, not some other units of convenience.

So, in the above scenario, the international body is the renderer that
_chose_ to publish them in SI.  Other renderers may have chosen (as did
the author of my high school physics text) to render them in other
units.

I encourage SI--but the fact that someone used it somewhere to publish
an "official" approximation for the speed of light or Avogadro's
constant*
is not a good argument for using SI

* Ever notice that Ada 95 Quality and Style contains two different
values 
  for Avogadro's constant?  (See 3.1.2 and 3.2.6)  Notice that neither
  mentions units, nor matches the 1998 CODATA value?

Re: Loss of Mars Climate Orbiter due to units of measurment conf

[Wes Groleau]

> > Actually, you can't "choose" what units the fundamental constants
> > are in.  They are decided on by an international body and the method
> > of measurement, the units used, the accuracy, etc are all part of
> > the published value.
> ..
> 
> I wouldn�t quite say that. Fundamental constants like the gravitational
> constant or the speed of light cannot be decided on and no one may
> decide, how they are to be measured.

Oh, yeah?  The Indiana legislature once "decided" that Pi equals 3.0...
(Or is that "urban legend"?)

Re: Loss of Mars Climate Orbiter due to units of measurment conf

[Gisle S�lensminde]

In article <38147DBD.8214F32C@ftw.rsc.raytheon.com>, Wes Groleau wrote:
>
>> > Actually, you can't "choose" what units the fundamental constants
>> > are in.  They are decided on by an international body and the method
>> > of measurement, the units used, the accuracy, etc are all part of
>> > the published value.
>> ..
>> 
>> I wouldn�t quite say that. Fundamental constants like the gravitational
>> constant or the speed of light cannot be decided on and no one may
>> decide, how they are to be measured.
>
>Oh, yeah?  The Indiana legislature once "decided" that Pi equals 3.0...
>(Or is that "urban legend"?)

Even worse. In according to an old Norwegian translation of Guinness 
book of records, they decided it to be exactly equal 4.0 The record was 
"the most unprecise estimate of pi". I don't have the book at hand now,
so I can't give further details. 

--
Gisle S�lensminde ( gisle@ii.uib.no )   

Re: Loss of Mars Climate Orbiter due to units of measurment conf

[Keith Thompson]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1266 bytes --]

gisle@lunde.ii.uib.no (Gisle S�lensminde) writes:
[...]
> Even worse. In according to an old Norwegian translation of Guinness 
> book of records, they decided it to be exactly equal 4.0 The record was 
> "the most unprecise estimate of pi". I don't have the book at hand now,
> so I can't give further details. 

This is *way* off-topic, but you can find more information in Petr
Beckmann's book "A History of pi".  In 1897, "A bill introducing
a new Mathematical truth" was introduced in the Indiana House of
Representatives.  The bill was full of internal contradictions;
one could conclude from the first paragraph that pi = 16/sqrt(3), or
about 9.2376.  After going through the committees on Swamp Lands and
Education, the House passed the bill 67-0.  Fortunately, a mathematics
professor happened to be visiting the state Senate on the day the
bill came up for debate there (after going through the Committee
on Temperance).  On his recommendation, further consideration of the
bill was postponed indefinitely.  It hasn't been on the agenda since.

-- 
Keith Thompson (The_Other_Keith) kst@cts.com  <http://www.ghoti.net/~kst>
San Diego Supercomputer Center           <*>  <http://www.sdsc.edu/~kst>
"Oh my gosh!  You are SO ahead of your time!" -- anon.

Re: Loss of Mars Climate Orbiter due to units of measurment conf

[Bryce Bardin]

Keith Thompson wrote:
> 
> gisle@lunde.ii.uib.no (Gisle S�lensminde) writes:
> [...]
> > Even worse. In according to an old Norwegian translation of Guinness
> > book of records, they decided it to be exactly equal 4.0 The record was
> > "the most unprecise estimate of pi". I don't have the book at hand now,
> > so I can't give further details.
> 
> This is *way* off-topic, but you can find more information in Petr
> Beckmann's book "A History of pi".  In 1897, "A bill introducing
> a new Mathematical truth" was introduced in the Indiana House of
> Representatives.  The bill was full of internal contradictions;
> one could conclude from the first paragraph that pi = 16/sqrt(3), or
> about 9.2376.  After going through the committees on Swamp Lands and
> Education, the House passed the bill 67-0.  Fortunately, a mathematics
> professor happened to be visiting the state Senate on the day the
> bill came up for debate there (after going through the Committee
> on Temperance).  On his recommendation, further consideration of the
> bill was postponed indefinitely.  It hasn't been on the agenda since.
> 
> --
> Keith Thompson (The_Other_Keith) kst@cts.com  <http://www.ghoti.net/~kst>
> San Diego Supercomputer Center           <*>  <http://www.sdsc.edu/~kst>
> "Oh my gosh!  You are SO ahead of your time!" -- anon.

The full story is reported on the following web page:
  http://www.acc.umu.se/~olletg/pi/indiana.html

It says in part:

   "The bill implies four different values for pi and one for sqrt(2),
as
   follows: pi' =3D 16/sqrt(3), 2 sqrt(5 pi/6), 16 sqrt(2)/7, 16/5 (
9.24 ,
   3.236 , 3.232 , 3.2 respectively.) sqrt(2)' =3D 10/7."

(I was quoted the "16/5" value by many of my colleagues when I was on
the
Physics faculty of Indiana University.)

P.S.  Keith, by the way, what is that I see in one of your e-mail
addresses 
about Shavian "fish" nets?