Re: signed/unsigned integer conversion for right shift seems

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



"Also sprach Jonathan Wakely:"
> > Correct. I am exact, which is the minimum anyone can be.
> 
> I prefer this from https://www.merriam-webster.com/dictionary/sophistry
> Thus sophist (which comes from Greek sophist?s, meaning "wise man" or
> "expert") earned a negative connotation as "a captious or fallacious

It is ambiguous in the sense that both it and its opposite are possible
dictionary meanings, a rare distinction.  Your choice was number 2 in
the usual dictionary list.  I preferred number 1 both on the basis of
accuracy in this instance and popularity as per the dictionary ordering.

I could care less ;)

It illustrated that you should be more careful, so I allowed it.

> reasoner." Sophistry is reasoning that seems plausible on a
> superficial level but is actually unsound, or reasoning that is used
> to deceive.

Unfortunately for your likely intention, my reasoning is both deep and
correct and it is on display preciely in order that you may be able to
take issue with it at any point in it.

Please go ahead.

Otherwise I'll take that as childish meaningless nonsense and be happy.

> >> The insight you finally gave me credit for wasn't even mine
> > It is.  Kindly point to somebody who says it before you?
> https://gcc.gnu.org/ml/gcc-help/2018-02/msg00019.html

(1) That doesn't work as a method of answering.  You are supposed to
    quote the precise passage you mean and 
(2) explicitly state the reasoning that forms your opinion of it

Otherwise you are being ambiguous.  I'll quote 19 in its entirety,
noting beforehand that it does not say it. My (1):

  See the definition of "usual arithmetic conversions" in the text of the
  standard and note that they are not applied to bitwise shift arguments
  (they undergo integer promotions individually).

Where in there do you see it said?  I don't.  Let me remind you that the
single keyword was BECAUSE. 

Here's my (2): He does not reveal any reasoning, so that is empty of
content because we require reasoning to connect the dots  between the
standard and the conclusion (no conversions).

To explain further the standard contains many statements, none of them
is that conversions shall not be applied in this case, so one requires
reasoning to connect things in the standard with the conclusion one
wants.  He does not provide it.

You did, with that ONE WORD ("because"), for the first time. Do you see
a "because" or a "therefore" or a "since" or an "as" or a "whence" etc.
in msg 19?

That single word "because" was enough of a clue for me to construct the
whole probable chain of reasoning that you couldn't be arsed to express.

How do you think people can follow your reasoning if you do not
enunciate it?

That "because" was enough on its own to start me looking through the
whole standard word by word first for the DATA (such as that paraphrased
in msg 19) that might permit the logic chain to be constructed on top of
it, and secondly for the RULES OF DEDUCTION that would form the body of
the logic chain.

Do you get it now?

6.1 provides the rule

   (8.3.18 =/> ~p /\ 6.5 =/> ~p) => p

That form of reasoning would be invalid without its authority. Invalid
reasoning is invalid reasoning, full stop. Don't explain your rule
of deduction and what authorizes it and you are reasoning invalidly.

If you cannot read that, or for some other reason you find your mind
just "skips over it" witout taking it in, please LET ME KNOW and I will
rephrase in less formal language.  It is not unusual for people to be
"symbol-blind".  I'm phrasing it in a form that I natively expect you to
be familiar and comfortable with, but if I am wrong with that guess,
you'll have to tell me.  It's no skin one way or another.

> https://gcc.gnu.org/ml/gcc-help/2018-02/msg00020.html

I'm not going to bother because your credibility is already zero with
these things, having shown so many times (e.g.  19 above) that you fail
to get it.

Oh logicks, count thy number that I have written, proved complete, and
sound.

> You may claim it wasn't precise enough for you, but that's your

I have no idea or opinion on preciseness. It is simply irrelevant,

> problem for being an idiot. Nobody here is under any obligation to

Reading anything I write should tell you and anyone what is really up.

I can't make you see what you appear not to be able to see although the
huge majority of people in the world can ...  All I can comfort you with
is that it is a recognized generic problem of engineers. And
administrators.  Also poor lawyers. Actually, poor anything.


Yea! It's common! Unfortunately, no cure is available, but it
begins with the realization that there may be something there that
you are missing. Treat it like one of those aliens you can only see if
you don't look straight at them ;).

> provide you with a formal education in the C standard. You were given

Nobody is doing so and nobody is obliged to do anything, except that it
is a simple and universal courtesy from and to human beings to explain
their reasoning.

To put it another less charitable way, people who can't see or explain
their reasoning are subhuman nazis, and so one aims to show one is not
by always showing ones reasoning.  That will leave it vunerable to
challenge and the idea is precisely that.

> the answers, and several clues how to interpret the standard if you

You still don't get it, but we know that. It is YOU that I am
interpreting, not the standard. YOU need to show YOUR reasoning.

Only that enables its worth to be judged, and that is the common
courtesy that one human being leaves another.

> cared to. You chose to ignore or misinterpret them, because they

I have done nothing of the kind. 

It seems to me that when I have asked you to show

   "What, specifically, do you see in 6.5.7 that allows the conversion
   specified by the general rule of conversions ... NOT to be applied?"

that YOU are misinterpreting that as a request to be pointed at WHERE 
in the standard is the BASIS for a deduction that conversion may not
be applied (in fact must not, as it turns out).

I did not ask that.

I asked for WHAT IS THE LOGICAL RULE. Nothing else can be of interest.

I was expecting you to point at what in the end turned out to be 6.3
first paragraph final sentence, where it announces the DEDUCTION RULE
that says it's either in 6.3.18 or in 6.5 or it is not allowed.

I put it down to your blindness to the fact that your own and the
standard's logic exists and are things. In the standard's case, the
rules of the logic have to be written down somewhere, because that
is what standards do. 

You kept on not referring me to it and claiming that you had. Indeed,
you gave no reasoning at all for your statements, thus saying
absolutely nothing.

It is a problem of "level".

> didn't suit you. Call that exactness if you wish, I call it arrogance
> and pissiness.

I call it normal, ordinary, everyday stuff, that every human being
needs and does in order to do their own stuff in a coherent fashion.  To
be human, you show you have the self-knowledge to be able to explain
your own mental workings. It's only machines (well, and animals) that
can't.

I hope that the penny is if not dropped, on it way into the slot.
Remember that ONE WORD ("because") was enough for me, and please
reflect.

Regards

PTB




[Index of Archives]     [Linux C Programming]     [Linux Kernel]     [eCos]     [Fedora Development]     [Fedora Announce]     [Autoconf]     [The DWARVES Debugging Tools]     [Yosemite Campsites]     [Yosemite News]     [Linux GCC]

  Powered by Linux