Re: Programmer's unpaid overtime.

From: Edward G. Nilges (spinoza1111_at_yahoo.com)
Date: 11/29/03 

Date: 28 Nov 2003 19:02:08 -0800

Richard Heathfield <invalid@address.co.uk.invalid> wrote in message news:<3fc5b0c6@news2.power.net.uk>...
> [Mr Nilges' article weighed in at 316 lines. Let's see how much of it was
> relevant to this newsgroup, by removing all the irrelevant stuff. See end
> of article for results!]
>
> Edward G. Nilges wrote:
>
> > Programmer Dude <Chris@Sonnack.com> wrote in message
> > news:<3FC51342.9FCE5E9C@Sonnack.com>...
> >> "Edward G. Nilges" wrote:
> >>
> >> >> I believe the rule is quite simple: time invested grants
> >> >> privilege to depart from the norms *ON* *OCCASION*.
> >> >>
> >> >> As Richard pointed out to you recently, it has to do with
> >> >> percentage.
> >> >
> >> > During the 2002 discussion of the US Data Quality Act, an
> >> > on-topic issue,...
> >>
> >> Oh, Ed, get over it.
> >>
> >> The DQA *itself* wasn't on-topic, because it applied to only
> >> one nation. And after several attempts to explore the
> >
> > Discussions of C apply only to one language.
>
> Very true, and so C discussion takes place in comp.lang.c, not in a general
> programming group. What we see discussed here is /programming/, in which
> sometimes C is used to make a concrete illustration possible. Sometimes
> people use C++, or BASIC, or assembly language, or Ada, or Lithp, or Perl,
> to write their illustrations. I don't hear you complaining about those.

This is a lie. You've discussed the C language in this ng as the C
language and you've promoted your outdated use of it.
>
> <snip of irrelevant stuff>
>
> >> > Richard contributed NOTHING except claims that the issue was
> >> > off-topic. His claims were off-topic because they were
> >> > meta-discussion.
> >>
> >> Incorrect on two counts. First, Richard *did* contribute;
> >> second, topicality is--by definition--on topic on amUSENET.
> >
> > Not to the extremes and in the bad faith it was done, it isn't.
>
> The bad faith was on your part, for introducing an off-topic discussion. I
> seem to recall that several of us tried to turn the discussion in topical
> directions, and even succeeded at one point - that tokenisation stuff where
> you ended up with egg all over your face, remember?

In your dreams, buddy boy.

Richard is referring to some code I found useful in Visual Basic. In
1993 I discovered that that then new language lacked anything like C's
strspn and strcspn, a tool not for finding a unique character or
string, but for finding one of a SET of alternative characters.

As I had done in assembler language as long ago as 1974, I developed a
VB tool which I still use to provide the missing functionality.

Based on this, which I called verify following PL/I and Rexx rather
than the more gnomic C names, I developed a very, very simple parser
for a very, very common task, and this is isolating blank and other
white space character delimited strings without worrying the user
about one-n blanks when n is small.

If memory serves, and because I coded the examples special for the
discussion, I made a trivial but serious error, and openly corrected
the error, whether before or after noticing it, or having it pointed
out to me.

The error did not bother me because in my actual production library,
now a stateless object in .Net, it had long since been fixed.

But because incompetent programmers delight in pointing out other
programmer's errors, several of Heathfield's buttboys capered and
gibbered about my "incompetence" for weeks afterward.

Didn't bother me simply because I've single handedly written several
compilers of 10000 lines or more which have been used by others
successfully. Indeed, it is to me the mark of the competent programmer
to list, to identify, to narrate and to celebrate each one of his OWN
errors whenever they occur, and to be in general silent about other
people's errors, unless he or she is engaged in a common project.
This is in part because you learn from your mistakes.

The error in question, if memory serves, was forgetting to allow for
blanks before the first token. The lesson is to either use a regular
expression (despite the deficiencies of regular expressions) or put
one in the comments to document the manual implementation of a regular
expression. The regular expression is (if blanks only constitute white
space) ([ ]*[^ ]+)*[ ]*.

There are many other valid solutions but this solution seems to have
the least psychological or mathematical ambiguity, which is important
because different regular expression processors give different results
for hard cases.

Richard is very emphatic about my "verbosity". This is because in
service to corporate power, programmers, as part of their
deprofessionalization, are no longer expected to be able to chat
amusingly in human language about code. Increasingly, data systems are
preferred that can NOT be fully and transparently explained to labor
or the public, and in recent years programming has become a silent
brotherhood (bruder schweigen) as a result.

>
> > Richard did not once ask for a justification of the topicality. He
> > claimed without much argument that the DQA was offtopic.
>
> The charter is very clear.
>
> > He refused to
> > "listen" to the rebut because my rebut did not alter his claim.
>
> That's a very interesting definition of "listen".
>
> <snip>
>
> >> > Also, as I have noted, Richard's posts concerning C are offtopic
> >> > in the sense of being more appropriate to comp.lang.c.
> >>
> >> We, in comp.programming, are fairly casual about topic so long as
> >> it has to do with the art and craft of Actual Compute Programming.
> >> Regular posters here are also allowed the occasional diversion.
> >>
> > What about the profession?
>
> If you feel the need to discuss the profession and cannot find an
> appropriate newsgroup in which to hold such discussions, you can take
> advantage of a well-established procedure for creating a new newsgroup for
> that purpose.

I asked why, given corporate lack of committment or resources to
software quality, corporations could in good faith demand that
governments (whether in the US or your tinpot 51st state with its
ridiculous monarchy) adhere to standards corporations do not enforce.

This remains a comp.programming issue unless programmers are now a
clerical class expected only to "code" rules understood to be
non-self-contradictory, perfectly clear, and not in violation of the
law.

Putting aside any political considerations, even as a technical
choice, this is a rejection of any insight into the unworkability or
self-contradiction of the rules derived from engagement with data
systems.

It is the social choice which creates the software crisis itself, and
it means that different social choices would result in clear and
maintainable code.

If this is not of interest to programmers, pal, then what is?
Deliberately using an outdated language? Buying a house?

>
> <snip>
>
> [Down to 84 lines, including my reply and sig block. Mr Nilges, you are
> being overly verbose. Please stick to the subject. Thank you.]