Software Design & Development Marathon [2012] (3 Viewers)

[SpiralFlex]

Pizza or should I say (Jb) and squishy was there

Wow a year has past

 

[brent012]

Pizza or should I say (Jb) and squishy was there

Ahh yeah, now that you mention it i remember Squishy's avatar and Pizzas posts lol. Those were the couple of others i was thinking about =P.

 

[ismeta]

Hey V.U . I hope that's sarcasm, because I can't stand that language at all. I don't like drag & drop. :/

Ohey, I don't know who you are! But then again it's not that hard to find out who I am, is it And yes, certainly sarcasm. I came across VB 6 on the school desktops at school and was shocked (not even .NET). At least my school will have python programmers left, had a bunch do the NCSS Challenge (beginners stream) in year 9 so that was nice.

 

[SpiralFlex]

Ohey, I don't know who you are! But then again it's not that hard to find out who I am, is it And yes, certainly sarcasm. I came across VB 6 on the school desktops at school and was shocked (not even .NET). At least my school will have python programmers left, had a bunch do the NCSS Challenge (beginners stream) in year 9 so that was nice.

Im a stalker that knows everybodys name in this thread. Get some sleep you got Physics tomorrow

 

[ArguablyMitch]

You are a lucky man.

Does anyone here have incompetent SDD teachers? Like your teacher read the textbook 5 years ago and that's all they know? It's such a joke at my school. Teacher doesn't know ANY language and gives us these stupid sheets. I'm fine with learning on our own, which is what we do anyway, but why should we be forced to learn VB on our own as opposed to something else.

Yep, same at my school. My teacher freaked during the last lesson before Trials because he realised that he'd forgotten to teach us a bunch of things that were in the exam...
I will be so happy when this exam is over.

 

[GoldyOrNugget]

Alright guys focus! What were the major syllabus changes that will make a practical difference to us in the exam?

 

[SpiralFlex]

Alright guys focus! What were the major syllabus changes that will make a practical difference to us in the exam?

Nothing really, all I could really see if for Option Topic 2 they added Unicode and Boolean Algebra. The questions should really be no different from last years. They only really renamed sections of the syllabus for clarity.

 

[GoldyOrNugget]

ALGORITHM QUESTION (this one is moderately difficult, but definitely within the scope of the course, and similar to a past HSC question):

You're given a sequential file of records representing item purchases. Each record contains the name of the customer (string) and the number of items they bought in that transaction (integer).

Output the name of the customer with the most purchases.

e.g. if the input is

  {name: "John Doe", count: 13},
  {name: "Susan", count: 5},
  {name: "John Doe", count: 7},
  {name: "Bobert", count: 16}

Here, the output would be "John Doe" as he has bought 20 items in total.

EDIT: assume efficiency doesn't matter

 

[GoldyOrNugget]

In terms of standard algorithms, are there limitations on how we choose to implement them? e.g. there are dozens of ways to implement a binary search. Are we allowed to use any of them?

 

[Cyberbully]

Who am i

 

[tranformer]

Just wondering, do we have to learn the algorithms for the searches and sorts? E.g. binary, insertion etc. Or do we simply have to learn how it works and how it can be applied along with its advantages & disadvantages?

 

[SpiralFlex]

Just wondering, do we have to learn the algorithms for the searches and sorts? E.g. binary, insertion etc. Or do we simply have to learn how it works and how it can be applied along with its advantages & disadvantages?

You need to understand the algorithms for both. And yes, I guess you could learn the advantages and disadvantages. But by that I assume you mean how effective it is in certain situations and assessing its performance and accuracy. You could look at something called Big O Notation, but it's beyond the scope of SDD.

 

[GoldyOrNugget]

In terms of big-O notation, all the SDD sorts are equivalent (i.e. in the worst case, they're all equally shitty). In terms of little-o notation however, selection sort is worse and insertion sort and bubble sort are equivalent (i.e. selection sort is always shitty, whereas the other two have the potential to be not so shitty if the input data is nice).

 

[Hpakins]

heres a question for all the programming paradigm peeps
state and briefly describe what programmers need to consider when choosing certain programming paradigms? (4 marks)

sorry if its not worded nicely, tried my best to input aha

 

[JaySimmo]

In terms of big-O notation, all the SDD sorts are equivalent (i.e. in the worst case, they're all equally shitty). In terms of little-o notation however, selection sort is worse and insertion sort and bubble sort are equivalent (i.e. selection sort is always shitty, whereas the other two have the potential to be not so shitty if the input data is nice).

not necessarily, any linear search runs at O n^2 whereas binary can run at O log(n)

 

[GoldyOrNugget]

All the sorts are O(n²). Linear search is O(n). Binary search is O(log n). The lower-bound for comparison sorts is provably O(n log n).

EDIT: this means that binary search will try at most comparisons. This was tested in last year's HSC multiple choice.

 

[sddmoustache]

In terms of big-O notation, all the SDD sorts are equivalent (i.e. in the worst case, they're all equally shitty). In terms of little-o notation however, selection sort is worse and insertion sort and bubble sort are equivalent (i.e. selection sort is always shitty, whereas the other two have the potential to be not so shitty if the input data is nice).

So like, if we're given a question on suggesting which sort to use, what I choose...since all the sorts are equally "crappy"?

 

[GoldyOrNugget]

- Selection sort requires at most n swaps, so it's effective on systems where memory manipulation is computationally expensive
- Insertion sort runs best on almost-sorted lists
- There's no real reason to use bubble sort, other than it being intuitive

 

[ahdil33]

ALGORITHM QUESTION (this one is moderately difficult, but definitely within the scope of the course, and similar to a past HSC question):

You're given a sequential file of records representing item purchases. Each record contains the name of the customer (string) and the number of items they bought in that transaction (integer).

Output the name of the customer with the most purchases.

e.g. if the input is

  {name: "John Doe", count: 13},
  {name: "Susan", count: 5},
  {name: "John Doe", count: 7},
  {name: "Bobert", count: 16}

Here, the output would be "John Doe" as he has bought 20 items in total.

EDIT: assume efficiency doesn't matter

Just talking logic, I would input all the data into an array. Search for the person's name, if not, create a new record. Then just go through the algorithm and find the maximum, and output the corresponding name. Is there a faster way to do it?

 

[ahdil33]

In terms of standard algorithms, are there limitations on how we choose to implement them? e.g. there are dozens of ways to implement a binary search. Are we allowed to use any of them?

Yes.