T
Tim Streater
spinoza1111 said:
Clap, clap. Another content-free post from Spinny. Tell me, have you
always had attention deficit disorder?
Clap, clap. Another content-free post from Spinny. Tell me, have you
always had attention deficit disorder?
S
Seebs
Yeah, pretty much.
This seems to be pretty much par for the course. It's entirely possible
that in some of his attempts to open a private channel, he used more
temperate language... But he has never, at any point, done anything
involving me that did not lead immediately to, if I showed any sign
of thinking he was anything but batshit insane, asserting that the right
thing for me to do is remove my page about C:TCR.
Why?
Because that page is why the Wikipedia people finally said "no, there isn't
any more citation needed for the thing saying Schildt's writings are not
well regarded among the technical community."
I would dearly love to know the *actual* reasons for which Nilges is so
incredibly obsessive on this topic. Did he get fired from a job for owning
a Schildt book, or for something he did which relied on information he
derived from one of Schildt's books? Did he lose some political battle
in which the ostensible subject was something to do with the stack, which
seems to be a particular point of obsession? I have no idea, but I've
always been curious. The reasons Nilges provides are consistently obviously
fake -- the mere fact that he switches from one to another with no hiccups
as soon as he realizes a given one won't play with a particular audience
makes it clear that they're all just excuses and rationalizations.
I'd love to know the real reason.
.... The interesting thing, BTW, is that I've been having multiple fascinating
discussions with people who disagree with me, while Nilges is the one who
calls someone an "asshole" for correcting him, then insists that he's not
being insulting.
I responded to him about twice before concluding that he was good only
for laughs. That said, there's a substantial point here -- Nilges has
never figured out how to view the entirety of long posts through Google
Groups, so far as I can tell, so substantive questions which are near
the end of a longish post are invisible to him. (This is why you sometimes
see Google's text for "and there's more to this post" quoted in his
replies.)
But I have no interest in privately communicating with someone who appears
to habitually lie about what he or other people said. I'd rather keep
the record out in the open. Not because this will prevent him from
blatantly misrepresenting things, but because it will prevent him from
fooling anyone*.
-s
[*] Obviously, Kenny is taken in by anything anyone writes which criticizes
people he considers "regs", but I don't think that counts.
S
Seebs
I would point out that his behavior is not particularly consistent with
attention deficit disorder. What he's showing is much too specialized;
he only loses interest in things when it starts looking like it won't make
him look good. His babbling about structured programming looked like a
great way to assert dominance over poor lil' uneducated me, but he forgot
that there are other posters here, some of whom have done a lot of formal
study of computer science.
-s
W
Willem
Seebs wrote:
)> Clap, clap. Another content-free post from Spinny. Tell me, have you
)> always had attention deficit disorder?
)
) I would point out that his behavior is not particularly consistent with
) attention deficit disorder. What he's showing is much too specialized;
) he only loses interest in things when it starts looking like it won't make
) him look good. His babbling about structured programming looked like a
) great way to assert dominance over poor lil' uneducated me, but he forgot
) that there are other posters here, some of whom have done a lot of formal
) study of computer science.
Narcissism perhaps ? I think that term came up before in this froup.
SaSW, Willem
--
Disclaimer: I am in no way responsible for any of the statements
made in the above text. For all I know I might be
drugged or something..
No I'm not paranoid. You all think I'm paranoid, don't you !
#EOT
)> Clap, clap. Another content-free post from Spinny. Tell me, have you
)> always had attention deficit disorder?
)
) I would point out that his behavior is not particularly consistent with
) attention deficit disorder. What he's showing is much too specialized;
) he only loses interest in things when it starts looking like it won't make
) him look good. His babbling about structured programming looked like a
) great way to assert dominance over poor lil' uneducated me, but he forgot
) that there are other posters here, some of whom have done a lot of formal
) study of computer science.
Narcissism perhaps ? I think that term came up before in this froup.
SaSW, Willem
--
Disclaimer: I am in no way responsible for any of the statements
made in the above text. For all I know I might be
drugged or something..
No I'm not paranoid. You all think I'm paranoid, don't you !
#EOT
T
Tim Streater
Seebs said:
Yes, quite possibly. What I was observing was his inability to respond
in a way that is germane to the issue.
B
blmblm
[ snip ]
Correct -- he doesn't. (The copying takes place in the *test harness*,
where there appears to be at least a marginally good reason for it.)
[ snip ]
Correct -- he doesn't. (The copying takes place in the *test harness*,
where there appears to be at least a marginally good reason for it.)
[ snip ]
B
blmblm
[ snip ]
How does that solve the problem Seebs mentions? JIT compilation
works pretty well for longish programs, where the time cost of
the compilation is small compared to the overall running time.
For short programs, however, my experience [*] is that precompiled
code can be noticeably faster.
[*] Comparing C code (compiled with gcc and optimization) with Java
(compiled to byte code and run with a runtime system that *supposedly*
does JIT compilation).
Maybe I'm overlooking something obvious here, but I'm not imagining
how this is possible. I mean, either the code is interpreted,
or it's compiled at *some* point, either as part of startup cost
or -- what?
[ snip ]
How does that solve the problem Seebs mentions? JIT compilation
works pretty well for longish programs, where the time cost of
the compilation is small compared to the overall running time.
For short programs, however, my experience [*] is that precompiled
code can be noticeably faster.
[*] Comparing C code (compiled with gcc and optimization) with Java
(compiled to byte code and run with a runtime system that *supposedly*
does JIT compilation).
Maybe I'm overlooking something obvious here, but I'm not imagining
how this is possible. I mean, either the code is interpreted,
or it's compiled at *some* point, either as part of startup cost
or -- what?
[ snip ]
B
blmblm
[ snip ]
And the subsequent discussion does not clear up the misunderstanding.
Ah well. No surprise, I guess.
I find this usage of "dear heart" rather patronizing. Just sayin'.
And I have *NO* idea what my comment has to do with being
objective. I was attempting to avoid giving the impression that
I agreed with your assessment of Seebs's code as ill-structured.
As is true for any number of subjects, I don't really have an
opinion about that one way or another.
(It occurs to me to wonder, by the way, why we're having this
discussion of Seebs's code in threads other than the one in which
he posted the URL for it. But, you know -- whatever.)
[ snip ]
And the subsequent discussion does not clear up the misunderstanding.
Ah well. No surprise, I guess.
I find this usage of "dear heart" rather patronizing. Just sayin'.
And I have *NO* idea what my comment has to do with being
objective. I was attempting to avoid giving the impression that
I agreed with your assessment of Seebs's code as ill-structured.
As is true for any number of subjects, I don't really have an
opinion about that one way or another.
(It occurs to me to wonder, by the way, why we're having this
discussion of Seebs's code in threads other than the one in which
he posted the URL for it. But, you know -- whatever.)
[ snip ]
B
blmblm
"Barneby"?
Your opinion is duly noted and will probably result in fewer responses
from me. However, sometimes he raises a somewhat-on-topic question
to which I would like an answer as well, and other times the temptation
to reply to some particular casting of aspersions is too much.
B
blmblm
[ snip ]
I strongly suspect that you are wrong about that, though I suppose it
might depend on what you mean by "rigorous proof".
If you include purely-mathematical (i.e., non-CS) proofs, then that
would seem to imply that no one else here has much in the way of
mathematical training, which I strongly suspect is not true. To name
the first example (other than myself [*]) who comes to mind, Ben
Bacarisse comes across, to me anyway, as someone who knows some math.
[*] And I am just not going to engage in some sort of competitive
brandishing of educational credentials.
If you mean that the proofs in question have to have something
to do with CS, well, still you would seem to be implying that no
one else here has taken the sort of CS course in which proofs are
required, and that also seems to me to be a pretty strong claim,
one unlikely to be true.
[ snip ]
I strongly suspect that you are wrong about that, though I suppose it
might depend on what you mean by "rigorous proof".
If you include purely-mathematical (i.e., non-CS) proofs, then that
would seem to imply that no one else here has much in the way of
mathematical training, which I strongly suspect is not true. To name
the first example (other than myself [*]) who comes to mind, Ben
Bacarisse comes across, to me anyway, as someone who knows some math.
[*] And I am just not going to engage in some sort of competitive
brandishing of educational credentials.
If you mean that the proofs in question have to have something
to do with CS, well, still you would seem to be implying that no
one else here has taken the sort of CS course in which proofs are
required, and that also seems to me to be a pretty strong claim,
one unlikely to be true.
[ snip ]
S
Seebs
Obviously, it would be impossible to perform a proper clinical diagnosis
over the internet, but:
* Exaggerated sense of own skills and accomplishments -- tends to believe
self substantially better-qualified than other people.
* Proudly claims about connections to people whose contributions are more
widely recognized, such as famous experts or Nobel prize winners.
* Theorizes that all lack of recognition of his superior qualities is due
to conspiracies or jealousy among obviously-inferior people.
* Invents idiosyncratic standards to show that other people are not "special".
* Regards any correction or criticism as a personal attack.
* Attacks people with unusual ferocity, then turns around and treats them
normally the next day; doesn't see anything wrong with this.
Basically, yeah, he sweeps the category. The thing where he'll totally lose
it and spew obscenities at someone, then treat them nicely the next day when
he thinks they have something he wants, is pretty much stereotypical; he
doesn't show any signs of having an awareness of other people except in terms
of how they presently are useful to him or failing to meet his needs.
Similarly, the obsessive trolling seems to be a classic way to try to obtain
narcissistic supply.
The kindest thing to do is probably to completely ignore him. It won't
actually help, but it may slow the decay down some. With one exception
(a guy with a badly-formatted but very insightful page on "malignant
self-love"), most therapists view clinical narcissism as very close to
untreatable, because fundamentally, you can't treat a mental disorder when
the patient cannot conceive of any form of acceptable life not rooted in
that set of behaviors.
.... Relevance to comp.lang.c, you ask? If we want the group to be useful,
we have to recognize that we have a Grade A Troll, and we're going to have
to develop coping strategies. Go look at what happened to soc.religion.quaker
to see why -- the group has essentially no content but an extreme narcissist
and a few people who obsessively attack his posts on the off chance that
anyone doesn't realize he's full of crap.
-s
S
Seebs
True.
Well, consider. His actual claim was stupid. He can't win it. Since he
can't use it to make himself look good, there's no reason for him to respond
to it.
-s
S
Seebs
Where on earth did I suggest anything of the sort?
I'd guess that my algorithm is extremely similar to the one in BSD
libc. Maybe slightly more complicated with the attempt to stash
the next location worth checking.
Yeah.
I think I see the problem: I picked the name "my_strstr()" for my strstr()
implementation, and "test_strstr()" for the function which tests a strstr()
implementation by comparing its results to those of the library one.
I'm guessing that, with his typical complete inattention to detail, Nilges
just saw a function named "something_strstr()" with two arguments and assumed
that it was my implementation. I sometimes forget just how far below the
norm his reading comprehension is when it comes to code. I had not realized
until just now that his behavior actually made sense if you assumed a mistake
that was actually stunningly incomprehensible, rather than merely laughable.
-s
I'd guess that my algorithm is extremely similar to the one in BSD
libc. Maybe slightly more complicated with the attempt to stash
the next location worth checking.
Yeah.
I think I see the problem: I picked the name "my_strstr()" for my strstr()
implementation, and "test_strstr()" for the function which tests a strstr()
implementation by comparing its results to those of the library one.
I'm guessing that, with his typical complete inattention to detail, Nilges
just saw a function named "something_strstr()" with two arguments and assumed
that it was my implementation. I sometimes forget just how far below the
norm his reading comprehension is when it comes to code. I had not realized
until just now that his behavior actually made sense if you assumed a mistake
that was actually stunningly incomprehensible, rather than merely laughable.
-s
S
Seebs
It doesn't.
But in his defense, I don't think he has any comprehension of Unix-like
environments. The big difference (only marginally topical, but bear with
me) is that Unix-like environments have INCREDIBLY low cost of process
spawning compared to Windows. So Windows systems tend to evolve large
programs which run for a long time, Unix systems tend to evolve tiny programs
which run often.
Well, a typical thing would be, say you want to run the same script hundreds
of times, you'd load the script once and then execute it hundreds of times.
(This is how any performance-oriented implementation of Ruby on Rails will
do it, for instance.)
Why, yes.
I'm not particularly sure the design was a good one, but I wasn't asked
for an opinion on that, merely presented with a very specific problem
and asked to make it work.
The scenario is as follows: We maintain a build system in which a number
of components are built once on a single machine and then shared out to
many other machines. Possibly stored in different locations on each machine.
There is a tool which, at build time, compiles in particular paths that
it will use for certain lookups. (It is beyond the scope of our project
to change this tool.) To run a prebuilt version of this tool on another
machine, you need to override that. You can override that by finding out
what directory the program is *actually* installed in, and shoving certain
values into the environment to override the compiled-in values.
So the task at hand was: Build a program which figures out the path it
lives in, then alters particular path values to contain that path (instead
of some other path), and put those path values into the environment.
This is done every time that other tool gets invoked, and that other tool
gets invoked hundreds of thousands of times. We can't do it by modifying
the binary of the other tool, as multiple systems could be sharing a single
binary copy of that other tool, but mounting it in different paths, so the
path computation has to be done "live".
As is usually the case, Nilges has chosen to invent a theory of what would
have been happening had I been an idiot, rather than asking questions to find
out what the circumstances were.
-s
S
Seebs
It seems deeply implausible to me. Both of my parents taught college
mathematics, and I retain to this day an instinctive distrust of empirical
methods as unreliable and shoddy.
-s
K
Keith Thompson
It's at least partly because *you* keep discussing it with him.
C
Colonel Harlan Sanders
My guess (we'll never know the "real" reason) based on some hints in
various posts, is that he was pontificating in a programming
newsgroup about 10 years ago and Heathfield contradicted/corrected or
otherwise bruised his ego. So he set out to "take him down" As he is
now trying to do with you. At some point of this Schildt's name came
up, and Heathfield disparaged him or his book. And so Nilges took the
opposite tack..... He has been sparring with Heathfield ever since.
A year or so ago, Nilges discovers Wikipedia, and Schildt's page
therein. He decides that this is all part of the conspiracy. The
criticisms of Schildt there cite you and Feather, and his attempts to
delete this are reversed, over and over. He decides you're the
instigator of the conspiracy, and attacks you both here.....
But behind all that, he sees you as a symbol of Korporate Amerikka,
which has rejected him, despite his obvious genius, as they're all
jealous of his ability to write poetry and run marathons while coding
in assembly on his laptop. Since you get paid to program, while he can
only do it as a hobby, you remind him of his failure. So you must be
destroyed.
Well, it's a bit more plausible than Nilges' conspiracy theories, at
least.
S
Seebs
This is plausible. He's certainly very consistent in adopting the opposite
of anything someone he dislikes says, without any regard to whether the
resulting position is consistent with his other claims.
This would explain his obsession with claiming that I don't program, and
then his revised obsession with claiming that I do it very badly.
Thanks. That makes a lot more sense. Also, now that I've found out that
he's not just *a* usenet kook, but *the* Usenet kook (hint: He's part
of the events from which we got "Godwin's Law"), up there in the company
of people like Archimedes Plutonium, I'm a lot more ready to accept that
his behavior is comprehensible. Loosely speaking.
-s
T
Tim Streater
Seebs said:
I know, I know. I shouldn't do it really. I picture an 18th century me,
with powdered wig and dressed like a fop, visiting Bedlam and seeing
Spinny behind bars, drooling. Very unkindly I then poke him with a
pointed stick.
S
spinoza1111
And this excuses an incorrect program how?
(Yawn) Terminate and stay resident IS, unlike the heap, an MS-DOS term
from way back, and there's nothing special with it here.
And this excuses an incorrect program how?
Hold that thought
"I was only following orders, so my code sucks". Nuremburg defense is
not allowed here, Hermann.
I created a compiler build system. Fortunately, I was able to rewrite
anything I wanted. These constraints "it is beyond the scope" are mind-
forg'd manacles.
However, I don't think you should rewrite anything until you've
completed first year computer science.
So the task at hand was: Build a program which figures out the path it
lives in, then alters particular path values to contain that path (instead
of some other path), and put those path values into the environment.
This is done every time that other tool gets invoked, and that other tool
gets invoked hundreds of thousands of times. We can't do it by modifying
the binary of the other tool, as multiple systems could be sharing a single
binary copy of that other tool, but mounting it in different paths, so the
path computation has to be done "live".
As is usually the case, Nilges has chosen to invent a theory of what would
have been happening had I been an idiot, rather than asking questions to find
out what the circumstances were.
Incompetent programmers like to be asked questions: it makes them feel
important. Nothing wrong with that. I felt important when important
faculty at Princeton (and Nash) asked me questions.
However, I get a sense here that you're working in The Ass End of a
Botched Situation, created by the nonexistence of people in your work
group competent to change what seems to be the path compiler.
Sure, if the "path compiler" (if that's an accurate description)
generates only %s, then % is a sufficient condition in your idiomatic
situation.
But why post this code?
The situation in which you wrote it is barbaric. Basically, a complex
piece of code (the "path compiler") was written offsite but your
employer, as usual, is too cheap to retain the best programmers to
maintain it, perhaps for the same reason Bell Northern Research
destroyed its compiler group in 1980.
The code if correct in your environment was not in any way exemplary.
Commencing with the election of Reagan, in programming, a deliberate
Kulturkampf was conducted against retaining senior and qualified
programmers and the result? People without computer science education
are hired but not given ability to change things, and develop ugly,
idiomatic approaches. When they see good code they think it's the
worst.