How many subscribe to scholarly things like 'Communications of the ACM'?
or the 'IEEE Spectrum'?
I then I took it one step further and sat at a table by myself down at the
club, and read it cover to cover. With a beer or two.
And something that hasn't happened in a long time, or well, ever, I got
called out as a sterotype: looking like a geek from San Francisco. Well,
that was what the woman, who also had no conversational partner, used to
strike up a conversation.
Anyway, back to the magazine, who would have thought, that in the CACM, you see such interesting things
like:
"Divination is the practice of occultic ritual as an aid in decision
making", on page 7 no less. Which goes on to say that "... divination
truly allows us to consult the divine, we can view it simply as a form of
randomization, which is recognized as a powerful construct in game theory,
and algorithm design." ... "randomization is a powerful way to deal with
incomplete information" ... "they are simply randomizing in the face of
uncertainty about rain, pests, and more, but this randomization comes with
a belief in the divine source of the decision".... "when the accept/reject
decision pivots on issues such as significance and interestingness, which
can be quite subjective".
Another article: "Assuring Software Quality by Preventing Neglect"....
which is a 'grayhat' problem when compared to blackhats (hackers who
deploy software as a weapon with malicious intent) and whitehats (setting
safeguards against defective products)". With a few interesting
conversation starters being, I think: ".... during maintenance cycles,
they do not correct the old source code comments, seeing such edits as
risky and presumptuous". ... "it is a failure of degree, a failure to pay
enough attention and take enough trouble" ... which leads to a theory of
"ethics of care which displaces the classical agent centered morality of
duty and justice, enduring patient-centered morality as manifest real-time
in relationships".
And, well, something which probably everyone knows, but has been
evolving through a series of articles in the magazine, is that machine
learning algorithms require training time via "generative adversarial
networks", which means you need to introduce some sneaky garbage to improve
the odds of appropriate machine learning and matching black boxes.
And another, in which I can personally associate: "Multitasking Without
Thrashing". "Human context switching is more complicated than computer
context switching. Whereas the computer context switch replaces a fixed
number of bytes in a few CPU registers, the human has to recall what was
'on the mind' at the time of the switch, and, if the human was interrupted
with no opportunity to choose a 'clean break', the human has to
reconstruct lost short term memory" ... "if you have several important
tasks, your brain can get stuck in a decision process that can take quite
a long time to decide -- a situation known as the choice uncertainty
problem". ... "thrashing happens to human multi-taskers when they have too
many incomplete tasks. They fall into mood of 'overwhelm' in which they
experience considerable stress" ..... "context switching is _not_ the
cause of thrashing. ... the cause of thrashing is the failure to give
every active task enough space for its working set" ... "when a task's
working set is in your workspace, protect it from being unloaded as long
as the task is active .. analog: protect working sets of active tasks and
do not steal from other tasks". ... which leads to the rather obvious conclusion: "to exit the thrashing state, you need
to reduce demand or increase your capacity".
Ok one more: "what we have instead is a society moving towards prosthetic
brains that can be monitored at all times by the state, without the
inconvenience of having to have everyone check in each day at the police
station". When you read this, were you thinking "this isn't me! This isn't now!"?
Well, even in present time, this is visible by what we write, where we write, what we read,
where we read, who we interact with, what we interact with, ....