Redesigning system classes

[ConvertFromOldNGs]

by JADE Kid - Ray Hidayat >> Mon, 15 Jan 2001 21:53:26 GMT

I don't understand you. Long words.

Did you know that two lines are perpindicular if they're gradients are negative reciprocals to each other?

What's an oxymoron?

--
Ray Hidayat
JADE Kid - 2000
www.jadekids.com

[ConvertFromOldNGs]

by Robert Barr >> Mon, 15 Jan 2001 22:47:32 GMT

I don't understand you. Long words.

I don't have a definition of "aggregration" to hand, but I suspect it implies multiplicity, which is at odds with one-to-one.

Did you know that two lines are perpindicular if they're gradients are negative reciprocals to each other?

Does this still apply to gradients defined in more than 2 dimensions?

What's an oxymoron?

Ox·y·mo·ron. (n) A combination of contradictory or incongruous words.

(From late Greek, oxymOros pointedly foolish, from Greek oxys sharp,
keen + moros foolish

Usually a figure of speech, references are somewhat tongue in cheek.

Some well known examples: "act naturally", "hard water", "tentative deadline", "ready-to-eat frozen foods", "athletic scholarship", "congressional ethics", "pretty ugly", "even odds", "Great Depression", "terribly nice", "planned serendipity", "military intelligence",
"freezer burn", "junk food", "jumbo shrimp", "political promise", "High-tech unions", "Carefully selected volunteers", "male/female intellect" (delete as appropriate), "religious tolerance", "serious computing", "network centered" ...

more?

http://www.specsci.com/donspage/htmldocs/oxymoron.htm, http://my.voyager.net/shotwell/oxymoron.html

[ConvertFromOldNGs]

by Warwick Hunt >> Mon, 15 Jan 2001 22:55:30 GMT

Surely one of the best is "Microsoft Works".
I'm beginning to wonder about "Jade Kid" as well.
Warwick

[ConvertFromOldNGs]

by JADE Kid - Ray Hidayat >> Mon, 15 Jan 2001 23:05:36 GMT

Wow. What a nice explanation.

I was just kidding about the grandients thing. It is true, but I just needed some long words.

--
Ray Hidayat
JADE Kid - 2000
www.jadekids.com

[ConvertFromOldNGs]

by Michael Jones >> Tue, 16 Jan 2001 0:02:34 GMT

one to one relationships are generally between two classes,
so the relationship is an aggregration made up of parts of those two classes i.e the keys. So IMO it is probably okay.

[ConvertFromOldNGs]

by Wilfred Verkley >> Tue, 16 Jan 2001 0:34:40 GMT

I don't have a definition of "aggregration" to hand, but I suspect it implies multiplicity, which is at odds with one-to-one.

I sense some duplicity in your accusation of multiplicity.

My understanding of aggregation in an Object Oriented sense is forming larger objects out of smaller ones. Its a "part-of" relationship, and there is a concept of ownership and dependency as well.

i.e. a car object has four wheel objects, a motor object, and a bumper object. All of these objects are "part-of" the car and in my system, have not independent existence outside of the car. The car<->motor relationship is an aggregation relationship even though its a one-to-one.

[ConvertFromOldNGs]

by Krull >> Tue, 16 Jan 2001 2:46:10 GMT

I don't have a definition of "aggregration" to hand, but I suspect it implies multiplicity, which is at odds with one-to-one.

The term 'aggregation' does seem to imply multiplicity, but this could be a multiplicity of 0, 1 or n. In object modeling 'aggregation' is sometimes used to describe a 'part-of' relationship. A 'car has one motor' is an example of a one-to-one aggregation in this context. In a 'component reuse' context, the term containment is also used to distinguish between a form of 'uses' or 'has-a' relationship were clients only deal with the 'outer object' and are unaware of the 'contained object', versus 'aggregation' where client access to the inner object is allowed.

Did you know that two lines are perpindicular if they're gradients are negative reciprocals to each other?

Does this still apply to gradients defined in more than 2 dimensions?

Hmmmm a bit off topic, but yes it applies in 3-dimensions at least in 'Euclidean Space', which like most specialisations still has its uses. This property can even be extended to a line intersecting a curved surface. I am not sure if this remains true in non-Euclidean geometry (e.g. Gaussian, Reimanian, Hilbert space etc. ), but on the surface (pun accidental) I can't see why it shouldn't hold true.

Perhaps Ray was thinking of the term 'orthogonal', sometimes used to mean two mutually independent or separate concepts and this lead him to recall this geometrical theorem containing a synonym to orthogonal, namely 'perpendicular' - the mind boggles

PS: Here is an exercise especially for Ray. Can you prove or just find a published proof that the theorem you stated is ' true ' in Hilbert Space, here is a starting point: http://www1.shore.net/~ndm/java/mmexplorer1/mmhil.html