27 Jan
2012
27 Jan
'12
9:21 p.m.
This is something I know nothign about either,
and it would certainly
help _me_ if somebody could explain what lambda calculus and
'closures' are useful for.
The lambda calculus is useful in the same way that any theoretical
discipline is useful. It does not itself solve problems; it informs
It's difficult to give a practical problem closures are useful for in
much the same way it's difficult to give a practical problem resistors
Actually, I totally disagee with that. I can give a doen problems where
resisotrs are useful. The potential divider being high up the list...
Let me quote The Art of Electronics (2nd edition is the one I have to hand).
'This is a long and interesting story. It is the heart of electronics.
Crudely speaking the name of the game is to make and use gadgets that
have interesitng and useful I-versus-V characteristics. Resisotrs (I
simply proportial to V), capacitors (I proportional to the rate of change
of V), diodes (I flows only in one direction).... are examples.'
'It is an interesting fact that the current throug ha metallic conductor
(or other partically conducting material) is proportional to the voltage
across it. ... This is by no means a universal law for all objects. For
instance the current through a neon bulb is a highly non-linear function
of the applied voltage (it is zero up to a critical voltage, at which
point it rises dramatically). The same goes for a variety of interesting
sepcial devices -- diodes, transistors, light bulbs, etc.'
'Resisotrs are truely ubiquitous. There are almost as many times as there
are applications. Resistors are usedi in amplifiers as loads for active
devices, in bias networks, and as feedback elements. In combination with
capacitors they establish time constnts nad act as filters. They are used
to set operating currewnts and signal levbels. Resistors are used in
power circuits to reduce voltages by dissipating power, to measure
currents, and to discharge capacitors after power is remved. They are
used in precision circuuts to establish currents, to provide accurate
votlage ratios and to set precise gain values. In logic circuits they act
as bus and line terminators and as "pull-up" and "pull-down" rsisotrs.
In
high-voltage circutis they are used to measure voltages and to equalize
leakage currents among diodes or capacitors connected in series. In radio
frequency circuits they are even used as coil forms for inductors'
'Roughtly speaking resisotrs are used to convert a votlage to a current
and vice vera. This may sound awfully trite, but you will soon see what
we mean.'
Now I don't for an instant claim that's all there is to resistors. But at
least it explains why you'd nee to know about them. Can the same be done
for lambda calculaus.
There is anotehr difference too. I guess it's _possible_ to have an
electronci device withotu a signle resisotr, but such things are not
common. It's (almost?) impossible to understnad the operation of any
circuit (if you don't treat the ICs as just black boxes0 without
understnadign the concept of reisstance (or soemthing equivalent to it,
like conductanmce).
That's why every book on electroncis covers resistors early on.
Now, I don't claim to be much of a programmer, and I certainly don't
claim to be a computer scientist. But Iv'e written the odd progrma, I've
read source listings of quite a few things, I've read quite a few books
on programming (OK, particulalry low-level stuff and OSes). But I've not
had to understnad lambda calculus or closures 9yet). OK, you'll tell me
that they're not applciaable to device drivers, or data acquisiton
systems, or... , whichI fully accpet. But what _are_ they applicable to.
-tony