Results 1 to 6 of 6

Thread: oo vs traditional methods

  1. #1
    Senior Member
    Join Date
    May 2005
    Posts
    434
    Thanks
    0
    Thanked
    0 times in 0 posts

    oo vs traditional methods

    hi can anyone help me on any information concerning the comparision between object oriented methods and structured methods i.e. SSADM, traditional waterfall method etc?

    Thoughts?

  2. #2
    ?!
    Join Date
    Sep 2004
    Posts
    1,045
    Thanks
    2
    Thanked
    6 times in 5 posts
    If you know and understand OO concepts enough, you should only need to look up structured/procedured programming to compare the two.

    http://en.wikipedia.org/wiki/Structured_programming

  3. #3
    Flower Child stytagm's Avatar
    Join Date
    Aug 2004
    Location
    London
    Posts
    754
    Thanks
    47
    Thanked
    23 times in 18 posts
    Are you talking about actual "software engineering methods" ie, a way of getting from an idea to some software (Like waterfall or iterative). The whole show from conception, through requriements, design, prototyping, testing, the lot...

    or

    Do you mean use OO design and programming techniques / ideaologies? Which only really apply to design and development?

    "Methods" like waterfall etc are sort of independent of whether you use a procedural or OO programming language. So you can use the waterfall / iterative / whatever method with OO or procedural. Some of the more recent methods evolved after OO, so are slightly more Data orientated, but a good method will give you the freedom to implement your final solution using whichever languages and techniques you see fit.

    I.E. Your requirements and specifications should outline the problem, without stepping on the toes of the design. For example by using OO terminology to describe a problem (classes objects inheritance), when you might actually need a small fast procedural solution (or vice versa, taling about functions / actions when you might need an OO solution).

    It's a classic university essay topic so the net should be full of good arguments on the subject.

    PS Mods - Should this thread be moved out of General discussion to the software development forums?
    They told me I was gullible ... and I believed them.

  4. #4
    Senior Member
    Join Date
    May 2005
    Posts
    434
    Thanks
    0
    Thanked
    0 times in 0 posts
    the methodologies i am talking about i.e software engineering methods

  5. #5
    Seething Cauldron of Hatred TheAnimus's Avatar
    Join Date
    Aug 2005
    Posts
    17,168
    Thanks
    803
    Thanked
    2,152 times in 1,408 posts
    dude, seriously read your lecture notes.

    Software models like this often have various interpretations applied, i found this out the hardware because i thought i knew one of the models because i'd used it before, turns out my lecturer had a differen't idea, and wasn't impressed.... even thou wikipedia and most other people would of agreed with me!
    throw new ArgumentException (String, String, Exception)

  6. #6
    Flower Child stytagm's Avatar
    Join Date
    Aug 2004
    Location
    London
    Posts
    754
    Thanks
    47
    Thanked
    23 times in 18 posts
    Nutshell thoughts (off the top of my head).

    Methods are basically there to manage the complexity of projects, other wise we'd all just start hacking, and still finish on time/budget.

    A procedural method conentrates on capturing, modelling and programming functions / precedures that do stuff, to data. Example, JSP (Jackson structured programming, not java server pages, Jackson was there first ), JSP use functional decomposition to decompose a problem in sub-problems based on what you want it to do (functions). In theory eventually you get to functions that are simple enough to understand, and design code (functions again) that will solve your problem. Also important might be sequence and state. Imagine a "number cruncher" program, we're going to worry about modelling the "Crunching".

    An OO method concentrates on modelling the objects, the things. Not just the data that has stuff done to it, but also the things that are doing the stuff to the data. So you would decompose big objects like "House" into smaller objects like "roof", "door" and "toilet", until you get to objects simple enough that you can model them in code as Classes, with data and methods, and data objects. So we're going to model the "number cruncher" and the "numbers", although the "number cruncher" can have both internal "numbers" and a "crunch" method. I can't think of an OO method by name of the top of my head, is the Rational unified process OO structly??

    Lastly you have methods that are not tied to procedural or OO, possibly because they are looser. For example I like Jackson's problem frames for requirements and specs, which use Heirarchical decomposition to describe the problem not the solution. Then when you get to design, you can still decide whether to concentrate on functions in C, or objects in Java, or a bit of both and code in any language known to man.

    Verbal dischage ended*

    * Nominal value of this post £0.02
    Edit* I added a bit more - apologies!
    Last edited by stytagm; 25-05-2006 at 10:43 PM.
    They told me I was gullible ... and I believed them.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Altenative broadband methods
    By Agent in forum Networking and Broadband
    Replies: 4
    Last Post: 02-02-2006, 04:32 PM
  2. [SERVER] Backup Methods
    By Vini in forum Software
    Replies: 14
    Last Post: 02-11-2005, 09:37 PM
  3. Contact methods
    By JK Ferret in forum General Discussion
    Replies: 27
    Last Post: 04-05-2005, 12:55 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •