Search Results

Search found 2 results on 1 pages for 'sobolan'.

Page 1/1 | 1 

  • Teacher demands excessive/unjustified use of Design Patterns

    - by SoboLAN
    I study computer science and I have a class called "Programming Techniques". Its purpose is to teach (us) good object oriented design principles. During the semester we have homeworks, programs that we must write to demonstrate what we've learned. The lab assistant demands for each of these homeworks that specific design patterns should be used. For example, the current homework is an application used for processing customer orders. We are demanded to use either "Factory Method" or "Abstract Factory" design patterns for this. It gets even worse: at the end of the semester we must write a program (something more complex) that must use at least one creational pattern, at least one structural pattern and at least one behavioural pattern. Is it normal to demand this ? I mean, forcing us to design our programs in such a way that a specific design pattern makes sense is just beyond what I consider ok. If I'm a car mechanic and have a huge tool box, then I will use a certain tool from that box if and when the situation demands it. Not more, not less. If my design of the application doesn't demand at all the use of "Abstract Factory" (for example), then why should I implement it ? I'm not sure yet if the senior lecturer agrees with what the lab assistant is demanding, but I want to talk to him about it and I need solid arguments to do so. How should I approach this problem with him ? PS: I'm sure there must be a better way to teach us these things. Maybe making us each week read about 3 design patterns and the next week giving us a test with small but specific programming or architectural situations/problems. The goal in that test would be to identify what design patterns would make sense and how they could be implemented. This way, he can see if we understand them. EDIT: These homeworks are not just 100-line programs, they have quite a lot of requirements and are fairly complicated. This is the reason we have about 2 - 3 weeks of deadline for each of them. I agree that practicing this is the best way to learn. But shouldn't smaller programs/applications be used for this ? Something just for demonstrating purposes. Not big programs with lots of requirements/classes/etc.

    Read the article

  • Why do you have to manually type variable names while debugging?

    - by SoboLAN
    I've seen this in a lot of IDEs (even in the most popular and heavily used ones, like Visual Studio): if you want to watch a variable's value, you have to manually type its name in the Watches section of the debugger. Why can't there just be a list of all of them with checkboxes next to them ? The developer can then just check the box next to the one he wants to watch and that's it. Variables with identical names can probably be numbered in some way (for example a, b, x(1), x(2), c, etc. I've seen some exceptions to this (NetBeans or BlueJ), but there are exceptions to everything, right ? Maybe it's a stupid question, maybe not, but I've always wondered why this is so.

    Read the article

1