Types

[Eric Sosman]

Why master, "foo31" of course, which is assigned as the value of the
string thingy 'foo'....


Which part of "obviously you can't do this in Standard C" was hard to
understand, Master? And also irrelevant since my example didn't
depend on a char*.


Newsflash: who said anything about arrays?

Ah, Grasshopper, you are so quick to take offense. Learn
patience like your cousin the locust, and you, too, may move
to the Wild West and demonstrate inner peace by kicking people's
heads off.

The message that launched all this, um, meditation was

> Bear in mind that several languages define string addition and most
> provide default conversion operators. It'd be fairly trivial to
> implement such an "addition" operator.

.... in which there is no mention of "Standard C." As a card-
carrying August Master, I naturally made certain allowances for
your preoccupation with matters temporal, and divined that you
believed that adding "string addition" to C would be a "fairly
trivial" extension to the language that the Transcendent Dennis
bestowed upon us. My response was in the nature of a Zen koan --
or maybe a Cen Coan -- intended to bring about that state called
"satori" in which you would suddenly and mystically come to know
that the data type "string" is more wrenching to C than grasshoppers
might suppose.

Of course, it might also bring about the state called "abhorri"
in which you suddenly and mystically bless your Revered Teacher
with the pleasant effluent of an AK-47. In that case, your R.T.
will demonstrate his unparalleled control over and indifference to
matters physical by swatting away the bullets like so many pesky
(but holy and precious in their own way) mosquitoes -- yet on the
whole, the R.T. would prefer not to put the matter to the test.

 

[Richard Heathfield]

Eric Sosman said:

Learn patience like your cousin the locust, and you, too, may move
to the Wild West and demonstrate inner peace by kicking people's
heads off.

When you can walk the callstack and leave no backtrace, then you will have
learned.

 

[Mark McIntyre]

Ah, Grasshopper, you are so quick to take offense.

Actually, I didn't take offense at all.

The message that launched all this, um, meditation was


... in which there is no mention of "Standard C."

Eggs act a mun doe.

As a card-
carrying August Master, I naturally made certain allowances for
your preoccupation with matters temporal, and divined that you
believed that adding "string addition" to C would be a "fairly
trivial" extension to the language

.... unfortunately your divining rod was in this case, broke...

--
Mark McIntyre

"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it."
--Brian Kernighan

 

[av]

--
Mark McIntyre

"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it."
--Brian Kernighan

this signature is OT, false and it irritate me: can you please change
it?

 

[Richard Heathfield]

av said:

this signature is OT,

It's a *sig block*. Since when did sig blocks have to be topical?

false

Really? Looks about right to me. But if you have a problem with it, take it
up with bwk, not Mark.

and it irritate me: can you please change it?

That's a reasonable request (although of course it's one that Mark is under
no obligation to meet). Here's another reasonable request, which you are
under no obligation to meet: can you please use a capital letter at the
beginning of each sentence, and use the appropriate verb ending that
matches the subject?

 

[Mark McIntyre]

this signature is OT,

Hardly.

false

The only person who can say that is bwk.

 

[av]

Hardly.

it not speak on "C language standard", it is about how difficult
should be to debug, an ot subject

The only person who can say that is bwk.

who is bwk? is bwk==Brian Kernighan?
the quote above is not valid at last for Who create the Universe
and possibly for many others

 

[av]

av said:

It's a *sig block*. Since when did sig blocks have to be topical?

for me no. but seeing all people says we have to be not ot and seeing
that the above quote is false too

Really? Looks about right to me. But if you have a problem with it, take it
up with bwk, not Mark.
bwk?


That's a reasonable request (although of course it's one that Mark is under
no obligation to meet). Here's another reasonable request, which you are
under no obligation to meet: can you please use a capital letter at the
beginning of each sentence, and use the appropriate verb ending that
matches the subject?

he is free to write what he wants but i say i not agree on that,
because my little experience shows that with some run time (for to
make to show these errors) and some debugging, is possible to correct
all showed errors; it is not matter the complexity of under code

the above quote could be right only for code not written by me or in
other not low languages i don't know
or if error doesn't show itself
or if the programmer write in an high language and not know assembly

 

[Chris Dollin]

this signature is OT, false and it irritate me: can you please change
it?

(a) It's a /signature/. It's topicality isn't at issue. Feel free to
complain about my signature(s) if you like.

(b) I certainly find debugging code harder than writing it. So it's
certainly not obviously false.

(c) It's a /signature/, not a peer-reviewed essay. It's pithy and
makes a point.

(d) If it irritates you, ignore it. Or introspect to find out why.

(e) It's a /signature/. It's not insulting, it's not offensive,
and it's relevant to programming and tells us something about
the mindset of one of the people most involved with C. I
think it's rather good, myself.

 

[Richard Heathfield]

av said:

for me no. but seeing all people says we have to be not ot

It's a sig block, for heaven's sake!

and seeing that the above quote is false too

Your claim that it is false is not self-evidently true.

bwk?

Brian W Kernighan, the author of the quote.

he is free to write what he wants but i say i not agree on that,
because my little experience shows that with some run time (for to
make to show these errors) and some debugging, is possible to correct
all showed errors; it is not matter the complexity of under code

Yeah, right.

 

[Bob Martin]

who is bwk? is bwk==Brian Kernighan?
the quote above is not valid at last for Who create the Universe
and possibly for many others

The only thing off-topic here is your attempt to bring religion into it.

 

[CBFalconer]

(a) It's a /signature/. It's topicality isn't at issue. Feel free to
complain about my signature(s) if you like.

(b) I certainly find debugging code harder than writing it. So it's
certainly not obviously false.

(c) It's a /signature/, not a peer-reviewed essay. It's pithy and
makes a point.

(d) If it irritates you, ignore it. Or introspect to find out why.

(e) It's a /signature/. It's not insulting, it's not offensive,
and it's relevant to programming and tells us something about
the mindset of one of the people most involved with C. I
think it's rather good, myself.

av is a pure Troll. Ignore it. Don't feed it. PLONK it.

 

[Random832]

this signature is OT, false and it irritate me: can you please change
it?

How's it false? Did Kernighan not actually say it?

 

[Random832]

(a) It's a /signature/.

Well - while everything else you said is valid... signatures are four
lines, not six.

 

[Chris Dollin]

2006-12-15 <[email protected]>,
Chris Dollin wrote:

Well - while everything else you said is valid... signatures are four
lines, not six.

Signatures are /supposed/ to be no longer than 4 lines, but I don't
think being six lines long makes it a non-signature; it just makes
it an /excessively long/ signature. (Does the grammar of signatures
really exclude long signatures, or is in the nature of an implementation
restriction?)

 

[Richard Heathfield]

Chris Dollin said:

Signatures are /supposed/ to be no longer than 4 lines, but I don't
think being six lines long makes it a non-signature; it just makes
it an /excessively long/ signature. (Does the grammar of signatures
really exclude long signatures, or is in the nature of an implementation
restriction?)

" - If you include a signature keep it short. Rule of thumb
is no longer than 4 lines. Remember that many people pay for
connectivity by the minute, and the longer your message is,
the more they pay."

---- RFC 1855

Whilst it is only a rule of thumb, it's a good one. And whilst this whole
RFC only has "informational" status, it is nevertheless worth taking
seriously - about as seriously as you would take an ISO/IEC 9899 footnote,
I guess. That is to say, okay, it's not normative, but it's still a
guideline that the clueful will observe under most circumstances.

 

[Mark McIntyre]

it not speak on "C language standard", it is about how difficult
should be to debug, an ot subject

maybe you didn't understand my sig in the post you are replying to. I
recommend you read it again. Meanwhile here's another that is quite
apposite...

 

[CBFalconer]

Signatures are /supposed/ to be no longer than 4 lines, but I
don't think being six lines long makes it a non-signature; it
just makes it an /excessively long/ signature. (Does the grammar
of signatures really exclude long signatures, or is in the nature
of an implementation restriction?)

Signatures are EVERYTHING that follows a "-- " sig marker. Some of
mine are considerably longer because they are pre-prepared help
messages, such as lists of references, or possibly designed to make
a political point.

s /supposed/recommended/

 

[Thad Smith]

That might be a better "first, tentative definition" for a type,
as it is the actual definition of a concrete type. But given a
specification, one can derive an algorithm ... and it depends
on whether one wants to start with the concrete or the abstract.

I should have been clearer. An algorithm maps an input sequence to an output
sequence. You can consider the bits within a object of a particular type to be an
input sequence, but without a definition of an output sequence, you can't define
an algorithm. Not all data types encode a real (or complex) number. A complex
type might contain the state of an order in process. It can be used for multiple
purposes and has no inherent output.

A type, to me, defines the meaning of the bits. What process (algorithm) you
apply to that depends on your objective.