My point was nothing to do with adoption - in fact I was pointing out that people used programming languages even when then did not intend to use or write a compiler for them.

Lack of a stack doesn't imply a machine is not a von Neumann architecture. For example, if I recall correctly, the Cray Y-MP series had no stack for procedure activation records and yet still had a shared instruction/data storage.

Your estimate is a bit too high. Plankalkül was developed between 1943 and 1945, and published in a paper in 1948. FORTRAN was implemented in around 1955. I ripped these dates from Wikipedia's History of programming languages article.

For that matter, Turing's famous and influential 1936 paper On Computable Numbers paper introduces an abbreviation system ("Inst{...}") for building Turing Machine configurations (on page 260) which might loosely be described as a higher-level language.
 

The city of Çatalhöyük existed from about 7500BCE with a large population (thousands). This is apparently several thousand years older than the wheel.

Are you actually looking for a new job? Doing what?

Drat, too much haste.

s/do different people/two different people/
s/The latter cares most about the privacy angle/The former cares most about the privacy angle/

If by "scares" you mean manufactured, misleading hyperbole, you're wrong. There are tens of thousands of adverse drug interactions annually in the UK (and more in the USA). Many of these are avoidable (they're not just drug-drug interactions, adverse drug-condition or drug-{age,procedure} interactions occur too) and key to avoiding this is delivering timely, accurate information to your healthcare providers.

Keeping yourself off the relevant clinical databases is a choice and a compromise of risks; on the one hand the risk that your data will be leaked and on the other hand that your choice to equip your clinicians with less information will cause you to get less effective treatment in the future.

In some senses this is a balancing of benefits to do different people; first, your healthy, vigorous, young self. Second, your elderly, sick, incapacitated self. The latter cares most about the privacy angle but I'm pretty sure the latter cares most about the quality of care. But it would too late for the elderly you to benefit their treatment by reversing the decision made by their younger self.

~/source/GNU/coreutils/coreutils$ git log | grep -c '^Author: Jim Meyering'
23652
~/source/GNU/coreutils/coreutils$ git log | egrep '^(Date:|Author: Jim Meyering)' | tail -n 2
Author: Jim Meyering
Date: Sat Oct 31 20:42:48 1992 +0000

That's a false dichotomy.

Google's infrastructure is too large to hire system administrators that do anything manually. This means even the "system administrators" are software developers.

The alternative to doing the research to identify which job opening appeals to you most is that they interview you for a job _they_ think might suit you, which isn't a good match for your skills, and so you don't get the job. Which one is it you prefer again?

Indeed, Google already published a paper describing their approach:
https://www.google.com/url?sa=t&source=web&cd=1&ved=0CA8QFjAA&url=http%3A%2F%2Fcode.google.com%2Fspeed%2Farticles%2Ftcp_initcwnd_paper.pdf&ei=xUciTKhpmOiUB_HL8bAP&usg=AFQjCNET-zahhIxtRlXe28xn_8QSXXLx6A&sig2=0mlyaOW1btaj7hUjlL1Opw