orthogonal?
Author Message
orthogonal?

Orthogonal in reference to a comp lang. Well, I've read this
for quit a while and finally cracked a dictionary and got
stuff like:
* mutually perpendicular
* statistically independent
*blank stare - low background buzzzz*
Could someone define orthogonal in reference to a comp lang so
even I might understand it.
Thanks,
M. Dearman
--

Thu, 07 Feb 2002 03:00:00 GMT
orthogonal?

Quote:

> Orthogonal in reference to a comp lang. Well, I've read this
> for quit a while and finally cracked a dictionary and got
> stuff like:
> * mutually perpendicular
> * statistically independent
> *blank stare - low background buzzzz*
> Could someone define orthogonal in reference to a comp lang so
> even I might understand it.

When things are "orthogonal" or "perpendicular", if
their scalar products 0.  If a computer language is C, and a topic is
T, then TC = 0 means there is no projection of T onto C; that is, it
represents a completely differmt dimension.

It is rare to find any two areas of work U and V for which UV = 0.  They
usually both have some non-zero components on some dimension.  I believe
that people use "orthogonal" loosely here to mean that |TC| < epsilon,
where epsilon is some arbitrarily assigned small value.

--

__________________________________________________________
Fight spam now!
Get your free anti-spam service: http://www.brightmail.com
--

Thu, 07 Feb 2002 03:00:00 GMT
orthogonal?

Quote:

> Orthogonal in reference to a comp lang. Well, I've read this
> for quit a while and finally cracked a dictionary and got
> stuff like:
> * mutually perpendicular
> * statistically independent
> *blank stare - low background buzzzz*
> Could someone define orthogonal in reference to a comp lang so
> even I might understand it.

Look in the Jargon File (aka _New Hacker's Dictionary_), available
online at http://www.tuxedo.org/~esr/jargon/:

Quote:
> [from mathematics] Mutually independent; well separated; sometimes,
> irrelevant to. Used in a generalization of its mathematical meaning
> to describe sets of primitives or capabilities that, like a vector
> basis in geometry, span the entire `capability space' of the system
> and are in some sense non-overlapping or mutually independent. For
> example, in architectures such as the PDP-11 or VAX where all or
> nearly all registers can be used interchangeably in any role with
> respect to any instruction, the register set is said to be
> orthogonal. Or, in logic, the set of operators `not' and `or' is
> orthogonal, but the set `nand', `or', and `not' is not (because any
> one of these can be expressed in terms of the others). Also used in
> comments on human discourse: "This may be orthogonal to the
> discussion, but...."

Actually, the second example (logic operators) isn't really good -- it
mixes independence with orthogonality.

You may find probability terminology a help: 2 random variables are
orthogonal if their covariance vanishes.  For finite probability
spaces, this implies that they're independent.  That is, "one has no
effect on the other".

This is often considered a Good Thing -- it means you can get away
with understanding A and B separately, without needing to understand
A+B.

--

Compugen Ltd.          |Tel: +972-2-6795059 (Jerusalem) \  NEW IMPROVED URL!
72 Pinhas Rosen St.    |Tel: +972-3-7658520 (Main office)`--------------------
Tel-Aviv 69512, ISRAEL |Fax: +972-3-7658555  http://3w.compugen.co.il/~ariels
--

Thu, 07 Feb 2002 03:00:00 GMT
orthogonal?

Quote:
>Could someone define orthogonal in reference to a comp lang so
>even I might understand it.

It refers to two properties of a language that do not interact in any
way.  I.e. a change to one will not result in problems with the other.

Francis Glassborow      Journal Editor, Association of C & C++ Users
64 Southfield Rd
Oxford OX4 1PA          +44(0)1865 246490
All opinions are mine and do not represent those of any organisation
--

Thu, 07 Feb 2002 03:00:00 GMT
orthogonal?

Orthogonal in reference to a comp lang. Well, I've read this
for quit a while and finally cracked a dictionary and got
stuff like:
* mutually perpendicular
* statistically independent
*blank stare - low background buzzzz*
Could someone define orthogonal in reference to a comp lang so
even I might understand it.

It might be best if you'd give us the context in which this term was
used.  Did it just say `C is orthogonal' or was there more?
--
"...dans ce pays-ci il est bon de tuer de temps en temps un amiral
pour encourager les autres."
--Voltaire, _Candide_
--

Thu, 07 Feb 2002 03:00:00 GMT
orthogonal?

MD> Could someone define orthogonal in reference to a comp lang so even
MD> I might understand it.

:orthogonal: [from mathematics] adj. Mutually independent; well
separated; sometimes, irrelevant to.  Used in a generalization of
its mathematical meaning to describe sets of primitives or
capabilities that, like a vector basis in geometry, span the
entire `capability space' of the system and are in some sense
non-overlapping or mutually independent.  For example, in
architectures such as the PDP-11 or VAX where all or nearly all
registers can be used interchangeably in any role with respect to
any instruction, the register set is said to be orthogonal.  Or, in
logic, the set of operators `not' and `or' is orthogonal,
but the set `nand', `or', and `not' is not (because any
one of these can be expressed in terms of the others).  Also used
in comments on human discourse: "This may be orthogonal to the
discussion, but...."

greetings,
Tom

--

Thu, 07 Feb 2002 03:00:00 GMT
orthogonal?

Quote:

> Orthogonal in reference to a comp lang. Well, I've read this
> for quit a while and finally cracked a dictionary and got
> stuff like:
> * mutually perpendicular
> * statistically independent
> *blank stare - low background buzzzz*
> Could someone define orthogonal in reference to a comp lang so
> even I might understand it.

My understanding or orthogonal in a computer is when any combination
of two different things are legal.

Examples might help the most.

C operators and types are orthogonal.  Any combination of the
operators ++, +, -, * works with with any of the types char, int,
float.  It would not be orthogonal if + were restricted to int.

The 68K register set is orthogonal in that all the address registers
A0-A7 work with all instructions and addressing modes.  The x86 is not
orthogonal since many of the CPU registers are restricted to certain

As I remember, the VMS command line interface is orthogonal since
the command flags are consistent across commands.

The UNIX commands are not orthogonal, since the flags vary with the
command.  rm -i means inquire before executing the command, while
rmdir does not accept -i.
--

--

Fri, 08 Feb 2002 03:00:00 GMT
orthogonal?

Quote:

> It refers to two properties of a language that do not interact in any
> way.  I.e. a change to one will not result in problems with the other.

As a concept borrowed and adapted from linear algebra,
orthogonality can be understood as "independence".
It most often arises in a context where lack of
orthogonality is being bemoaned.  E.g. in C, there
are conceptually orthogonal notions of scope, lifetime,
and linkage, but the language doesn't provide separate
keywords for each of these attributes; rather, it
overloads some keywords such as "static" and depends on
context to determine the intended meaning.
--

Sat, 16 Feb 2002 03:00:00 GMT
orthogonal?

writes:

Quote:

>C operators and types are orthogonal.  Any combination of the
>operators ++, +, -, * works with with any of the types char, int,
>float.  It would not be orthogonal if + were restricted to int.

This is my understanding of 'orthogonal' when referring to languages.  However
I am used to hearing this term in reference to assembly languages.
Chick Racer
--

Sat, 23 Feb 2002 03:00:00 GMT

 Page 1 of 1 [ 9 post ]

Relevant Pages