[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: casual contributors

From: Graham Percival
Subject: Re: casual contributors
Date: Fri, 23 Mar 2012 17:02:33 +0000
User-agent: Mutt/1.5.21 (2010-09-15)

On Fri, Mar 23, 2012 at 05:26:11PM +0100, David Kastrup wrote:
> Graham Percival <address@hidden> writes:
> That was about the failure rate in the Theoretical Electrotechnics and
> the Electronic Parts exams.
> Parts of the process
> were ridiculous, and sort of a competition in ugliness among professors
> where each considered their course the most important of the whole
> study, of course expecting the students to work much harder than for any
> other course.

That's what I expected electrical engineering to be.

> > 100% of the fourth-year students would fail, and the department
> > would fire me.  I don't think I can make you understand just how
> > much of a difference there is between UK engineering students in
> > 2012 and German engineering students 20-30 years ago.
> It is not the students.  That's like saying that the average human
> nowadays is bred worse for sports than those from a century ago.  It's
> the conditions.

I totally agree that it's the conditions!  I'm not going to single
out any one of them -- low standards in high schools, ease of
distactions, general cultural disinterest in math+science, quality
of parenting, etc -- but they combine into a mess.

> > [1] telling us "that's impossible due to the Shannon-Nyquist
> > theorem" was bonus marks.  BONUS MARKS.  We then asked those
> > students to solve it for 5000 Hz instead, still giving poles at
> > 1/2 and 3/2 pi.
> Uh, it's impossible not due to Shannon-Nyquist, but because a sampled
> signal is not a sine wave, period.  If your reconstruction filter is a
> lowpass at 10000Hz, you get a sine of 5000Hz out.  If your
> reconstruction filter is a bandpass admitting 10000Hz to 20000Hz, you
> get a sine of 15000Hz out.  That is not just hypothetical: quadrature
> mirror filters (?) work partly by subsampling signals above the Nyquist
> frequency of the subsampling.  Nyquist just limits the _bandwidth_ of
> what you can sample, not the absolute frequencies.

Fair point.  If a student had discussed _that_ in the oral exam, I
would have been overjoyed to give them bonus marks.  :)

> > If anything, it's worse -- universities are
> > increasingly being run as businesses, and it doesn't make business
> > sense to turn away customers, right?
> The definition of a customer for me is a person receiving something
> valuable.  A person merely believing to receive something valuable is a
> sucker instead.

Oh, they *are* receiving something valuable.  They're receiving an
engineering university degree from a 500+-year-old university.
This gives them additional social standing and job opportunities.
They gain a _bit_ of knowledge, but at first guess I'd say that
their entire four-year degree is equal to one year at a German

> Engineering and "hard sciences" are
> going that way today, and the damage is quite more direct.
> We first lost sight of the goals of living.  Now we are losing sight of
> the means of living.

I hate to sound like a grumpy old man -- no wait, who am I
kidding, I love being a curmudgeon :) -- but I agree.  There are
some people who do amazing stuff these days: spaceX, machine
learning at google, the plummeting cost of DNA analysis, etc.  But
there seems to be a growing rift between the top
scientists+engineers and everybody else.

Anyway, this is rather far afield from getting help for casual
contributors to lilypond.  :)

- Graham

reply via email to

[Prev in Thread] Current Thread [Next in Thread]