DevX Home    Today's Headlines   Articles Archive   Tip Bank   Forums   

Results 1 to 8 of 8

Thread: A Web site for all you OO pessimists out there

  1. #1
    T. Hoskins Guest

    A Web site for all you OO pessimists out there


    Programming Wisdom Center http://www.geocities.com/SiliconVall...888/index.html
    Since the recent thread titled "OO Considered Harmful" seemed to generate
    quite a few responses, I thought that I would add some fuel to the fire by
    posting a URL to an anti OO web site. While I don't agree with everything
    this person has to say on the subject of OOP software development, I did
    find his web site content to be an interesting read.

    Personally, I am a big fan of component-based development. Currently, I am
    reading a book titled Java Modeling with UML by Peter Coad. I find it kind
    of amusing how one of the biggest proponents of OOAD and OOP is using composition
    rather than inheritance in this book. Don't get me wrong, for some applications
    I think the classic OOP approach where polymorphism is provided by inheritance
    is a useful and effective solution (especially when applied to small-scale
    development tasks). Even so, when it comes to n-tier and web-based software
    development, I think that stateless component-based development is the best
    solution in most circumstances. IMHO, traditional stateful objects (classic
    OOP) simply doesn't cut it in most circumstances.

    This URL http://pub2.ezboard.com/fcomputercon...picID=85.topic
    will lead you to a topic titled "Sanity Check: Is VB Good for COM development?"
    at Realrates.com BBS. This topic seems very relevant to the discussions going
    on the forums here at DevX and I was hoping that someone will take the time
    to read this post and then comment on what they are discussing over there
    in that thread.

    Note: I cross-posted part of this post (the second URL) to the Enterprise
    Development Client/Server Development forum. That forum seems to be farily
    silent lately so ...

  2. #2
    Richard Curzon Guest

    Re: A Web site for all you OO pessimists out there

    All the OOP=inheritance misinformation is just from the johnny-come-lately
    hypesters. OOP never was really about that, from its roots to modern best
    practices.

    Truly OOP-literate Books like the Gang of Four, whose authors are really
    aware of the raison d'etre and the history, promote composition over
    inheritance whenever possible. (i.e., they say that explicitly).

    Trace back the heritage of such programming-literate people... you find
    yourself with people like David Parnas in the 70's and 80's. The smart
    people then and now agree, on how to use composition and information hiding
    (encapsulation) to build robust maintainable programs.

    Even people like Booch (who embrace inheritance more than GOF) tried to
    caution newbies, saying that their was a "tension" between inheritance and
    encapsulation, that designers should beware. Unfortunately the stuff he
    said about inheritance was very well understood, the cautions not
    understood. So, entirely ignored by the hypesters.

    It's only the opportunists and shams who ever told us that OOP was about
    inheritance in the first place, encouraged fools to rush in, after dropping
    a few thousand on training of course. <g>

    Which leads to the unfortunate and ill-informaed "backlash" now against OOP,
    *because* it's about inheritance. Go back and read the books, burn the "3
    day expert" rip-off material, and all will be clear.. <g>

    regards
    Richard.


    "T. Hoskins" <starter@movemail.com> wrote in message
    news:3a74c7b9$1@news.devx.com...
    >
    > Programming Wisdom Center

    http://www.geocities.com/SiliconVall...888/index.html
    > Since the recent thread titled "OO Considered Harmful" seemed to generate
    > quite a few responses, I thought that I would add some fuel to the fire by
    > posting a URL to an anti OO web site. While I don't agree with everything
    > this person has to say on the subject of OOP software development, I did
    > find his web site content to be an interesting read.
    >
    > Personally, I am a big fan of component-based development. Currently, I am
    > reading a book titled Java Modeling with UML by Peter Coad. I find it kind
    > of amusing how one of the biggest proponents of OOAD and OOP is using

    composition
    > rather than inheritance in this book. Don't get me wrong, for some

    applications
    > I think the classic OOP approach where polymorphism is provided by

    inheritance
    > is a useful and effective solution (especially when applied to small-scale
    > development tasks). Even so, when it comes to n-tier and web-based

    software
    > development, I think that stateless component-based development is the

    best
    > solution in most circumstances. IMHO, traditional stateful objects

    (classic
    > OOP) simply doesn't cut it in most circumstances.
    >
    > This URL

    http://pub2.ezboard.com/fcomputercon...topicID=85.top
    ic
    > will lead you to a topic titled "Sanity Check: Is VB Good for COM

    development?"
    > at Realrates.com BBS. This topic seems very relevant to the discussions

    going
    > on the forums here at DevX and I was hoping that someone will take the

    time
    > to read this post and then comment on what they are discussing over there
    > in that thread.
    >
    > Note: I cross-posted part of this post (the second URL) to the Enterprise
    > Development Client/Server Development forum. That forum seems to be

    farily
    > silent lately so ...




  3. #3
    Richard Dalton . Guest

    Re: A Web site for all you OO pessimists out there




    "Richard Curzon" <ssi@interNOSPAMlog.com> wrote:
    >burn the "3 day expert" rip-off material, and all will be
    >clear.. <g>


    I feel the chill of a breath of fresh air around my ankles.



    -Rd

  4. #4
    Richard Dalton . Guest

    Re: A Web site for all you OO pessimists out there


    "T. Hoskins" <starter@movemail.com> wrote:

    Hi, I'll read the anti-OO page later. I've just had a
    quick look at the VB/COM discussion. Thanks for that.

    It echo's sentiments which I think are becomming clear from
    this board, although don't seem to have fully taken root
    in the industry.

    I am convinced that a lot of people building objects in VB
    and running them in COM/MTS "think" they are getting object
    pooling. The myth is propagated further because in some
    cases they may be getting ADO pooling and/or other
    caching, but they are not getting MTS object pooling if the
    objects are in VB.

    The big reason for using MTS/COM+ as I see it, is the ability
    to manage transactions from the Business layer with your
    DB layer perhaps running on another box.

    This works with VB but you dont get any extra performance. If
    you want MTS/COM+ to contribute to scalability you need to look at
    VC++

    Of course .Net will make this argument go away, but until then
    if anyone tells you that you need MTS/COM+ for
    performance/scalability/object pooling then they either don't
    know what they are talking about, or they hope you don't.


    -Richard





  5. #5
    Richard Dalton . Guest

    Re: A Web site for all you OO pessimists out there


    "T. Hoskins" <starter@movemail.com> wrote:
    >While I don't agree with everything this person has to say on >the subject

    of OOP software development, I did
    >find his web site content to be an interesting read.



    I think you and I probably find the same parts of his site
    to be obectionable.

    He makes one or two interesting points. On the whole he
    comes accross as a bit too eager. In fact it sounds like
    a typical "I don't like what I don't understand" kind of
    argument. I can regognise this type of argument because
    I frequently make them myself, it's a weakness I suppose.

    Hid article is heavy on the "Some Experts" and "Experts Say"
    type of references. I'd prefer to see names. When people
    use the term "Some experts" they often mean the team of
    imaginary experts living in their head.

    Some of the things he points out such as for instance ad-hoc
    queries, have been discussed here, and can be accomodated
    quite easily in an OO paradigm.

    His claim that OO became popular "because of" GIU interfaces
    is MAD. Jacobson was using interfaces, components and
    tiered architectures in the 60's.

    He also makes another mad point that I have to call him on:

    "I am saddened to see non-OO research, training, and tools stop advancing
    because everybody focuses on OOP instead. OOP's alleged advantages are becoming
    a self-fulfilling-prophecy because of this. They should prove that OO is
    better for everything first before pulling the plug on the others."

    By the very existance of his article he shows that there are
    people who do not agree with OO. Why aren't these people
    researching non-OO techniques? He seems to want people who
    think OO is a good idea to devote some time to non-OO
    techniques. Let the non-OOers like himself do the non-OO research.

    I agree with him that the examples in books etc are stupid.
    They don't even remotely approach the complexity of real world
    problem. But this isn't an OO problem. It's a book problem.
    How many "Hello World" and "Temperature Convertor's" have you
    written. These are the mainstay of procedural books.

    Actually I've said enough.

    Interesting article. Some good points overshadowed by some
    not so good points and dodgy reasoning.

    Thanks for pointing out the link.

    -Richard




  6. #6
    Thomas Eyde Guest

    Re: A Web site for all you OO pessimists out there

    He has one good point: OO does not solve every problem. But what does?

    What I miss is counterproof that procedural is best. I can only find
    suggestions on how OO is not optimal, but nothing on procedural being
    optimal.

    He is the procedural expert, he should provide lots of them.

    /Thomas



  7. #7
    Gunnar Skogsholm Guest

    Re: A Web site for all you OO pessimists out there

    Both of these extremes are incorrect. OOP is encapsulation, inheritance,
    and polymorphism. It's a tripod.

    "Richard Curzon" <ssi@interNOSPAMlog.com> wrote in message
    news:3a74d1e4@news.devx.com...
    > All the OOP=inheritance misinformation is just from the johnny-come-lately
    > hypesters. OOP never was really about that, from its roots to modern best
    > practices.
    >
    > Truly OOP-literate Books like the Gang of Four, whose authors are really
    > aware of the raison d'etre and the history, promote composition over
    > inheritance whenever possible. (i.e., they say that explicitly).
    >
    > Trace back the heritage of such programming-literate people... you find
    > yourself with people like David Parnas in the 70's and 80's. The smart
    > people then and now agree, on how to use composition and information

    hiding
    > (encapsulation) to build robust maintainable programs.
    >
    > Even people like Booch (who embrace inheritance more than GOF) tried to
    > caution newbies, saying that their was a "tension" between inheritance and
    > encapsulation, that designers should beware. Unfortunately the stuff he
    > said about inheritance was very well understood, the cautions not
    > understood. So, entirely ignored by the hypesters.
    >
    > It's only the opportunists and shams who ever told us that OOP was about
    > inheritance in the first place, encouraged fools to rush in, after

    dropping
    > a few thousand on training of course. <g>
    >
    > Which leads to the unfortunate and ill-informaed "backlash" now against

    OOP,
    > *because* it's about inheritance. Go back and read the books, burn the "3
    > day expert" rip-off material, and all will be clear.. <g>
    >
    > regards
    > Richard.
    >
    >
    > "T. Hoskins" <starter@movemail.com> wrote in message
    > news:3a74c7b9$1@news.devx.com...
    > >
    > > Programming Wisdom Center

    > http://www.geocities.com/SiliconVall...888/index.html
    > > Since the recent thread titled "OO Considered Harmful" seemed to

    generate
    > > quite a few responses, I thought that I would add some fuel to the fire

    by
    > > posting a URL to an anti OO web site. While I don't agree with

    everything
    > > this person has to say on the subject of OOP software development, I did
    > > find his web site content to be an interesting read.
    > >
    > > Personally, I am a big fan of component-based development. Currently, I

    am
    > > reading a book titled Java Modeling with UML by Peter Coad. I find it

    kind
    > > of amusing how one of the biggest proponents of OOAD and OOP is using

    > composition
    > > rather than inheritance in this book. Don't get me wrong, for some

    > applications
    > > I think the classic OOP approach where polymorphism is provided by

    > inheritance
    > > is a useful and effective solution (especially when applied to

    small-scale
    > > development tasks). Even so, when it comes to n-tier and web-based

    > software
    > > development, I think that stateless component-based development is the

    > best
    > > solution in most circumstances. IMHO, traditional stateful objects

    > (classic
    > > OOP) simply doesn't cut it in most circumstances.
    > >
    > > This URL

    >

    http://pub2.ezboard.com/fcomputercon...topicID=85.top
    > ic
    > > will lead you to a topic titled "Sanity Check: Is VB Good for COM

    > development?"
    > > at Realrates.com BBS. This topic seems very relevant to the discussions

    > going
    > > on the forums here at DevX and I was hoping that someone will take the

    > time
    > > to read this post and then comment on what they are discussing over

    there
    > > in that thread.
    > >
    > > Note: I cross-posted part of this post (the second URL) to the

    Enterprise
    > > Development Client/Server Development forum. That forum seems to be

    > farily
    > > silent lately so ...

    >
    >




  8. #8
    Gunnar Skogsholm Guest

    Re: A Web site for all you OO pessimists out there

    Yes, this is a very interesting point of view. I have an article that has
    documented some OO resuse cases, and found that when done correctly, there
    have been some big OO resuse success. Using an OOP approach, we created an
    application framework that resulted in an 8-1 code reduction. The increase
    in maintainability resulted in being able to leverage a much smaller
    development staff to maintain a large amount of functionality.

    "T. Hoskins" <starter@movemail.com> wrote in message
    news:3a74c7b9$1@news.devx.com...
    >
    > Programming Wisdom Center

    http://www.geocities.com/SiliconVall...888/index.html
    > Since the recent thread titled "OO Considered Harmful" seemed to generate
    > quite a few responses, I thought that I would add some fuel to the fire by
    > posting a URL to an anti OO web site. While I don't agree with everything
    > this person has to say on the subject of OOP software development, I did
    > find his web site content to be an interesting read.
    >
    > Personally, I am a big fan of component-based development. Currently, I am
    > reading a book titled Java Modeling with UML by Peter Coad. I find it kind
    > of amusing how one of the biggest proponents of OOAD and OOP is using

    composition
    > rather than inheritance in this book. Don't get me wrong, for some

    applications
    > I think the classic OOP approach where polymorphism is provided by

    inheritance
    > is a useful and effective solution (especially when applied to small-scale
    > development tasks). Even so, when it comes to n-tier and web-based

    software
    > development, I think that stateless component-based development is the

    best
    > solution in most circumstances. IMHO, traditional stateful objects

    (classic
    > OOP) simply doesn't cut it in most circumstances.
    >
    > This URL

    http://pub2.ezboard.com/fcomputercon...topicID=85.top
    ic
    > will lead you to a topic titled "Sanity Check: Is VB Good for COM

    development?"
    > at Realrates.com BBS. This topic seems very relevant to the discussions

    going
    > on the forums here at DevX and I was hoping that someone will take the

    time
    > to read this post and then comment on what they are discussing over there
    > in that thread.
    >
    > Note: I cross-posted part of this post (the second URL) to the Enterprise
    > Development Client/Server Development forum. That forum seems to be

    farily
    > silent lately so ...




Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
HTML5 Development Center
 
 
FAQ
Latest Articles
Java
.NET
XML
Database
Enterprise
Questions? Contact us.
C++
Web Development
Wireless
Latest Tips
Open Source


   Development Centers

   -- Android Development Center
   -- Cloud Development Project Center
   -- HTML5 Development Center
   -- Windows Mobile Development Center