|
|
Posted by Tony Marston on 11/29/06 11:01
"Jerry Stuckle" <jstucklex@attglobal.net> wrote in message
news:VomdnXOSpecHdvHYnZ2dnUVZ_oadnZ2d@comcast.com...
> Michael Fesser wrote:
>> .oO(Tony Marston)
>>
>>
>>>I disagree. it is *not* necessary for the simple reason that the code
>>>will perform exactly the same function whether methods and properties are
>>>marked as public/private/protected or not.
>>
>>
>> Then why do we use OOP and high-level languages like PHP at all? Pure
>> hand-written assembler code will perform exactly the same function.
>>
>>
>>>>It prevents developers from doing things that shouldn't be done, for
>>>>example calling an internal method out of context. I don't want all my
>>>>methods being publicly available, simply in order to avoid errors and
>>>>unpredictable results.
>>>
>>>That is a matter for programmer discipline, it is not a matter of
>>>additional functionality. The code will do exactly the same with or
>>>without it.
>>
>>
>> The code written in a language like Delphi for example will also do
>> exactly the same with all type checks, range checks, overflow checks
>> etc. turned off. But does it make sense to do that and just rely on
>> "programmer discipline"? No, it doesn't, because it will lead to
>> erroneous code on the long run.
>>
>> Compilers are able to automatically check a lot of things and warn the
>> developer if he made a mistake. Such checks and restrictions don't add
>> any functionality, but are necessary in order to write reliable code.
>>
>> The same goes for visibility declarations. I don't rely on discipline or
>> a comment like "please don't call this method". If a method is not meant
>> to be called directly then it's declared as such - problem solved.
>>
>> Micha
>
> Micha,
>
> Tony and I have been into this before. He breaks into conversations
> trying to spout his version of OO, with a few blogs from people no one
> every heard of to back him up.
I see. So in your opinion Martin Fowler is of of these "people no one ever
heard of "? He says, like I do, that "Encapsulation Wasn't Meant To Mean
Data Hiding" at http://homepage.mac.com/keithray/blog/2006/02/22/
Are you saying that YOU are more of an expert than Martin Fowler? What
arrogance!
> It's not worth getting into the argument. He's just a troll with
> delusions of competency.
If everyone who disagrees with you is incompetent then the world is full of
idiots. Your opinion is not the only opinion, and there are plenty of
"experts" who have opposing views.
--
Tony Marston
http://www.tonymarston.net
http://www.radicore.org
Navigation:
[Reply to this message]
|