S
spinoza1111
It is emerging that on the day of the accident that caused the spill,
BP executives were bullying technicians and engineers to "fix"
problems too fast and too cheaply while having a party on the rig.
I worked in software for Standard Oil in Chicago and Tulsa for a brief
and unhappy period. I was doing OS code for the IBM (mainframe)
Virtual Machine operating system. I was applying secret Standard Oil
changes to the VM source code to secure its seismic data, and writing
new OS code, using assembler. I worked as consultant from a Chicago
firm.
[I would remark that VM/360 prefigured modern "open source" since it
was shipped in source form, and customers could make their own
modifications. VM/360 provided, on dumb terminals, the illusion of
having a personal computer. While I was working with Standard Oil, a
brilliant English programmer, Mike Cowlishaw, developed the Rexx
programming language to write scripts for VM.]
I got the changes done on time and they worked in Chicago, on "my" VM.
VM had been designed to allow systems programmers to fully test new
editions of the OS privately, just like modern Virtual Machine
software.
At the same time, my marriage was heading south because, as I now
realize, my former wife had unrecognized depression. Nonetheless, I
was "forced" to go to Tulsa to demonstrate to the customer that the
software worked, and, more important, that I was a good old boy type
of programmer, which I'm not.
The Tulsa visit was a disaster, although the software worked. The good
old boys thought I was some sort of homo. I didn't like oil patch
culture at all; for one thing, all you could drink was 3.2 beer. The
customer was "offended".
I noticed a culture of "normalized deviance" (a term invented by
anthropologist Dianne Vaughan): a male/macho culture in which thinking
was suspect despite the fact that as early as 1981, the technical
apparatus was scaling up to a point that required thought...including
the realization that my company was wasting time and jet fuel sending
me to Tulsa to look good.
Accounts of the day of the Deepwater Horizon accident are trickling
out. There was, it now appears, an excess of arrogant SOB British
Petroleum executives on the rig, having a party to celebrate their
safety record (!). When problems arose they made idiotic suggestions
which made the problem worse.
But this is well known. What's less commented on is that the
engineers, rather like the engineers on the spot in Chernobyl, did not
push back against the apparatchiks. Instead, they did things offline
and in secret that made the situation worse. Viewing themselves as
engineers using their education and skills to conquer nature, in
another part of their psyche, they'd incorporated the proposition that
"no matter what, managers and owners are our customers, and, the
customer is always right".
I'd protested that I needed to be with my wife and to help her with
our two babies, and that the OS modifications I'd made worked. I was
told that "the customer is always right".
Here, "programmers" like Seebach have incorporated the axioms to the
extent that they can always justify incredibly poor practice by
reference to inchoate ideas of practicality that have nothing to do
with the problem. For example, we were supposed to accept code as
exemplary that fails if a percent is followed by a different character
than p. We were supposed to accept fall through switch when that made
it appear that what was an error, was not, and vice versa.
At the same time, just as on the rig, any one engineer who objected to
the strikingly crude procedures in use could be isolated and mocked,
Schildt is mocked by people who cannot do a proper technical review.
The bespoke-suited executives, underneath a veneer, turn out to be
savages and barbarians with no education or culture above the World
Cup.
Just as Marxism as actually practiced in the Soviet Union resulted in
Chernobyl, Deepwater Horizon is capitalism's Chernobyl. The messes
here, including C standardization, are smaller instances.
To be a "man" on the Deepwater Horizon platform, you had to be...less
than a man, and never speak out, even when ordered to pump seawater
when you knew that that would make it worse. You must never defend
your coworkers. You must find the target du jour and join the mob.
After Exxon Valdez, I was told, in the same hysterical terms used here
in defense of deep C nonsense, that it wasn't practical to double hull
tankers. Likewise, it's not practical to convert to Java.
BP executives were bullying technicians and engineers to "fix"
problems too fast and too cheaply while having a party on the rig.
I worked in software for Standard Oil in Chicago and Tulsa for a brief
and unhappy period. I was doing OS code for the IBM (mainframe)
Virtual Machine operating system. I was applying secret Standard Oil
changes to the VM source code to secure its seismic data, and writing
new OS code, using assembler. I worked as consultant from a Chicago
firm.
[I would remark that VM/360 prefigured modern "open source" since it
was shipped in source form, and customers could make their own
modifications. VM/360 provided, on dumb terminals, the illusion of
having a personal computer. While I was working with Standard Oil, a
brilliant English programmer, Mike Cowlishaw, developed the Rexx
programming language to write scripts for VM.]
I got the changes done on time and they worked in Chicago, on "my" VM.
VM had been designed to allow systems programmers to fully test new
editions of the OS privately, just like modern Virtual Machine
software.
At the same time, my marriage was heading south because, as I now
realize, my former wife had unrecognized depression. Nonetheless, I
was "forced" to go to Tulsa to demonstrate to the customer that the
software worked, and, more important, that I was a good old boy type
of programmer, which I'm not.
The Tulsa visit was a disaster, although the software worked. The good
old boys thought I was some sort of homo. I didn't like oil patch
culture at all; for one thing, all you could drink was 3.2 beer. The
customer was "offended".
I noticed a culture of "normalized deviance" (a term invented by
anthropologist Dianne Vaughan): a male/macho culture in which thinking
was suspect despite the fact that as early as 1981, the technical
apparatus was scaling up to a point that required thought...including
the realization that my company was wasting time and jet fuel sending
me to Tulsa to look good.
Accounts of the day of the Deepwater Horizon accident are trickling
out. There was, it now appears, an excess of arrogant SOB British
Petroleum executives on the rig, having a party to celebrate their
safety record (!). When problems arose they made idiotic suggestions
which made the problem worse.
But this is well known. What's less commented on is that the
engineers, rather like the engineers on the spot in Chernobyl, did not
push back against the apparatchiks. Instead, they did things offline
and in secret that made the situation worse. Viewing themselves as
engineers using their education and skills to conquer nature, in
another part of their psyche, they'd incorporated the proposition that
"no matter what, managers and owners are our customers, and, the
customer is always right".
I'd protested that I needed to be with my wife and to help her with
our two babies, and that the OS modifications I'd made worked. I was
told that "the customer is always right".
Here, "programmers" like Seebach have incorporated the axioms to the
extent that they can always justify incredibly poor practice by
reference to inchoate ideas of practicality that have nothing to do
with the problem. For example, we were supposed to accept code as
exemplary that fails if a percent is followed by a different character
than p. We were supposed to accept fall through switch when that made
it appear that what was an error, was not, and vice versa.
At the same time, just as on the rig, any one engineer who objected to
the strikingly crude procedures in use could be isolated and mocked,
Schildt is mocked by people who cannot do a proper technical review.
The bespoke-suited executives, underneath a veneer, turn out to be
savages and barbarians with no education or culture above the World
Cup.
Just as Marxism as actually practiced in the Soviet Union resulted in
Chernobyl, Deepwater Horizon is capitalism's Chernobyl. The messes
here, including C standardization, are smaller instances.
To be a "man" on the Deepwater Horizon platform, you had to be...less
than a man, and never speak out, even when ordered to pump seawater
when you knew that that would make it worse. You must never defend
your coworkers. You must find the target du jour and join the mob.
After Exxon Valdez, I was told, in the same hysterical terms used here
in defense of deep C nonsense, that it wasn't practical to double hull
tankers. Likewise, it's not practical to convert to Java.