Mike Raiford wrote:

> Whoops, missed that must have duped the word when editing my post :/

I know how it happens: You start writing. You briefly do something else. 
You come back to finish writing, and don't realise how much you've written.

Annoying, isn't it? :-P

Post a reply to this message

Invisible wrote:
> As you probably all know by now, I hold an honours degree in Computer 
> Science. (Upper-second class, no less.)
> 
> As far as I can tell, whoever created this cause was of the opinion that 
> "software" is a construct for storing, processing and retreiving 
> business data. In other words, you have a UI at the front (possibly 
> web-based), a large database at the back, and some complex business 
> logic in the middle.
> 
> The idea that there might exist "software" which doesn't fit this model 
> appears to have not occurred to anybody here. For example:
> 
> - Embedded software.
> - Device drivers.
> - Computer games.
> - Mathematical moddeling.
> - Signal processing.
> - Artificial intelligence.
> 
> None of these really fit the world-view described above. Most of them 
> don't usually involve any kind of database. Device drivers usually don't 
> have a *user* interface of any kind. Computer games might be 3-tier, but 
> it's a hell of a lot different to a stock control system! And DSP and AI 
> are 90% mathematics.
> 
> Given the world-view above, it should come as no surprise that we 
> learned about things like project management, object-oriented analysis 
> and design, UML, CASE tools, relational databases, SQL, HTML, CSS, Java, 
> JavaScript, XSLT, TCP/IP, double-entry book keeping (I'm not making this 
> up), a little bit of management theory (Taylor, Maslow, etc.), and 
> systems testing.
> 

IMHO, for computer science that sounds right.  You learn about 
programming at a higher level.  You are taught the tools to write 
programs typically for the end user.  Business is a very large area that 
uses this type of talent - and programming in that environment typically 
involves data storage.


You are right, the areas that you list also use software, and you have 
been taught a lot of the logic and thinking to be able to program in 
those areas.  But a lot of those areas also require specialized 
instruction - hence degrees in electrical engineering, computer 
engineering, math.

The areas you listed do require programming, but usually from other 
areas of expertise - computer engineering, electrical engineering, math, 
etc...




> There were also two modules taught by Mr Apathy. Mr Apathy was tasked 
> with teaching us about computer hardware, and later about operating 
> systems. In Mr Apathy's opinion, knowing about binary is "pointless" 
> because "the computer will do it all for you anyway". He believed that 
> "20 years ago it might have been necessary to know this stuff, but in 
> the modern world you're really never going to need this information. But 
> it *is* in the exam, so I have to teach it to you." I cannot tell you 
> what an inspirational motivation for learning this was.
> 

To a point he is right - programming in upper level languages does not 
require a knowledge of the very low level happenings.  That's part of 
why they are there - you can program faster without having to worry 
about as much as it is taken care of for you (e.g. garbage collection)

In a sense it would be like teaching you about electron flow in a diode 
and FET so that you can type the word 'print' better.

At level do you stop digressing?

He decided that binary is not really necessary.
He could have presented his case better so that it wasn't so discouraging.

> 
> Actually, I say only one formula... some of the final year optional 
> modules I did also had a little more math in them than that.
> 
> 
> This module did involve some non-trivial math. It was so badly-explained 
> that I couldn't possibly tell you how advanced it was. Much of the 
> module revolved around feed-forward ANNs trained by back-propogation, 
> which is a kind of "grædiɛnt æsɛnt miθæd". For back-propogation to work, 
> the transfer function needs to be non-linear. And there was something 
> about a "næbl" operator.
> 
> IIRC, the part about generic algorithms lasted about an hour.
> 
> It's a pitty really, because it seems like a really interesting subject...
> 
> 

Computer Science is not about math, it's about laying down code, user 
interface, and bringing it together.
The thought is that if you need math, someone will show what they need.




> 
> So anyway, that's what *I* did during my degree. Anybody out there do 
> anything more interesting?
> 

To heck with my course work.
Built and raced 2 solar powered racing cars.
Think 70mph and 'for highway use'.


> 
> 
> My mum keeps telling me I should do a mathematics course. Personally, 
> I'm not sure where the hell I'd get the time or the money from. (It's 
> not exactly cheap.) I guess it *is* kind of amazing that I know about 
> the Laplace transform, given that I have never at any point in my life 
> been "taught" anything beyond simple arithmetic.
> 
> Similarly, I know about how to wire logic gates together to make 
> interesting devices. And I know about Dijkstra's shunting algorithm. 
> (Unfortunately, I'm never sure exactly how to spell the name though!)

It sounds like you think very technically.  What is you interest level 
in electronics.  There's plenty of heavy lifting math there ;-)

Have you ever wanted to build your own 'light blinker' (a device that 
does something that you made it do it)?  This can be done without lots 
of heavy lifting math.



Later... Tom

Post a reply to this message

Invisible wrote:

> Annoying, isn't it? :-P

Yup :)

-- 
~Mike

Post a reply to this message

> So if you wanted to describe the motion of a complex arrangement of rigid 
> components (e.g., a car gearbox), you'd use kinematics?

Yes, for analysis of how fast each bit spins, but that part is actually very 
simple and well known about gearboxes.  Kinematics would be more useful to 
design the windscreen wiper mechanism for instance, where it is assumed that 
the motor can generate enough torque to maintain a constant speed.  What the 
designer must consider is how the constant angular velocity of the electric 
motor is converted to that movement you see the wiper blades making.

> So... the resonant modes of a effectively 1D system (a string, a 
> gas-filled pipe, etc.) would just be harmonics of the main resonant 
> frequency?

Haha I should have known this had something to do with organ pipes!  Yes 
pretty much.

> Heh. So you know how an ideal gas is different from a real gas then? ;-)

From what I recall, ideal gasses are ones that follow the PV=nRT equation 
and a host of other equations.  Which works pretty well for most "normal" 
gasses at "normal" temperature and pressure.  IIRC it doesn't work for steam 
though, which is why we needed hideous fold-out tables and charts.

> If I take a piece of paper and hold it horisontal, it flops under its own 
> weight. But if I fold it down the middle, now it *can* stand up under its 
> own weight. (But only if you hold it the right way.)

You can do that without folding it too, along the edge that you are holding 
it, just push down in the middle with your thumb above the paper with two 
fingers underneath either side.  By introducing that slight curvature in the 
paper you are making it very difficult to bend downwards without stretching 
or ripping the paper, so of course then the paper does not have enough 
weight to do that by itself.  I don't know what category this would come 
under, dynamics or thin bodies or something ;-)

> At the same time, a straight metal rod is very strong, but once bent it 
> becomes drastically weaker,

Actually it usually becomes stronger after the 1st bend because you have 
work hardened it at that point.

> and it seems that nothing will restore it to its original condition.

Because when you try to bend it back, it just wants to bend at a different 
point rather than where the original bend was, because that point is now 
stronger!  Eventually of course if you repeatedly bend it enough fatigue 
will set in and it will break.

> Wait - there's a way to *solve* differential equations?? o_O

Use Laplace transforms, makes things way easier for non-trivial differential 
equations.

> OMG, the first time I watched this on TV, I killed myself laughing. All 
> those hours to build, and it ****ed itself to pieces in seconds! :-D

Hehe yeh I love that program.  Our robot almost did the same, when I 
disconnected the control interface to the PC while the motor that lifted the 
arm was still running.  Of course it was in software that the motor is 
stopped when it hits the switch at the end of the arm's travel, so with no 
more control input the motor continued to wind up and start bending the 
metal arm before I could pull the power :-)  The mechanics guys weren't 
pleased that they had to rebuilt it.

> By the way... how much of the stuff you learnt do you actually *use* now 
> anyway? ;-)

Around 10% probably, but I would imagine every job would have a slightly 
different 10% so I certainly don't regret doing such a wide range of 
subjects within Engineering.

Post a reply to this message

Tom Austin wrote:

>> There were also two modules taught by Mr Apathy. Mr Apathy was tasked 
>> with teaching us about computer hardware, and later about operating 
>> systems. In Mr Apathy's opinion, knowing about binary is "pointless" 
>> because "the computer will do it all for you anyway". He believed that 
>> "20 years ago it might have been necessary to know this stuff, but in 
>> the modern world you're really never going to need this information. 
>> But it *is* in the exam, so I have to teach it to you." I cannot tell 
>> you what an inspirational motivation for learning this was.
> 
> To a point he is right - programming in upper level languages does not 
> require a knowledge of the very low level happenings.  That's part of 
> why they are there - you can program faster without having to worry 
> about as much as it is taken care of for you (e.g. garbage collection)
> 
> In a sense it would be like teaching you about electron flow in a diode 
> and FET so that you can type the word 'print' better.
> 
> At level do you stop digressing?

Well now, it all depends on what you're trying to do.

How many times have you seen a program that behaves strangely given a 
large enough number as input? Many people seem confused that large 
positive numbers come out negative. But this is a simple and obvious 
consequence of 2s complement arithmetic. I don't care what language 
you're programming in, you need to have a basic high-level understanding 
of this stuff.

(As to whether you need to know exactly how many bits are assigned to 
the mantissa of an IEEE double-precision float, or the exact bit 
patterns for demonals... er, no, most people will never actually need to 
know that.)

The stuff about how hardware interrupts work is fairly irrelevant to 
most programming exercises, but if you were doing something slightly 
more specialised it would become highly relevant.

In the end, what it boils down to is that low-level hardware details 
interest me, whereas Mr Apathy did has absolute level best to discourage 
anybody from even *attempting* to put any effort into learning this stuff.

(FWIW, this is the same guy who told us that the rebuilding of Colossus 
was "pointless" because "it will never be the same machine as the 
original".)

> Computer Science is not about math, it's about laying down code, user 
> interface, and bringing it together.
> The thought is that if you need math, someone will show what they need.

Actually, "Information Technology" is about applying computers to solve 
real-world problems. "Computer Science" is the abstract study of 
theoretical models of computation, which results are actually 
computable, computer algorithms, and so forth.

> It sounds like you think very technically.  What is you interest level 
> in electronics.  There's plenty of heavy lifting math there ;-)

My *interest level* is moderately high. My *knowledge level* is very low.

(As in, I know how it's *supposed* to work. It just doesn't work that 
way when *I* do it.)

Post a reply to this message

scott wrote:
>> So if you wanted to describe the motion of a complex arrangement of 
>> rigid components (e.g., a car gearbox), you'd use kinematics?
> 
> Yes, for analysis of how fast each bit spins, but that part is actually 
> very simple and well known about gearboxes.  Kinematics would be more 
> useful to design the windscreen wiper mechanism for instance, where it 
> is assumed that the motor can generate enough torque to maintain a 
> constant speed.  What the designer must consider is how the constant 
> angular velocity of the electric motor is converted to that movement you 
> see the wiper blades making.

So things like cams and eccentrics then?

>> So... the resonant modes of a effectively 1D system (a string, a 
>> gas-filled pipe, etc.) would just be harmonics of the main resonant 
>> frequency?
> 
> Haha I should have known this had something to do with organ pipes!  Yes 
> pretty much.

Well, it was a simple example of something I've already looked at a fair 
bit.

Where I actually heard it mentioned was in relation to "the resonant 
modes of [the air in] this room".

Presumably a 2D surface like a drum head would have quite a lot of 
possible resonant modes?

How is all this related to standing waves?

>> Heh. So you know how an ideal gas is different from a real gas then? ;-)
> 
>  From what I recall, ideal gasses are ones that follow the PV=nRT 
> equation and a host of other equations.  Which works pretty well for 
> most "normal" gasses at "normal" temperature and pressure.  IIRC it 
> doesn't work for steam though, which is why we needed hideous fold-out 
> tables and charts.

...because steam can condense? (Or at least is likely to under typical 
conditions, whereas air isn't.)

>> If I take a piece of paper and hold it horisontal, it flops under its 
>> own weight. But if I fold it down the middle, now it *can* stand up 
>> under its own weight. (But only if you hold it the right way.)
> 
> You can do that without folding it too, along the edge that you are 
> holding it, just push down in the middle with your thumb above the paper 
> with two fingers underneath either side.  By introducing that slight 
> curvature in the paper you are making it very difficult to bend 
> downwards without stretching or ripping the paper, so of course then the 
> paper does not have enough weight to do that by itself.  I don't know 
> what category this would come under, dynamics or thin bodies or 
> something ;-)

Ah yes - just curving the paper makes it behave quite differently. (See 
corrigated sheet metal.) But why, I wonder?

Similarly, a hollow tube responds differently to a solid rod. And again, 
a tear tends to propogate along a sheet, but if you put a hole in the 
sheet, it actually stops the tear. WTF?

>> At the same time, a straight metal rod is very strong, but once bent 
>> it becomes drastically weaker,
> 
> Actually it usually becomes stronger after the 1st bend because you have 
> work hardened it at that point.
> 
>> and it seems that nothing will restore it to its original condition.
> 
> Because when you try to bend it back, it just wants to bend at a 
> different point rather than where the original bend was, because that 
> point is now stronger!  Eventually of course if you repeatedly bend it 
> enough fatigue will set in and it will break.

Really? OK, well that's even weirder!

Presumably there's some molecular-level *reason* for all of this?

>> Wait - there's a way to *solve* differential equations?? o_O
> 
> Use Laplace transforms, makes things way easier for non-trivial 
> differential equations.

...because the Laplace transform turns differential equations into 
algebraic equations?

Presumably there are still equations which can't be "solved" in this way 
though? (It seems quite easy to come up with a set of equations which 
yield absurdly complicated behaviour...)

>> OMG, the first time I watched this on TV, I killed myself laughing. 
>> All those hours to build, and it ****ed itself to pieces in seconds! :-D
> 
> Hehe yeh I love that program.

Did you see Scrappy Races, where the same guy tried to "tune" a 6 L 
diesel engine, and the govener fell off? (I don't know what a govener 
is, but it sounds important...)

> Our robot almost did the same, when I 
> disconnected the control interface to the PC while the motor that lifted 
> the arm was still running.  Of course it was in software that the motor 
> is stopped when it hits the switch at the end of the arm's travel, so 
> with no more control input the motor continued to wind up and start 
> bending the metal arm before I could pull the power :-)  The mechanics 
> guys weren't pleased that they had to rebuilt it.

Niiiice! :-D

>> By the way... how much of the stuff you learnt do you actually *use* 
>> now anyway? ;-)
> 
> Around 10% probably, but I would imagine every job would have a slightly 
> different 10% so I certainly don't regret doing such a wide range of 
> subjects within Engineering.

Mmm, true...

Post a reply to this message

scott wrote:

> The best part was the robot design project.  We were put into teams of 6 
> and told to make a robot that drove around a track marked by white lines 
> on a black board and picked up containers.  The containers would either 
> be empty or full, and we had to move the containers to the appropriate 
> bin.  In most teams 2 people actually made the robot mechanics, 2 did 
> the electronics (motor drive control, interface to PC, etc) and 2 did 
> the software.  Was really good fun, especially as the quicker your robot 
> completed the task the more points you got.

We did a "software development project". This is where a team of (IIRC) 
6 of us got together to plan, design, build and test a "large" piece of 
software, and exhibit it at a show.

Of course, of the 6 people in the group, I was the only one who knew how 
to program. So this "project" consisted of the following:

- We all sat round a table, and we're like "OK, so what are we gonna make?"
- After a while, we decide what to build, and generate a few design 
ideas. "Do you think you can do that, Andrew?" "Yeah, probably." "OK 
mate, sounds like a plan."
- I sit in a corner by myself, building and testing the program. Every 
now and then another team member will casually ask me how it's going. I 
tell them it's going OK. They smile and nodd, and wander off.

Fortunately, for this particular module, most of the marks are for 
producing all the pretty GANT charts and all the project-management 
garbage, and not for actually producing a working product.

[The urge to insert something about software consultants here is 
overwhelming!]

Which is just as well - my "product" was a trivial job searching engine 
that would have taken about 3 seconds if I had used a real database 
engine rather than manually full-scanning flat files held entirely in 
RAM. >_<

Can you spell "not scalable", "lost update problem" and "security"?

Post a reply to this message

> So things like cams and eccentrics then?

Yep, or just plain linkages of rods and stuff, some of them can get quite 
complex.  On my old Peugeot 206 it had some pretty funky wiper blade 
mechanism to that it would cover much more of the window than a simple arc.

> How is all this related to standing waves?

A standing wave is just a wave that stays in position, ie it's frequency and 
spatial phase remain constant.  If you see a resonant mode by itself, it 
will look like a standing wave and you can easily identify the stationary 
points that remain at rest.

> Presumably a 2D surface like a drum head would have quite a lot of 
> possible resonant modes?

Yep, see these photos that show the stationary points, try to imagine how 
the non-stationary parts would bend up and down.

http://www.meta-synthesis.com/webbook/34_qn/2d_waves.jpg

> Ah yes - just curving the paper makes it behave quite differently. (See 
> corrigated sheet metal.) But why, I wonder?

To curve a flat sheet of paper all you need to do physically is to stretch 
the top half of the thickness by a tiny amount and compress the bottom half 
by an equally tiny amount.  Once you start curving it in other directions as 
well the amount of stretching and squashing you need to perform becomes 
orders of magnitudes higher.  You can see this just by looking at the 
geometry of a curved piece of paper and thinking about how it must deform in 
order to curve it in the other direction.

> Similarly, a hollow tube responds differently to a solid rod.

For a hollow tube and a rod of the same weight, the tube will always be 
stronger and stiffer because it will have a higher 2nd moment of area. 
Actually the I-beam is pretty much the best useful shape you can have, which 
is why they are so common (tubes are not so easy to join together and funny 
angles).

> And again, a tear tends to propogate along a sheet, but if you put a hole 
> in the sheet, it actually stops the tear. WTF?

It's to do with the radius at the tip of the tear (very small) compared to 
the radius of the hole (much bigger).  It's why the windows in aeroplanes 
are round and not square (they had some square ones to start with but they 
got cracks at the corners once the pressure difference went up).

> Presumably there's some molecular-level *reason* for all of this?

Yeh, something to do with dislocations in the crystal lattice from what I 
remember.

> ...because the Laplace transform turns differential equations into 
> algebraic equations?

Yup.

> Did you see Scrappy Races, where the same guy tried to "tune" a 6 L diesel 
> engine, and the govener fell off? (I don't know what a govener is, but it 
> sounds important...)

No I didn't see that one, the governer stops the engine going too fast if 
there is no load.  Diesel engines are quite capable of self-destruction if 
you try to run them flat out with no load.

Post a reply to this message

scott wrote:
>> How is all this related to standing waves?
> 
> A standing wave is just a wave that stays in position, ie it's frequency 
> and spatial phase remain constant.  If you see a resonant mode by 
> itself, it will look like a standing wave and you can easily identify 
> the stationary points that remain at rest.

I see... So the resonant modes of something are all possible standing 
waves that can be created in that space?

>> Presumably a 2D surface like a drum head would have quite a lot of 
>> possible resonant modes?
> 
> Yep, see these photos that show the stationary points, try to imagine 
> how the non-stationary parts would bend up and down.
> 
> http://www.meta-synthesis.com/webbook/34_qn/2d_waves.jpg

Ooo... so there are both concentric and radial modes, and combinations 
thereof?

Presumably for a square surface it would be different though.

>> Ah yes - just curving the paper makes it behave quite differently. 
>> (See corrigated sheet metal.) But why, I wonder?
> 
> To curve a flat sheet of paper all you need to do physically is to 
> stretch the top half of the thickness by a tiny amount and compress the 
> bottom half by an equally tiny amount.  Once you start curving it in 
> other directions as well the amount of stretching and squashing you need 
> to perform becomes orders of magnitudes higher.  You can see this just 
> by looking at the geometry of a curved piece of paper and thinking about 
> how it must deform in order to curve it in the other direction.

So it happens due to the infintesimal but finite thickness of the sheet? 
Interesting....

>> Similarly, a hollow tube responds differently to a solid rod.
> 
> For a hollow tube and a rod of the same weight, the tube will always be 
> stronger and stiffer because it will have a higher 2nd moment of area. 
> Actually the I-beam is pretty much the best useful shape you can have, 
> which is why they are so common (tubes are not so easy to join together 
> and funny angles).

"2nd moment of area"? ._.

Well anyway, it strikes me that a hollow tube of the same weight would 
be very much larger, and hence obviously stronger.

The reason I mention this is that apparently trees that have rotted 
hollow tend to survive storms better. Thus, the fungi that rot trees 
actually *prolong* their life, not shorten it. (It's amazing what you 
can find out from wildlife books!)

>> And again, a tear tends to propogate along a sheet, but if you put a 
>> hole in the sheet, it actually stops the tear. WTF?
> 
> It's to do with the radius at the tip of the tear (very small) compared 
> to the radius of the hole (much bigger).  It's why the windows in 
> aeroplanes are round and not square (they had some square ones to start 
> with but they got cracks at the corners once the pressure difference 
> went up).

How interesting...

>> Presumably there's some molecular-level *reason* for all of this?
> 
> Yeh, something to do with dislocations in the crystal lattice from what 
> I remember.

Ouch. "Doctor, I've got a dislocated crystal in my lattice. It hurts 
when I bend over..."

>> ...because the Laplace transform turns differential equations into 
>> algebraic equations?
> 
> Yup.

I see...

>> Did you see Scrappy Races, where the same guy tried to "tune" a 6 L 
>> diesel engine, and the govener fell off? (I don't know what a govener 
>> is, but it sounds important...)
> 
> No I didn't see that one, the governer stops the engine going too fast 
> if there is no load.  Diesel engines are quite capable of 
> self-destruction if you try to run them flat out with no load.

Guy was tinkering with this fuel-injected diesel engine. (The engine is 
about the size of a small bath tub.) Obviously the camera crew weren't 
paying too much attention. (This was during build time.) Suddenly we got 
a shot of the machine in the distance with ****-off black Clouds Of Doom 
spewing out of the exhaust pipes, and the team of hill-billies running 
for their lives in every direction like startled rats.

Apparently the governer just fell off, and the engine is now running at 
full tilt in neutral with the fuel injection system at non-default settings.

A few moments later the team run back to the machine. One of them is 
trying to stuff his shirt into the air intake. Another is trying to 
physically rip the leads off the battery. (Uh, how does that help?) 
Another is trying to squeeze the fuel lines with his hands. I don't know 
quite what happened, but after a few moments the machine fell silent.

Cut to a scene 20 minutes later, and they've taken the head block off 
the engine, and the team captin is holding up bits of pistol rods, and 
other items which look like they really ought to be *attached* to 
something, but clearly aren't...

Post a reply to this message

> I see... So the resonant modes of something are all possible standing 
> waves that can be created in that space?

Yes I guess so.

> "2nd moment of area"? ._.

Determines the geometric stiffness of a cross-section.  Basically the more 
"material" further away from the axis of bending the stiffer it will be.  So 
if you have an area budget of 10cm^2, better to put it all as far as 
possible away from the centre (ie a circle), or if it only needs to bend in 
one direction, then two parallel plates (usually connected though, so that 
will be an I-beam then).

> Apparently the governer just fell off, and the engine is now running at 
> full tilt in neutral with the fuel injection system at non-default 
> settings.

I had a similar thing happen to my radio controlled car, the throttle fell 
off, literally the bit that regulates the air-flow into the engine fell out 
of the air intake.  Of course then the car is going full throttle in a 
street full of parked cars and all I can control is the steering!  Finally I 
think it got wedged underneath a real car and not too much damage.  I found 
the throttle too and put it back in, with some loctite this time!

> Cut to a scene 20 minutes later, and they've taken the head block off the 
> engine, and the team captin is holding up bits of pistol rods, and other 
> items which look like they really ought to be *attached* to something, but 
> clearly aren't...

Try running a lawn mower with no oil, part way through cutting the grass it 
stopped, not like the running out fuel stopped, but instant stop from 200rpm 
or whatever to silence.  Took the head off, a big bit of wood and a big 
hammer freed up the piston from the side of the cylinder, then a good dose 
of oil and it was working again!

Post a reply to this message