Lecture 11: What to do when things go wrong? Richard Buckland UNSW

we start by discussing the skater guy from last lecture. intrinsic vs extrinsic motivation. deep and surface learning. class reps revisited briefly.

what can go wrong. what courses of action are open to you to deal with things going wrong in your code? we look at 4 general strategies and discuss their relative strengths and weaknesses.

assert as a way of documenting assumptions and of checking them. contracts. who is to blame. what can you expect other parties to do? what do you need to allow for yourself.

more types: char for representing characters. ascii, relation to 8917, signed vs unsigned. typecasts. implicit type conversions (widening) vs explicit type conversions, typedef. c is lax with types.

getting fit - why is my plan not working. motivation. behaviour change.

Category:

Education

Tags:

• char
• intrinsic
• extrinsic
• motivation
• deep
• surface
• learning
• assert
• errors
• exceptions
• exception
• handling
• contract
• typedef
• richard
• buckland
• UNSW
• computer
• science
• comp1917
• university
• lecture
• fit
• health
• report
• behaviour
• change

[kid29a]

I wonder if more advanced material can be that fun ;) for example: some java database web application programming. I just love the way wich this guy is teaching :)

[SilkSwe]

Could be. :)

Either way I'm just thankful to them for posting the videos at all.

[flowewritharoma]

awesome lecture

[NiGhtMarEs0nWax]

@TeckMod that doesn't make sense. the high order bit is reserved as a flag, when all the bits are off, zero can be + or - , although negative zero isn't used it's function cannot be given to another number. if that bit was set, how could that number be interpreted as a positive or negative value?

i was talking about the number range. its actually just (2^7)-1 for an 8 bit sequence; this gives a maxiumum. ((2^7)*2)-1 or (2^8)-1 [same thing] to find the total amount of true representable values.

[TeckMod]

@NiGhtMarEs0nWax I know this is 2 years old,but let me explain as I understand it.Let's say we have 4 bits; that means that you can go from 0 to 15 (2^[# of bits]). Now notice that we have an even number of numbers we can represent in 4 bits.If we want to have signed numbers we half that quantity(half positive,half neg/),so now we go from 0 to 7 and from -0 to -7. There's a neg zero, so we take it out and use its place to store another negative number.

[NiGhtMarEs0nWax]

also i noticed that he was referring to a 256 bit number as 1, 2, 4, 8, 16, ~ 256. when in reality its 1, 2, 4, 8, 16, ~ 128. 255 is the sum of all teh bits together + the 0 to make 256 unique combinations. when a byte is signed it ranges from -64 to +64, giving it still a possible of 256 combinations. is that correct?

i noticed that ona previous lecture aswell, maybe i am missing something :P

[NiGhtMarEs0nWax]

why do youtube videos always stop and start as "loading" when im only 50% of twh ay through the buffer?

[0121ryanh117]

This video is only 80 MB. Good compression

[khalimseye]

that's a stupid question

[SauronsEye]

Going to Uni and the usual HECS/HELP fee at the end is a burden, that alone will prevents many attending Uni..

So thankyou to the Uni of NSW for these videos.

I'd probably never do this course for real and 2ndly the Uni is so far from my home for a daily commute but I still get to learn something that I will use.

So thank you again.

[theLichKing1]

they have like 55 videos for only the Higher Programming course.. it would take an enormous amount of effort to edit videos and mix things together.. be thankful that they have those available for the world