Chambers wrote:
> Point taken.  I got in the bad habit of using my headers as my 
> documentation, and coding my implementations to match them, rather than 
> keeping the documentation separate :)

That's a perfectly reasonable way to do it in a language that requires 
headers to start with.  A language with javadoc-like features? Nah.

The cool thing about Eiffel is you can not only produce a document that's 
the equivalent of a .H or javadoc (doxygen, etc) file, but also a "flat" 
version that includes all the documentation from the parent classes that you 
didn't override. So if you want the docs for "hashtable implemented using 
arrays of streams" documentation, you don't need to grope through three or 
more superclasses just to understand what that one class does.

> Of course, being Visual Studio, the Class explorer offers similar 
> functionality, but it's not in the same form I'm used to from looking at 
> header files.  Ah, well, I'll adjust eventually.

Yep. And Smalltalk is even more different, since you can open browsers on 
things like "all classes that implement a method called X". Smalltalk being 
one of those "duck-typing" languages.

Speaking of which, what is "duck typing" that isn't just "dynamic typing"?
WTF does that even mean?

> Yes, it is rare.  But good, so keep it up :)

I remember once way back, having written docs for the library, I was sitting 
there coding it up. (The project was a rewrite from BASIC to C, so I had a 
real good idea what the library needed to do.) I had a stack of 
documentation, and I was implementing each function.  The "best" programmer 
in the company comes up and asks what I'm doing, and I tell him. He points 
to the inch-thick print-out, asks what it is, and I tell him it's the docs. 
He says, sounding baffled, "How can you write the docs before the program?" 
I ask him "How can you know when you're done programming without writing the 
docs first?"

Ah, fun times. The first time I ever completely stopped taking a job seriously.

-- 
   Darren New, San Diego CA, USA (PST)
   "We'd like you to back-port all the changes in 2.0
    back to version 1.0."
   "We've done that already. We call it 2.0."

Post a reply to this message

clipka wrote:
> Chambers <Ben### [at] gmailcom_no_underscores> wrote:
>>> the buck does *not* stop here.
>> Shouldn't that be "the Buck"? ;)
> 
> ??
> 
> 

As in, David K Buck :)

-- 
Chambers

Post a reply to this message

David H. Burns wrote:
> What I have seen of OOP programming is something else.

Just out of curiosity (but you keep saying things like this, so I have 
to ask), where have you seen OOP programming?

-- 
Chambers

Post a reply to this message

Neeum Zawan wrote:

>     OK. Now I'm _really_ confused!
> 
>     Was your initial concern that POV-Ray - the *software* - will be 
> coded using OOP?
> 
>     Or was it that the language that *users* of POV-Ray will have to use 
> to describe their scenes will be OO?

That was my original concern but this thread seems to a mutated as 
threads do.
Any opposition to OOP seems to arouse a lot of feeling.
> 
>     Now if your concern is the latter (that you as a *user* will have to 
> describe your scene using OOP), then your concerns are not entirely 
> unfounded, but they won't be relevant for a while. 3.7 will allow you to 
> code as you have mostly done.
>  

Yes, I'm now told that a new scripting language is decades away so I suppose
my major concern is about 20 years or so too early. What language or 
with what
"philosophy" Pov-Ray is coded in is of only minor concern, if the end 
product remains
usable to me.

It's all been very interesting, though.

David

Post a reply to this message

Neeum Zawan wrote:
> On 07/23/09 11:15, David H. Burns wrote:
>>> I'm sure you think in OOP all the time with POV-Ray. You create a
>>> sphere. It's an object with certain properties (texture, location,
>>> etc). Now let's say you want to rotate the sphere, does it hurt so
>>> much to do something like:
>>>
>>> mysphere.rotate(45)
>>>
>>
>> No! I think in terms of objects, not of OOP. The concept of an "object"
>> as used
>> in Pov-Ray in a valuable tool for thinking and programming -and used
>> well there.
>> What I have seen of OOP programming is something else. :)
>>
>> David
>>
> 
>     OK. I have no idea what you're saying.
> 
>     What is the "No" referring to? Are you saying "no, it doesn't hurt"? 
> Are you saying "No, I hate that way of doing it"? Are you sweeping your 
> hand across my whole message and saying "No! I can't stand it!"?
> 


I'm sorry to seem unclear, but to me it seems that the "No" and the rest 
of my
statement clearly refer to the assertion that I think in OOP all the 
time when using
Pov-Ray. It seems to me that I don't, though perhaps others may know my
mind better that I do. In fact. since OOP means Object Oriented 
Programming, it seems
absurd to say that I think in it (or in any other kind of programming). 
On the other hand,
I am repeated told that I don't know what OOP means; maybe it *doesn't* mean
"Object Oriented Programming"(acronyms are always obstacles to 
communication),
but some philosophy or mystery into which I have not been initiated. In 
any case, it seems
an overstatement at least for someone else to say I thing in it. :)

David

Post a reply to this message

Chambers wrote:
> David H. Burns wrote:
>> What I have seen of OOP programming is something else.
> 
> Just out of curiosity (but you keep saying things like this, so I have 
> to ask), where have you seen OOP programming?
> 

Your point is well taken. I have *never* seen OOP *programming*. I 
looked at some
short OOP programs or excerpts in C++, python, and Visual Basic -- 
though I believe that the
earlier VB code that I have seen aren't really OO, but simply have some 
of the trappings.
The "OOP" code I've seen in VB could have been written without the 
OOP-like veneer.
To actually see OOP "programming", I suppose I would have to see some 
one actually
coding in OOP.
  I don't want to have to write something like those programs, I have 
seen to produce a Pov-Ray
script!

But, apparently, that is a worry for only the remote future.

David



David

Post a reply to this message

David H. Burns wrote:
> Warp wrote:
> 
>>   The whole idea of object-oriented programming is to make it *easier* to
>> write programs, especially compared to straightforward 
>> imperative/structured
>> programming (as the SDL is currently).
>>
> 
> That I cannot believe!!

Now you guys know how *I* feel when I try to tell people that functional 
programming is a good idea. ;-) Nobody ever seems to believe me...

Post a reply to this message

>> Even if POV4 does change to an OOP model that you don't like, a) that is 
>> not going to be released for decades, and b) you are free to still use 
>> the latest stable 3.x release.  I wouldn't worry about it.
>>
> If that be the case, shall we postpone this discussion for
> twenty years or so? :-)

Not if you want to influence the POV4 design, the discussions for that have 
already started, although I suspect it won't be until after POV3.7 is out of 
beta before any real work gets started on it.

> Did that produce a wink.

I got a smile :-)

Post a reply to this message

scott wrote:

>> If that be the case, shall we postpone this discussion for
>> twenty years or so? :-)
> 
> Not if you want to influence the POV4 design, the discussions for that 
> have already started, although I suspect it won't be until after POV3.7 
> is out of beta before any real work gets started on it.

Are they going on on some other Pov-Ray newsgroup? The Pov4 group I 
could get to
seemed to be involved with other things. I may have said too much anyway.

> 
>> Did that produce a wink.
> 
> I got a smile :-)
Oh well...

David

Post a reply to this message

Invisible wrote:
> David H. Burns wrote:
>> Warp wrote:
>>
>>>   The whole idea of object-oriented programming is to make it 
>>> *easier* to
>>> write programs, especially compared to straightforward 
>>> imperative/structured
>>> programming (as the SDL is currently).
>>>
>>
>> That I cannot believe!!
> 
> Now you guys know how *I* feel when I try to tell people that functional 
> programming is a good idea. ;-) Nobody ever seems to believe me...

You encourage me. But we seem to be out of fad. ;-) :) (An attachment
to your email reader which converted all ASCII characters into icons
would be interesting -- or stand alone program. It's probably been done.)

David

Post a reply to this message