Why Is OOP So Hard?

[Phil Weber]

Grist for the mill...
http://www.fawcette.com/javapro/2002...awan_10_15_02/
--
Phil Weber

[MarkN]

To add to it, check out the interview with Alan Cooper

http://www.ftpconferences.com/cooper..._interview.asp

Especially this:

<SNIP>
<FTP>
We started out discussing the .NET platform. Let’s get back to that. What
do you think about the .NOT movement among VB6ers?
</FTP>

<COOPER>
I’ve been buying books on C# and .NET for the past few weeks. .NET is without
a doubt extremely complex and extremely difficult. Then again, what programmers
do is extremely complex and extremely difficult. I agree with Grady Booch
when he says programming is hard and intellectually demanding and always
will be and always was. A lot of people got sucked into the programming world
during the .COM boom. Now they’re discovering it’s not easy. I think that
as we build bigger and more complex software systems we’re going to find
that the tools are going to give us more power and reach but they’ll never
make things easier.
<COOPER>
</SNIP>



As for the article Phil Posted:

Some people fall into the category it describes. I think, though, that some
can't learn OO even if they can currently program (my opinion is that OO
is better because it makes for more maintainable and tunable code). Some
can, but don't want to. Some are but don't know they really are. Sure,
most can and do use objects in their coding - but to create their own, that
is a different story. They probably could handle tangible or concrete objects
(Customer, Invoice) but things that aren't tangible or are abstract are difficult
to objectize by most. This being said, I think the problem is not with OOD/P
but those involved in programming. Not everyone is cut out to be a doctor.
We shouldn't be painting diagrams on the patients so Joe Shmoe can do surgery
because he knows how to gut a deer. I'm not saying everything should be
done OO - not saying it shouldn't be either.

[jon ogden]

But if programming is difficult, and intellectually demanding, then managers
won't be able to order up a couple of dozen programmers from their favorite
H1B importer, without worrying about whether they'd do better to hire people
with brains and talent. <grin >


"MarkN" <vb.@127.0.0.1> wrote:
>
>To add to it, check out the interview with Alan Cooper
>
>http://www.ftpconferences.com/cooper..._interview.asp
>
>Especially this:
>
><SNIP>
><FTP>
>We started out discussing the .NET platform. Let’s get back to that. What
>do you think about the .NOT movement among VB6ers?
></FTP>
>
><COOPER>
>I’ve been buying books on C# and .NET for the past few weeks. .NET is without
>a doubt extremely complex and extremely difficult. Then again, what programmers
>do is extremely complex and extremely difficult. I agree with Grady Booch
>when he says programming is hard and intellectually demanding and always
>will be and always was. A lot of people got sucked into the programming
world
>during the .COM boom. Now they’re discovering it’s not easy. I think that
>as we build bigger and more complex software systems we’re going to find
>that the tools are going to give us more power and reach but they’ll never
>make things easier.
><COOPER>
></SNIP>
>
>
>
>As for the article Phil Posted:
>
>Some people fall into the category it describes. I think, though, that
some
>can't learn OO even if they can currently program (my opinion is that OO
>is better because it makes for more maintainable and tunable code). Some
>can, but don't want to. Some are but don't know they really are. Sure,
>most can and do use objects in their coding - but to create their own, that
>is a different story. They probably could handle tangible or concrete objects
>(Customer, Invoice) but things that aren't tangible or are abstract are
difficult
>to objectize by most. This being said, I think the problem is not with
OOD/P
>but those involved in programming. Not everyone is cut out to be a doctor.
> We shouldn't be painting diagrams on the patients so Joe Shmoe can do surgery
>because he knows how to gut a deer. I'm not saying everything should be
>done OO - not saying it shouldn't be either.
>
>

[MarkN]

Typing code in a certain syntax to make a computer do a certain thing isn't
that difficult. Using tools wisely and architecting and building good systems/applications
is.


"jon ogden" <jon_ogden@tjx.com> wrote:
>
>But if programming is difficult, and intellectually demanding, then managers
>won't be able to order up a couple of dozen programmers from their favorite
>H1B importer, without worrying about whether they'd do better to hire people
>with brains and talent. <grin >
>
>
>"MarkN" <vb.@127.0.0.1> wrote:
>>
>>To add to it, check out the interview with Alan Cooper
>>
>>http://www.ftpconferences.com/cooper..._interview.asp
>>
>>Especially this:
>>
>><SNIP>
>><FTP>
>>We started out discussing the .NET platform. Let’s get back to that. What
>>do you think about the .NOT movement among VB6ers?
>></FTP>
>>
>><COOPER>
>>I’ve been buying books on C# and .NET for the past few weeks. .NET is without
>>a doubt extremely complex and extremely difficult. Then again, what programmers
>>do is extremely complex and extremely difficult. I agree with Grady Booch
>>when he says programming is hard and intellectually demanding and always
>>will be and always was. A lot of people got sucked into the programming
>world
>>during the .COM boom. Now they’re discovering it’s not easy. I think that
>>as we build bigger and more complex software systems we’re going to find
>>that the tools are going to give us more power and reach but they’ll never
>>make things easier.
>><COOPER>
>></SNIP>
>>
>>
>>
>>As for the article Phil Posted:
>>
>>Some people fall into the category it describes. I think, though, that
>some
>>can't learn OO even if they can currently program (my opinion is that OO
>>is better because it makes for more maintainable and tunable code). Some
>>can, but don't want to. Some are but don't know they really are. Sure,
>>most can and do use objects in their coding - but to create their own,
that
>>is a different story. They probably could handle tangible or concrete
objects
>>(Customer, Invoice) but things that aren't tangible or are abstract are
>difficult
>>to objectize by most. This being said, I think the problem is not with
>OOD/P
>>but those involved in programming. Not everyone is cut out to be a doctor.
>> We shouldn't be painting diagrams on the patients so Joe Shmoe can do
surgery
>>because he knows how to gut a deer. I'm not saying everything should be
>>done OO - not saying it shouldn't be either.
>>
>>
>

[Patrick Troughton]

Don't forget Cooper also states......

<quote>
Without a doubt .NET is one of the great simplifying tools in the world of
software.
</quote>

/Pat
---------------------------
Get .NET, it's hella phat.
---------------------------

"MarkN" <vb.@127.0.0.1> wrote:
>
>To add to it, check out the interview with Alan Cooper
>
>http://www.ftpconferences.com/cooper..._interview.asp
>
>Especially this:
>
><SNIP>
><FTP>
>We started out discussing the .NET platform. Let’s get back to that. What
>do you think about the .NOT movement among VB6ers?
></FTP>
>
><COOPER>
>I’ve been buying books on C# and .NET for the past few weeks. .NET is without
>a doubt extremely complex and extremely difficult. Then again, what programmers
>do is extremely complex and extremely difficult. I agree with Grady Booch
>when he says programming is hard and intellectually demanding and always
>will be and always was. A lot of people got sucked into the programming
world
>during the .COM boom. Now they’re discovering it’s not easy. I think that
>as we build bigger and more complex software systems we’re going to find
>that the tools are going to give us more power and reach but they’ll never
>make things easier.
><COOPER>
></SNIP>
>

[Steve]

Nice contradiction. Cooper is an old wind bag who can't finish a thought.

In one sentence he proclaims it (.Net) is complicated and in the next that
it is simple. In one sentence he shows how he feels, in the next he remembers
he's a Bill Gates whipping boy and that Microsoft has him in their back pocket.

"Patrick Troughton" <vb.@127.0.0.1> wrote:
>
>Don't forget Cooper also states......
>
><quote>
>Without a doubt .NET is one of the great simplifying tools in the world
of
>software.
></quote>
>
>/Pat
>---------------------------
>Get .NET, it's hella phat.
>---------------------------
>
>"MarkN" <vb.@127.0.0.1> wrote:
>>
>>To add to it, check out the interview with Alan Cooper
>>
>>http://www.ftpconferences.com/cooper..._interview.asp
>>
>>Especially this:
>>
>><SNIP>
>><FTP>
>>We started out discussing the .NET platform. Let’s get back to that. What
>>do you think about the .NOT movement among VB6ers?
>></FTP>
>>
>><COOPER>
>>I’ve been buying books on C# and .NET for the past few weeks. .NET is without
>>a doubt extremely complex and extremely difficult. Then again, what programmers
>>do is extremely complex and extremely difficult. I agree with Grady Booch
>>when he says programming is hard and intellectually demanding and always
>>will be and always was. A lot of people got sucked into the programming
>world
>>during the .COM boom. Now they’re discovering it’s not easy. I think that
>>as we build bigger and more complex software systems we’re going to find
>>that the tools are going to give us more power and reach but they’ll never
>>make things easier.
>><COOPER>
>></SNIP>
>>
>

[MarkN]

Not everyone is right about everything.

I wonder was he talking about .Net the platform or VS.Net the tool or both?

Notice he said 'one of'. Not 'the only'. Not 'the first'.

"Patrick Troughton" <vb.@127.0.0.1> wrote:
>
>Don't forget Cooper also states......
>
><quote>
>Without a doubt .NET is one of the great simplifying tools in the world
of
>software.
></quote>
>
>/Pat
>---------------------------
>Get .NET, it's hella phat.
>---------------------------
>
>"MarkN" <vb.@127.0.0.1> wrote:
>>
>>To add to it, check out the interview with Alan Cooper
>>
>>http://www.ftpconferences.com/cooper..._interview.asp
>>
>>Especially this:
>>
>><SNIP>
>><FTP>
>>We started out discussing the .NET platform. Let’s get back to that. What
>>do you think about the .NOT movement among VB6ers?
>></FTP>
>>
>><COOPER>
>>I’ve been buying books on C# and .NET for the past few weeks. .NET is without
>>a doubt extremely complex and extremely difficult. Then again, what programmers
>>do is extremely complex and extremely difficult. I agree with Grady Booch
>>when he says programming is hard and intellectually demanding and always
>>will be and always was. A lot of people got sucked into the programming
>world
>>during the .COM boom. Now they’re discovering it’s not easy. I think that
>>as we build bigger and more complex software systems we’re going to find
>>that the tools are going to give us more power and reach but they’ll never
>>make things easier.
>><COOPER>
>></SNIP>
>>
>

[MarkN]

I'm guessing he meant:

- Developing maintainable, maluable and multifaceted distributed 'applications'
is very difficult.

- .Net in and of itself is difficult (we're not talking about whipping out
single focus apps).

- But doing the above apps in .Net makes them much simpler.

Of course you could be right.

"Steve" <steveo@steveo.com> wrote:
>
>Nice contradiction. Cooper is an old wind bag who can't finish a thought.
>
>In one sentence he proclaims it (.Net) is complicated and in the next that
>it is simple. In one sentence he shows how he feels, in the next he remembers
>he's a Bill Gates whipping boy and that Microsoft has him in their back
pocket.
>
>"Patrick Troughton" <vb.@127.0.0.1> wrote:
>>
>>Don't forget Cooper also states......
>>
>><quote>
>>Without a doubt .NET is one of the great simplifying tools in the world
>of
>>software.
>></quote>
>>
>>/Pat
>>---------------------------
>>Get .NET, it's hella phat.
>>---------------------------
>>
>>"MarkN" <vb.@127.0.0.1> wrote:
>>>
>>>To add to it, check out the interview with Alan Cooper
>>>
>>>http://www.ftpconferences.com/cooper..._interview.asp
>>>
>>>Especially this:
>>>
>>><SNIP>
>>><FTP>
>>>We started out discussing the .NET platform. Let’s get back to that. What
>>>do you think about the .NOT movement among VB6ers?
>>></FTP>
>>>
>>><COOPER>
>>>I’ve been buying books on C# and .NET for the past few weeks. .NET is
without
>>>a doubt extremely complex and extremely difficult. Then again, what programmers
>>>do is extremely complex and extremely difficult. I agree with Grady Booch
>>>when he says programming is hard and intellectually demanding and always
>>>will be and always was. A lot of people got sucked into the programming
>>world
>>>during the .COM boom. Now they’re discovering it’s not easy. I think that
>>>as we build bigger and more complex software systems we’re going to find
>>>that the tools are going to give us more power and reach but they’ll never
>>>make things easier.
>>><COOPER>
>>></SNIP>
>>>
>>
>

[Arthur Wood]

<SNIP> Then again, what programmers
>do is extremely complex and extremely difficult. I agree with Grady Booch
>when he says programming is hard and intellectually demanding and always
>will be and always was.</SNIP>

just to add another idea, this is a quote from Tom Hanks in "A League of
Their Own" - toward the end of the movie, Geena Davis's character complains
in tears that what she has been doing has been hard, and she doesn't want
to do it any more, and Hanks responds:

"Of course it's hard - it's supposed to be hard. If it weren't hard, then
everybody would do it".

Arthur Wood

[Jason]

"Steve" <steveo@steveo.com> wrote:
>
>Nice contradiction. Cooper is an old wind bag who can't finish a thought.
>
>In one sentence he proclaims it (.Net) is complicated and in the next that
>it is simple. In one sentence he shows how he feels, in the next he remembers
>he's a Bill Gates whipping boy and that Microsoft has him in their back
pocket.
>
>"Patrick Troughton" <vb.@127.0.0.1> wrote:

But .NET is both very complex and simplifying at the same time. If I have
a hammer and a chisel, I can carve a table out of a redwood tree. It will
take me some time and not be very efficient, but I can do it. Hammers and
chisels are simple, and the operation of them is simple, but making the table
will require some skill.

Now if you have a sawmill and a woodworking shop with modern electrical tools,
the tools just got more complex. Any one of them used alone isn't that bad,
but learning all of them can take some time.

When you know most of the new tools, making a table is not only simpler,
but you can do a better job in much less time. And you save a lot of trees.
So things can be both very complex and simpifying at the same time, right?

Sorry. I just felt an analogy coming on.

[Mike Mitchell]

On Fri, 18 Oct 2002 00:43:39 -0700, "Phil Weber"
<pweber@nospam.fawcette.com> wrote:

>Grist for the mill...
>http://www.fawcette.com/javapro/2002...awan_10_15_02/

I added this to the replies under that rubric:

OOP is complex. Procedural programming is simple. OOP wants to hide as
much as it can. Procedural programming hides only the bits you don't
want to see all the time. A procedural program is easier to understand
than a OOD program. OOP is like peeling an onion while walking across
a minefield in the dark. Have you reached that final layer of
(in)comprehension yet, or have you already blown your last synapse to
smithereens? Layers of inheritance make OOP slow and bloated. Fewer
layers makes procedural programming fast and compact.

Extracting the sense from OOP is like trying to separate the eggs from
a ham omelette. Impossible! Procedural programming on the other hand
is like ham and eggs - two separate ingredients on the dish. Easy to
separate the eggs. Or add another if one is hungry. Or substitute
bacon for the ham. Or add mushrooms. But only for mom, because pop
doesn't like mushrooms. We all rise from the table replete, nourished,
ready for whatever life will throw at us - even OOP! While with OOP we
just eat the omelette and keep stumm.

Procedural programming makes sense. OOP doesn't. That's why OOP is
hard. Hard to understand. Hard to teach. Hard to measure. Hard to
justify. When we tidy the house, we go to the cupboard and get a
broom. One broom used everywhere - the kitchen, the living room, the
bedrooms, the stairs. In an OOP world each room would have its own
broom, which would actually be derived from a more fundamental kind of
broom, like a brush. When we're in the kitchen we use the
kitchen.broom. But as soon as we leave the kitchen, we must stop using
that specific broom, and use, say, the stairs.broom. Not only that, we
can *only* use the specific broom! Conversely, with a procedural,
generalised broom, should the handle snap or a burglar break in and
steal the whole broom, we can simply replace it with one borrowed from
the neighbours, even if it isn't exactly the same kind of broom we had
originally.

Try that with kitchen.broom! It won't work. The OOP broom has to be
derived from the base broom, which was a brush.

In other words, procedural programming is not only simple, it is also
flexible and adaptable. OOP is far more rigid, illogical and
unforgiving. Deriving, inheriting, polymorphing, encapsulating,
overriding, overloading - everything bar the kitchen.sink!

That's why I say, OOP is POOP!

****************************************

Then I read the interview with AC. Man, he's certainly not backward in
coming forward, is he?!! But he does raise some interesting points, so
I suppose one just has to read all of it to pull out the fundamentals.

Couple [of] things caught my eye:

"Electricians aren’t professionals because they’re externally
regulated. That’s why dentists get paid a lot more than electricians."

Oh, so it's not because dentists are operating drills at high speeds
and moving other sharp implements around inside our mouths, then? Or
can recognise pre-cancerous lesions? Or pull a tooth without ripping
half the gum out? Or install fillings, crowns, bridges, and false
teeth? Or take X-Rays of our embedded wisdom teeth? No, it's because
they're externally regulated!! Duh! Why am I just not as perceptive as
Uncle Al?

And:

"You have to understand that BASIC is an acronym for Beginners
All-purpoSe Instruction Code."

Al must have just forgotten! What it REALLY stands for - and I know
this is nit-picking of the first order - what it actually stands for
is:

Beginner's All-purpose Symbolic Instruction Code

And:

"I just think OOP is hard."

Yeah, Al! You and me both!

MM

[Tom Shelton]

[snip]

Mike,

I just have to tell you - any one who knows even the rudiments of OOD is
going to laugh their posteriors off at your broom example (I know I did)...
You have it totally backwords my man.

Tom Shelton

[MarkN]

Yes, OO can be complex. And procedural is 'simple'. But what we do is complex.
And using something that is simple to do something complex actually makes
it more it more complex. But with OO the complexity doesn't change and it
makes the more complex things simple. With an OO language you can do both
- procedural and OOP. With a procedural type language - you have no choice.

For your egg and ham example - you've got it backwards. In proceedural everything
is mixed together and you can't easily separate it. OO allows you to separate
concerns (and AOP allows it even further).

Once one understands OOP, it makes sense. Not everything does at first especially
with abstract things. Sometimes I will write a little procedural code if
things are a little too abstract but then objectize it (refactor it ).
It also helps to think of how something can be used two ways.

Proceedural is everything you say OOP is and OOP is everything you say procedural
is.


But what is the point. You won't ever believe it till you see it and you
can't see it. I've done and seen enough proceedural on the mainframe and
PC to know the advantage of OO.



Mike Mitchell <kylix_is@yahoo.co.uk> wrote:
>On Fri, 18 Oct 2002 00:43:39 -0700, "Phil Weber"
><pweber@nospam.fawcette.com> wrote:
>
>>Grist for the mill...
>>http://www.fawcette.com/javapro/2002...awan_10_15_02/
>
>I added this to the replies under that rubric:
>
>OOP is complex. Procedural programming is simple. OOP wants to hide as
>much as it can. Procedural programming hides only the bits you don't
>want to see all the time. A procedural program is easier to understand
>than a OOD program. OOP is like peeling an onion while walking across
>a minefield in the dark. Have you reached that final layer of
>(in)comprehension yet, or have you already blown your last synapse to
>smithereens? Layers of inheritance make OOP slow and bloated. Fewer
>layers makes procedural programming fast and compact.
>
>Extracting the sense from OOP is like trying to separate the eggs from
>a ham omelette. Impossible! Procedural programming on the other hand
>is like ham and eggs - two separate ingredients on the dish. Easy to
>separate the eggs. Or add another if one is hungry. Or substitute
>bacon for the ham. Or add mushrooms. But only for mom, because pop
>doesn't like mushrooms. We all rise from the table replete, nourished,
>ready for whatever life will throw at us - even OOP! While with OOP we
>just eat the omelette and keep stumm.
>
>Procedural programming makes sense. OOP doesn't. That's why OOP is
>hard. Hard to understand. Hard to teach. Hard to measure. Hard to
>justify. When we tidy the house, we go to the cupboard and get a
>broom. One broom used everywhere - the kitchen, the living room, the
>bedrooms, the stairs. In an OOP world each room would have its own
>broom, which would actually be derived from a more fundamental kind of
>broom, like a brush. When we're in the kitchen we use the
>kitchen.broom. But as soon as we leave the kitchen, we must stop using
>that specific broom, and use, say, the stairs.broom. Not only that, we
>can *only* use the specific broom! Conversely, with a procedural,
>generalised broom, should the handle snap or a burglar break in and
>steal the whole broom, we can simply replace it with one borrowed from
>the neighbours, even if it isn't exactly the same kind of broom we had
>originally.
>
>Try that with kitchen.broom! It won't work. The OOP broom has to be
>derived from the base broom, which was a brush.
>
>In other words, procedural programming is not only simple, it is also
>flexible and adaptable. OOP is far more rigid, illogical and
>unforgiving. Deriving, inheriting, polymorphing, encapsulating,
>overriding, overloading - everything bar the kitchen.sink!
>
>That's why I say, OOP is POOP!
>
>****************************************
>
>Then I read the interview with AC. Man, he's certainly not backward in
>coming forward, is he?!! But he does raise some interesting points, so
>I suppose one just has to read all of it to pull out the fundamentals.
>
>Couple [of] things caught my eye:
>
>"Electricians aren’t professionals because they’re externally
>regulated. That’s why dentists get paid a lot more than electricians."
>
>Oh, so it's not because dentists are operating drills at high speeds
>and moving other sharp implements around inside our mouths, then? Or
>can recognise pre-cancerous lesions? Or pull a tooth without ripping
>half the gum out? Or install fillings, crowns, bridges, and false
>teeth? Or take X-Rays of our embedded wisdom teeth? No, it's because
>they're externally regulated!! Duh! Why am I just not as perceptive as
>Uncle Al?
>
>And:
>
>"You have to understand that BASIC is an acronym for Beginners
>All-purpoSe Instruction Code."
>
>Al must have just forgotten! What it REALLY stands for - and I know
>this is nit-picking of the first order - what it actually stands for
>is:
>
>Beginner's All-purpose Symbolic Instruction Code
>
>And:
>
>"I just think OOP is hard."
>
>Yeah, Al! You and me both!
>
>MM

[Rune Bivrin]

"Jason" <jason@hotmail.com> wrote in news:3db05c5c$1@tnews.web.devx.com:

> But .NET is both very complex and simplifying at the same time. If I
> have a hammer and a chisel, I can carve a table out of a redwood tree.
> It will take me some time and not be very efficient, but I can do it.
> Hammers and chisels are simple, and the operation of them is simple,
> but making the table will require some skill.
>
> Now if you have a sawmill and a woodworking shop with modern
> electrical tools, the tools just got more complex. Any one of them
> used alone isn't that bad, but learning all of them can take some
> time.
>
> When you know most of the new tools, making a table is not only
> simpler, but you can do a better job in much less time. And you save
> a lot of trees.
> So things can be both very complex and simpifying at the same time,
> right?
>
> Sorry. I just felt an analogy coming on.
>
>

ROTFLMAO!

I will use this analogy. It's brilliant!

--
Rune Bivrin
- OOP since 1989
- SQL Server since 1990
- VB since 1991

[Zane Thomas]

On Fri, 18 Oct 2002 21:01:44 +0100, Mike Mitchell <kylix_is@yahoo.co.uk>
wrote:

>OOP is complex.

Mike is slow.


--

*--------={ Fine Art for .NET }=--------*
| .Net Components @ www.abderaware.com |
*---------------------------------------*

Turn on, tune in, download.
zane a@t abderaware.com