C as a Platonic pathology

[spinoza1111]

This may all be true "in law", but you are the only one threatening
international law suits.  No-one else has "charged" negligence;  people
have simply, clearly, and repeatedly demonstrated unprofessional
carelessness by Schildt and his publisher.

In common practice, the same common practice that you promote so
loudly when discussing the way actual programming occurs,  it's a
simple matter to be negligent without having an intent to be negligent.
It's termed carelessness and, if being performed in an environemnt where
professionalism is required, then it demonstrates a likely unprofessional
approach to the task.

Nobody outside programming recognizes the programmer self-image,
because the tools programmers use are made for exchange value (profit)
and not for use value, and access to employment is controlled by
nonprogrammers who value subservience over skill (cf Phillip Kraft,
PROGRAMMERS AND MANAGERS, Springer 1978). Programmers are covered by
traditional employment law in which "the employee is worthy of his
hire" and Reagan-era rewriting of the (US) Uniform Commercial Code
which today holds software vendors harmless for errors.

Therefore neither Schildt nor McGraw Hill can be seriously charged
with negligence and to make this charge is criminal and civil libel.
In my opinion, he did the best job he could for the bulk of his
audience, and his laudatory Amazon reviews confirm that his haters are
the sort of people who can't stand Microsoft, who set up www.meankids.org.

To make errors, receive genuine criticisms, and to not correct those
errors is further demonstration of that unprofessionalism.  And to then
make your livelihood based on those uncorrected errors....  ??

It is a common programmer fantasy, inside corporations which do not
recognize them to be professionals, and who treat them shabbily, to
believe that they work within a *laager* of professionalism, and in
these fantasy *laagers* men set themselves up as experts. But when the
content of their expertise is investigated one discovers it to be a
mass of mostly disconnected rules which they are unable to express in
clear English.

 

[spinoza1111]

spinoza1111said:

spinoza1111said:

<snip>

Most of [Schildt's] "mistakes" are matters of interpretation and
of style, and as such McGraw Hill's editors decided that they
didn't rise to the level of errata.

I predict (for you are very predictable) that you will post no
evidence whatsoever to support your (insupportable) assertion.

Prediction confirmed. I conclude that you recognise the
insupportability of your claim.

<snip>

You are treating a mass of trivial details about fuckups as
knowledge, and you want to be respected for a knowledge of
unconnected details.

No, I'm treating a non-standard header as a non-standard header, and
noting that you are incapable of recognising it as such even after
having been told about it. You are unlikely to be taken seriously by
even the most addled and easily-led newbie unless you can demonstrate
the ability to learn.

I think the demonstration is lost on you.

Awww, now I'm going to have to take my irony meter in for repair.

You see, Dijkstra felt
that use of natural language and mastery of its syntax indicated
programming ability,

Probably true. You aren't seriously claiming that you are a master of
natural language, are you? You should warn people before making such
claims, even if only to give them time to put down their coffee.

Yes, I feel that I can better express ideas than you. However, writing
skills track reading skills, and if your reading skills are low,
you're not qualified to judge writing.

You give little evidence of reading skills. Typically, you seize upon
one interpretation of what others have written and treat that as the
only possible interpretation. This is the mark of an uneducated
person.

Nor is it irrational to believe that one text can have more than one
meaning. You interpreted Aho/Sethi et. al as saying that a "compiler"
must generate an external file of object code. That is only one
possible interpretation, since traditionally, processors such as the
Purdue Fast Fortran Translators were called "compilers" since they
fulfilled the need for a "compiler" by generating interpreted code,
while protecting their host systems from student crashes.

I don't know of anyone bright who believes that.

There are very few "bright" programmers. In my experience, most
genuinely intelligent people have left, or have been driven from, the
field, because as programming was "rationalized" commencing circa 1970
(that is, drained of actual reason in favor of bureaucracy), people
who expressed themselves well quickly gained reputations as
troublemakers. As early as 1978, sociologist Phillip Kraft had
discovered that programming status was reorganizing itself around the
willingness to be subservient to management and capitalism.

 

[spinoza1111]

spinoza1111said:


Down this corridor, fourth on the left. There's a notice on the door
saying "Gentlemen", but ignore that, and go right on in.

You set out to destroy people here in order to commercially promote an
unsaleable book, ass wipe, so don't you dare lecture me about courtesy
or collegiality.

I don't see why. It's not a standard header, so whether to use quotes
or chevrons is really a style issue. A good rule of thumb is "for


I don't know why you are bothering to use the header at all - you
don't need it. Nor do I know why you can't learn to ask politely. But

I will ask with collegiality when you reply with collegiality. OK,
it's not needed because it's in stdlib. You might have said that some
time ago.

the way in which C distinguishes between chevrons and quotes is
explained in K&R2 on p88, and the C Standard also explains it in
3.8.2 (C89) or 6.10.2 (C99).


There are no secrets here. You just have to learn to read. It
shouldn't take you more than a few years.

<nonsense and Shakespeare snipped>

Many thug Britons and pub bores really, really hate it when I quote
Shakespeare because it's always offensive to remind people deracinated
and decultured by an inferior and class-ridden educational system of
what they've been deprived of, especially when one's an American.

Sharif don't like it. You know he really hates it.

 

[spinoza1111]

spinoza1111said:



If such hypocrisy were intentional, it might be amusing. But I really
don't think it is intentional. I think you really can't see the
contradictions inherent in what you post. And that's rather sad.

Ass wipe, don't you dare, don't you dare lecture me about courtesy or
collegiality
You are using this newsgroup dishonestly, shamefully, and unethically.
I'd prefer people use "bad" language, since language recognizable by a
primitive scan
Is better than the deliberate attempt to destroy a man.

That would be "always", then.

Bullshit. "Collegiality" means questioning the professional bonafides
of a person only after working with them in meat space and after a
great deal of diligent research, UNLESS you question it basis on
violation of basic human norms of decency. You don't know whether
Schildt or Navia are bad persons, and you deliberately focus on
peephole images of their performance, isolated "errors" as opposed to
any bigger picture. Based on this limited input, you questioned
Navia's professional competence, and then used his response to
complete your case.

I question YOUR professional competence on violation of basic norms of
decency.

 

[Julienne Walker]

I will ask with collegiality when you reply with collegiality.

That literally doesn't follow. Assuming you live with the same
perception of time as the majority of us, one has to ask a question
before receiving a reply to that question. Therefore it's impossible
to know whether the reply will be polite before asking. In fact,
asking the question without any hint of civility is a damning action
because people have a tendency to act in kind. If you are rude, others
will be rude to you.

Personally, I don't feel you have the right to demand collegial
treatment when you yourself only act collegial when it's convenient.
It's a simple matter[1] to ignore the tendency to act in kind and
continue being civil. You'll find that sticking to your guns (your
guns in this case being collegial behavior) even in the face of
adversity will garner far more respect than winning the argument at
any cost.


[1] Simple in theory. In reality, refusing to act in kind is
surprisingly difficult for most because most feel (assuming they feel
as I do) that failing to defend oneself against all attacks hurts
one's pride. It's because of this difficulty that not giving in is so
worthy of respect. I believe it's a sign of good character.

 

[spinoza1111]

I will ask with collegiality when you reply with collegiality.

That literally doesn't follow. Assuming you live with the same
perception of time as the majority of us, one has to ask a question
before receiving a reply to that question. Therefore it's impossible
to know whether the reply will be polite before asking. In fact,
asking the question without any hint of civility is a damning action
because people have a tendency to act in kind. If you are rude, others
will be rude to you.

Personally, I don't feel you have the right to demand collegial
treatment when you yourself only act collegial when it's convenient.
It's a simple matter[1] to ignore the tendency to act in kind and
continue being civil. You'll find that sticking to your guns (your
guns in this case being collegial behavior) even in the face of
adversity will garner far more respect than winning the argument at
any cost.

[1] Simple in theory. In reality, refusing to act in kind is
surprisingly difficult for most because most feel (assuming they feel
as I do) that failing to defend oneself against all attacks hurts
one's pride. It's because of this difficulty that not giving in is so
worthy of respect. I believe it's a sign of good character.

I have already said that most people who make these kinds of
observations fail to diligently examine posts and read one or two of
the most recent posts in an exchange. They conclude from any verbal
self-defense that the victim is to be blamed, because people in
corporations have lost the ability to defend themselves.

In other words, Julienne, you're no den mother. You need to do your
homework. We have here a pattern: an unsuccessful computer author,
Richard Heathfield, is using this ng as a basis for deliberate attacks
on people based on his limited knowledge of computer science, attacks
that he is aware are amplified by the people he enables.

His behavior has ruined the utility of this newsgroup for its intended
purpose in precisely the same way I saw loudmouthed thugs at Bell-
Northern Research destroy the utility of structured walkthroughs.

If I type "**** this shit" it is easy to use the most basic form of
parsing (a regular expression) to determine that I've said it, and it
is a facile inference that I'm the deviant.

But it's too facile, since Heathfield has transformed this newsgroup
into a normalized deviance WITHOUT saying "**** this shit".

Do your homework and do not dare to lecture me.

 

[spinoza1111]

I will ask with collegiality when you reply with collegiality.

That literally doesn't follow. Assuming you live with the same
perception of time as the majority of us, one has to ask a question
before receiving a reply to that question. Therefore it's impossible
to know whether the reply will be polite before asking. In fact,
asking the question without any hint of civility is a damning action
because people have a tendency to act in kind. If you are rude, others
will be rude to you.

Personally, I don't feel you have the right to demand collegial
treatment when you yourself only act collegial when it's convenient.
It's a simple matter[1] to ignore the tendency to act in kind and
continue being civil. You'll find that sticking to your guns (your
guns in this case being collegial behavior) even in the face of
adversity will garner far more respect than winning the argument at
any cost.

This is nonsense, Julienne. People here who act with decency and in
good faith fail to phrase things the way the thugs here want find
things posted about them which contain no profanity but cost them
their jobs, and Heathfield enables and participates in this. When they
defend their interpretations, they are then labeled "drama queens".
These people are usually people who've accomplished something while
the residents here have accomplished nothing. That's why they are
here.

So **** this shit.

[1] Simple in theory. In reality, refusing to act in kind is
surprisingly difficult for most because most feel (assuming they feel
as I do) that failing to defend oneself against all attacks hurts
one's pride. It's because of this difficulty that not giving in is so
worthy of respect. I believe it's a sign of good character.

 

[Julienne Walker]

That literally doesn't follow. Assuming you live with the same
perception of time as the majority of us, one has to ask a question
before receiving a reply to that question. Therefore it's impossible
to know whether the reply will be polite before asking. In fact,
asking the question without any hint of civility is a damning action
because people have a tendency to act in kind. If you are rude, others
will be rude to you.

Personally, I don't feel you have the right to demand collegial
treatment when you yourself only act collegial when it's convenient.
It's a simple matter[1] to ignore the tendency to act in kind and
continue being civil. You'll find that sticking to your guns (your
guns in this case being collegial behavior) even in the face of
adversity will garner far more respect than winning the argument at
any cost.

[1] Simple in theory. In reality, refusing to act in kind is
surprisingly difficult for most because most feel (assuming they feel
as I do) that failing to defend oneself against all attacks hurts
one's pride. It's because of this difficulty that not giving in is so
worthy of respect. I believe it's a sign of good character.

I have already said that most people who make these kinds of
observations fail to diligently examine posts and read one or two of
the most recent posts in an exchange. They conclude from any verbal
self-defense that the victim is to be blamed, because people in
corporations have lost the ability to defend themselves.

In other words, Julienne, you're no den mother. You need to do your
homework. We have here a pattern: an unsuccessful computer author,
Richard Heathfield, is using this ng as a basis for deliberate attacks
on people based on his limited knowledge of computer science, attacks
that he is aware are amplified by the people he enables.

His behavior has ruined the utility of this newsgroup for its intended
purpose in precisely the same way I saw loudmouthed thugs at Bell-
Northern Research destroy the utility of structured walkthroughs.

If I type "**** this shit" it is easy to use the most basic form of
parsing (a regular expression) to determine that I've said it, and it
is a facile inference that I'm the deviant.

But it's too facile, since Heathfield has transformed this newsgroup
into a normalized deviance WITHOUT saying "**** this shit".

Do your homework and do not dare to lecture me.

For future reference (though I intend never to speak with you again),
I don't lecture from a position of ignorance. This is not the first
time you've told me to do my homework, and it's not the first time
I've either implied or directly explained that I've done my homework.
So at this point I've concluded that you're no different from those
pitiful people who love to hear themselves talk and refuse to listen
to anything else. Thus, you're not worth talking to.

Good bye, and good luck on your path to destruction.

 

[Dennis \(Icarus\)]

Possibly. Still, it's important in law to demonstrate intent in cases
of negligence, and the anti-Schildt creeps are charging negligence
without establishing intent. If they cannot prove intent, and because
they act with malice, any global claims they make about Schildt's
competence become criminal libels, made with intent to destroy his
livelihood.

If you're talking about law, that may be the case. I was referring simply
to...negligence, neglectful, carelessness, inattention.

Heathfield's only published one typo if that is what it is.

A typo is a an error.

It appears you know very little about compiler internals. Were you to

And you'd be wrong.

have worked inside a compiler you would know that in actuality, a

I wrote one for my master's thesis.

program ordinarily called an "interpreter" by its users will in all
cases (save the case I mentioned of Mouse, a language using characters
in reverse Polish that literally needed neither lexical nor
syntactical analysis) produce *** an intermediate representation ***.
This *** intermediate representation *** might be a file, or a data
structure in memory, or it might even be single messages passed from a
compiler coroutine to the interpretive coroutine, but in all case the
"interpreter" does "compilation".

Not at all, as that's an "***intermediate representation ***. In other
words, the *** intermediate representation*** is not an end-product.

A c compiler generates an object file on most platforms, which is then used
by the linker to generate an executable. Could generate an asm file just as
easily. While one culd argue that an obj file or asm file is also an
intermediate representation, the key point to remenber is that compilation
is finished.

A java compiler generates the java VM instructions. The JV can use a process
called just-in-time compilation to translate the VM instructions into
machine code. Note that they dn't call it "just-in-time interpretation".
A .net compiler can generate the MSIL, or platform-dependent instructions.

<irrelevant stuff snipped>

Dennis

 

[Dennis \(Icarus\)]

On Aug 21, 9:11 am, "Dennis \(Icarus\)" <[email protected]>
wrote:

You missed "traditionally" unless like Tevye you like tradition.

No, I clearly didn't miss that since...well...I typed it.
continue reading


Using compiler tech is different from being a compiler.

You may've missed this, so I left it intact. Feel free to disagree with Aho
et. al. though.



Dennis

 

[spinoza1111]

spinoza1111said:
Explain WHY in response to this post, because
I forget horseshit rules that derive from poor language design.
Nor do I know why you can't learn to ask politely.
I will ask with collegiality when you reply with collegiality.
That literally doesn't follow. Assuming you live with the same
perception of time as the majority of us, one has to ask a question
before receiving a reply to that question. Therefore it's impossible
to know whether the reply will be polite before asking. In fact,
asking the question without any hint of civility is a damning action
because people have a tendency to act in kind. If you are rude, others
will be rude to you.
Personally, I don't feel you have the right to demand collegial
treatment when you yourself only act collegial when it's convenient.
It's a simple matter[1] to ignore the tendency to act in kind and
continue being civil. You'll find that sticking to your guns (your
guns in this case being collegial behavior) even in the face of
adversity will garner far more respect than winning the argument at
any cost.
[1] Simple in theory. In reality, refusing to act in kind is
surprisingly difficult for most because most feel (assuming they feel
as I do) that failing to defend oneself against all attacks hurts
one's pride. It's because of this difficulty that not giving in is so
worthy of respect. I believe it's a sign of good character.

I have already said that most people who make these kinds of
observations fail to diligently examine posts and read one or two of
the most recent posts in an exchange. They conclude from any verbal
self-defense that the victim is to be blamed, because people in
corporations have lost the ability to defend themselves.

In other words, Julienne, you're no den mother. You need to do your
homework. We have here a pattern: an unsuccessful computer author,
Richard Heathfield, is using this ng as a basis for deliberate attacks
on people based on his limited knowledge of computer science, attacks
that he is aware are amplified by the people he enables.

His behavior has ruined the utility of this newsgroup for its intended
purpose in precisely the same way I saw loudmouthed thugs at Bell-
Northern Research destroy the utility of structured walkthroughs.

If I type "**** this shit" it is easy to use the most basic form of
parsing (a regular expression) to determine that I've said it, and it
is a facile inference that I'm the deviant.

But it's too facile, since Heathfield has transformed this newsgroup
into a normalized deviance WITHOUT saying "**** this shit".

Do your homework and do not dare to lecture me.

For future reference (though I intend never to speak with you again),

You'll be back.

I don't lecture from a position of ignorance. This is not the first
time you've told me to do my homework, and it's not the first time
I've either implied or directly explained that I've done my homework.

Where is the evidence that you've done your homework?

What you see and what is experienced by people like Navia here is
RADICALLY different. You look at the top of a LIFO stack and view n<5
layers.

Whereas Navia and other Heathfield victims have to see post after post
which questions their personal and professional standing in full view
of all and in which their words are twisted. Their reality is ignored
because the people that Heathfield enables don't read anything but the
top of the stack, and note that even Heathfield, since he's always
here (his publisher possibly paying him to blog), learns a thing or
two. For example, Heathfield has stated that I'm not a troll, details
at eleven, because he's obsessive enough to read posts. But as a
source of negative energy here, he enables others who don't, like you.

Just as the only responsible way for the United Nations to reconstruct
the experience of Serbia's victims in the Yugoslavian war was to build
an elaborate data base linking the top of the chain of command to
specific incidents, you will have to construct, as I and my attorney
may have to construct, a data base showing a pattern over time, and
although you express yourself in dulcet tones, and speak with the
tongues of men and angels...you have not done so.

The Puritan goal of an Anglo-Saxon net is to expose the soul to the
crowd, each member of which gets his fucking rocks off by baying for
the blood of the Chosen One or Drama Queen. As such, the Chosen one is
the only one to actually experience what's going down. Other people
have a fractured view whose specific contents depend on what posts
they have read, and you, baby, are no exception.

So at this point I've concluded that you're no different from those
pitiful people who love to hear themselves talk and refuse to listen
to anything else. Thus, you're not worth talking to.

Again, an image here is substituted for reality because you access
this resource LIFO, and in one ear and out the other. As it happens,
most of my numerous friends will confirm that I'm a devastatingly
charming good listener and most of my enemies pub ranters and bores.

Good bye, and good luck on your path to destruction.- Hide quoted text -

Because it's write-only, lifo, and in one ear and out the other, you
use mythology to stagger to your absurd conclusion. What am I to you,
Hecuba? Some Sixties-dreamtime character like Jim Morrison?

 

[spinoza1111]

On Aug 21, 9:11 am, "Dennis \(Icarus\)" <[email protected]>
wrote:



No, I clearly didn't miss that since...well...I typed it.
 continue reading

<snip>




Using compiler tech is different from being a compiler.




You may've missed this, so I left it intact. Feel free to disagree with Aho
et. al. though.

Dennis

The root words of both "compiler" and "interpreter" are verbs. To
compile is to use lexical and grammatical analysis to translate a
program to another form. This can be object code in a classic
compiler, p-code in a compiler-interpreter, a parse tree (sometimes
known as a DAG) for an optimizer, or even a source program in the case
of a source transliterator, such as a program to translate C to C
Sharp (rotsa ruck on that one).

To interpret is to build a paper or "nutty professor" computer which
exists only as simulated by your interpreter.

Compilers such as Schildt's, mine, the Purdue University Fast Fortran
Translator of old, or the IBM 1401 Fortran compiler, which generate
code for an interpreter, form a smallish but honorable subset of
compilers both IMO and in the literature. In a court of law, Richard
Heathfield's ignorance and misuse of terminology for commercial and
libelous ends will be exposed for what it is.

 

[Dennis \(Icarus\)]

The root words of both "compiler" and "interpreter" are verbs. To

Very good.

compile is to use lexical and grammatical analysis to translate a
program to another form. This can be object code in a classic

That's what I've always heard and read as lexical analysis and parsing.

compiler, p-code in a compiler-interpreter, a parse tree (sometimes
known as a DAG) for an optimizer, or even a source program in the case

Also known as an abstract syntax tree (AST).

of a source transliterator, such as a program to translate C to C
Sharp (rotsa ruck on that one).

To interpret is to build a paper or "nutty professor" computer which
exists only as simulated by your interpreter.

Also called...a simulator, or emulator.

So you agree now that when the interpreter has finished builing
theintermediate form, and thn interprets that result.....it's still an
interpreter, and not a compiler?

Compilers such as Schildt's, mine, the Purdue University Fast Fortran
Translator of old, or the IBM 1401 Fortran compiler, which generate
code for an interpreter, form a smallish but honorable subset of
compilers both IMO and in the literature. In a court of law, Richard
Heathfield's ignorance and misuse of terminology for commercial and
libelous ends will be exposed for what it is.

Good to see you consider them to be compilers.

Dennis

 

[spinoza1111]

Very good.

Don't patronize me about matters you don't fully understand, Dennis.

That's what I've always heard and read as lexical analysis and parsing.

"Necessary" conditions for compiling (to compile): lexical analysis
and parsing (both are forms of parsing kept apart for convenience and
because lexical analysis can be performed simply, in one pass, without
a memory).

Generation of code for a real machine is not a necessary condition for
compiling because in common parlance by native English speaking
computer technicians and scientists, intepretive compilers such as the
examples I gave are said to be compilers, that compiled.

Learn linguistics (including what constitutes correct usage) and logic
(including the concept of necessary and sufficient condition) before
posting your views.

I agree that the usage of "compiler" to refer to interpretive
compilers (such as early Pascal and my examples) is in some desuetude
owing to the fashion for mindless speed and the evolution of backend
theory which has made retargetable code generation feasible.
Nonetheless, contemporary historians of technology refer to old
compiler/interpreters as compilers. Cf. my article in the Spring
Summer edition of IEEE Transactions in the History of Software
(Nilges).

Also known as an abstract syntax tree (AST).

Thank you.

Also called...a simulator, or emulator.

So you agree now that when the interpreter has finished builing
theintermediate form, and thn interprets that result.....it's still an
interpreter, and not a compiler?

You cannot properly label it an interpreter if it compiles, viz., does
lexical and syntactical analysis.

Therefore, Richard Heathfield has libeled Schildt on this matter and
many other ways.

Good to see you consider them to be compilers.

Well, I suppose you could get the computing Tsar of some mythical
kingdom to declare that no Compiler may be so labeled unless it
generates object code, but rotsa ruck on that one.

 

[Dennis \(Icarus\)]

On Aug 22, 10:58 am, "Dennis \(Icarus\)" <[email protected]>
wrote:

Don't patronize me about matters you don't fully understand, Dennis.

Your perception on the latter is faulty at best. If you want to avoid my
being patronizking towards you, I'd suggest you stop being patronizing
towards me. After all, surely you remember all those posts abou how you'll
act in the manner in whihc you're being treated?

"Necessary" conditions for compiling (to compile): lexical analysis
and parsing (both are forms of parsing kept apart for convenience and
because lexical analysis can be performed simply, in one pass, without
a memory).

Generation of code for a real machine is not a necessary condition for
compiling because in common parlance by native English speaking
computer technicians and scientists, intepretive compilers such as the
examples I gave are said to be compilers, that compiled.

I never said that it was. A compiler took X and generated Y as the target.
Aho & Sethi used the final result to distinguish compilers from
interpreters.

Learn linguistics (including what constitutes correct usage) and logic
(including the concept of necessary and sufficient condition) before
posting your views.

I have, and i can provide many, manhy references in support of my views on
what constitutes a compiler vs an interpreter. Did so in another post.

Thank you.

Not sure what you're thanking me for.

You cannot properly label it an interpreter if it compiles, viz., does
lexical and syntactical analysis.

Incorrect.
Lexical analysis and synatical analysis has to be done by both an
interpreter and a compiler, hence you cannot use that as the basis to
distinguish one from the other.

Dennis

 

[spinoza1111]

spinoza1111wrote:

The body of known compiler work for Schildt (or you) is extremely limited.. From what I have seen little more than a parser, a task
that takes a few man days when writing a compiler compared to several man months to complete most compilers.

The recursive descent parser for Quick Basic that I wrote took a few
man-days. The interpreter, a few more. The assembler (which translated
the symbolic labels in the p-code) a few more. The graphical user
interface which allows the user to see what's happening in detail, a
few more. Oh yes, the lexical analyzer, a few more.

The complete and literate comments? A few more.

It adds up.

By the way, here is what one anonymous Amazon reviewer of the
associated book said:

"Most books contain code written just for the book: small and simple
examples. This book contains a huge amount of code that has obviously
been under development and evolving for a long time. The code has a
level of documentation, error checking and self-consistency testing
that is rare to see even in commercial code, much less sample code for
a book. The author may have written the original compiler in a few
days, but he himself mentions that he then went on to rewrite it
completely as clearly obvious by looking at the sample code."

Neither Schildt nor I would claim that our compilers are comparable in
scale to Navia's lccwin32, or to a commercial compiler. A compiler
that generates object code is a valuable asset, and takes longer. I
produced one such at Bell Northern Research, including complete object
code generation, and although it didn't take an order of magnitude
longer, it took ... longer.

However, I believe that these compilers that you mention, which take
"months" are modern myths and urban legends. Commercial compiler shops
don't only use evolutions of yacc and lexx they also use code
generators for the back-end, and these tools are valuable assets in
themselves.

To speak as if there are compilers different in kind merely because
they are written by mythical "real" programmers, "real" men, is a
Lacanian symptom with which I am very familiar in technology. Deprived
(cf Susan Faludi, "Stiffed: the Betrayal of the American Male) of any
chance at an independent existence by corporate life, programmers have
to posit a phallus somewhere so as to legitimate the economic
arrangements (no affordable health insurance in the richest country in
the world) which oppress them lest they go bat shit.

In fact, Navia comes closest to this mythical creature...and look how
you treat him!

 

[spinoza1111]

Your perception on the latter is faulty at best. If you want to avoid my
being patronizking towards you, I'd suggest you stop being patronizing
towards me. After all, surely you remember all those posts abou how you'll
act in the manner in whihc you're being treated?

Our perceptions, which are often more valid, take place in time. We
come in here to have the resident thugs trash us when we disagree with
their opinions. We defend ourselves like men, and then are called
drama queens and trolls.

It only seems like it takes place outside of time, and it only seems
that all are guilty.

I never said that it was. A compiler took X and generated Y as the target..
Aho & Sethi used the final result to distinguish compilers from
interpreters.


I have, and i can provide many, manhy references in support of my views on
what constitutes a compiler vs an interpreter. Did so in another post.

I will agree that owing to the greater power of full-feature
compilers, many younger programmers prefer to refer to interpretive
compilers as interpreters. Nonetheless, as I've shown (did you miss
it) all "interpreters", save for the limit case of a language like
Mouse (do you know it?) do compilation in Aho/Sethi's sense, therefore
Richard Heathfield is guilty of libel qed.

Not sure what you're thanking me for.

Reminding me that a DAG is an AST, and a Snark is a Boojum.

Incorrect.
Lexical analysis and synatical analysis has to be done by both an
interpreter and a compiler, hence you cannot use that as the basis to
distinguish one from the other.

OK, so for you, object code generation for a real machine is a
necessary condition for being a compiler. That's not common use.

 

[spinoza1111]

comp.lang.c FAQ list · Question 10.8a

Q: What's the difference between #include <> and #include "" ?

http://c-faq.com/cpp/inclkinds.html

-Beej

No, you have to explain why. I write fast. You should do. The prose at
the link is beastly.

Anyway, the angle bracket includes from the standard location, the
quote includes from a location that can be defined and can hold your
"own" includes. This BTW is just ridiculous. Such a detail does NOT
belong in a programming language for grownups.

In OO, the standard library is not replaceable. Instead, it is
overridable in the source code. While this doesn't prevent all
problems it prevents many accidental errors.

 

[spinoza1111]

Your perception on the latter is faulty at best. If you want to avoid my
being patronizking towards you, I'd suggest you stop being patronizing
towards me. After all, surely you remember all those posts abou how you'll
act in the manner in whihc you're being treated?






I never said that it was. A compiler took X and generated Y as the target..
Aho & Sethi used the final result to distinguish compilers from
interpreters.


I have, and i can provide many, manhy references in support of my views on
what constitutes a compiler vs an interpreter. Did so in another post.




Not sure what you're thanking me for.




Incorrect.
Lexical analysis and synatical analysis has to be done by both an
interpreter and a compiler, hence you cannot use that as the basis to
distinguish one from the other.

BTW, this is a MASSIVE petitio principii, or question begging
argument:

(Ex)(Ey)[isAnInterpreter(x) && isACompiler(y) &&
doesLexicalAndSyntaxAnalysis(x) && doesLexicalAndSyntaxAnalysis(x) &&
x!= y]

therefore (you conclude) that doesLexicalAndSyntaxAnalysis "cannot" be
used.

Absurd, and heart-breakingly so.

In the first line, you assert without proof that x, which interprets
bytecodes after it does lexical analysis and parsing,
isAnInterpreter...and not a compiler.

But this is what is at issue. You haven't refuted my claim that
lexical analysis and parsing add up to a compiler (I concede that a
compiler with lexical analysis, parsing, and object code generation to
a real machine is in more cases a larger project).

Instead, you've assumed what you wanted to prove.

Nice work...if you can get it.

 

[Dennis \(Icarus\)]

On Aug 22, 8:42 pm, "Dennis \(Icarus\)" <[email protected]>
wrote:

Incorrect.
Lexical analysis and synatical analysis has to be done by both an
interpreter and a compiler, hence you cannot use that as the basis to
distinguish one from the other.

BTW, this is a MASSIVE petitio principii, or question begging
argument:

(Ex)(Ey)[isAnInterpreter(x) && isACompiler(y) &&
doesLexicalAndSyntaxAnalysis(x) && doesLexicalAndSyntaxAnalysis(x) &&
x!= y]

therefore (you conclude) that doesLexicalAndSyntaxAnalysis "cannot" be
used.

Absurd, and heart-breakingly so.

It is heartbreakingly absurd, in that, what I said, what that both
inteprerters and compilers have to perform lexical analysis and synatical
aalysis, hence you cannot claim an interpreter is a compiler on that basis.
Go reread your claim above where you state
"You cannot properly label it an interpreter if it compiles, viz., does
lexical and syntactical analysis."

In the first line, you assert without proof that x, which interprets
bytecodes after it does lexical analysis and parsing,
isAnInterpreter...and not a compiler.

How the interpreter is implemented, and what it does with the statement
afterwards, is up to the system.

But this is what is at issue. You haven't refuted my claim that
lexical analysis and parsing add up to a compiler (I concede that a
compiler with lexical analysis, parsing, and object code generation to
a real machine is in more cases a larger project).

They'll also add up to an interpeter, as lexical analysis and syntax
analysis are phases of the general problem - laguage translation.

Dennis