Interesting article about AI

[440bx]

@VisualLab

If you can, stop the babbling and show some code that makes your point.

I put my code on the line.  No more lame excuses, do the same. 


@Joanna,

You want to look at one of the files ?... Easy!!  download the archive and look at any one file you want.  Take your pick.

Is downloading a compressed file an insurmountable challenge to you ?  (it isn't to most people)


Lastly, this stuff really isn't on topic.  But, you are behaving like an A.I bot, posting stuff that's besides the point while implicitly  claiming it satisfies the requirements.  VisualGPT ???

[VisualLab]

@VisualLab

If you can, stop the babbling and show some code that makes your point.

I put my code on the line.  No more lame excuses, do the same. 

Dear 440bx, I have explained to you why your request is nonsensical. You do not understand, too bad. I don't have the slightest desire to fulfill your whims. There are better and more pleasant ways to waste time.

Lastly, this stuff really isn't on topic.  But, you are behaving like an A.I bot, posting stuff that's besides the point while implicitly  claiming it satisfies the requirements.  VisualGPT ???

Yes, you're right, all this talk about "OOP religion" and "comparing two programs written as proc-struc and OOP" is off-topic. It's a pity you only realized it now. However, the explanation I wrote was necessary to show the uselessness of your statements (to put it mildly, because I don't want to inflame the "discussion").

[440bx]

I don't have the slightest desire to fulfill your whims.

I'm pleased you acknowledge being all talk and most definitely no code.  Just as expected.

I don't want to inflame the "discussion").

Showing some code that supports your points would not be inflammatory but that's way too much to expect.

[ackarwow]

(...)
I'm never got on the OOP bandwagon. I've seem some practical and maintainable use of OOP, but I've seem far more use of OOP that made things far more complicated and harder to maintain than they should have been. I don't reject it outright, because any programming paradigm can be abused and also used effectively.
(...)

Can we say that OOP is better/worse than structural or machine (assembly) programming? Generally - we can't, because the answer depends on what we are supposed to use a given approach/tool ​​for. That is, it depends on the needs. Generally we can't say that a hammer is worse/better than a chisel or a file.

[440bx]

Generally we can't say that a hammer is worse/better than a chisel or a file.

True but, one hammer can definitely be better than another. 

[VisualLab]

(...)
I'm never got on the OOP bandwagon. I've seem some practical and maintainable use of OOP, but I've seem far more use of OOP that made things far more complicated and harder to maintain than they should have been. I don't reject it outright, because any programming paradigm can be abused and also used effectively.
(...)

Can we say that OOP is better/worse than structural or machine (assembly) programming? Generally - we can't, because the answer depends on what we are supposed to use a given approach/tool ​​for. That is, it depends on the needs. Generally we can't say that a hammer is worse/better than a chisel or a file.

As you can see, you can try to argue with arguments. You can give a lot of them. And still the answer will be something like: "yes, but no". Why "no"? Because... "no". And that's it.

There may be programmers somewhere who treat OOP as sacred. Except that those using C# and Java have no choice (because everything there is an object). But on this forum we have the opposite example - an example of an "anti-OOP" religion.

[TRon]

@VisualLab:
Thank you for your elaborations on the AI part of this discussion (which seems a bit faded away now). Indeed it is not about deliberate ill intend rather the incentive behind it that makes some people blind for what they are actually doing (mostly in a rush in order to try proof the point).

And it is not about fear but concern. I can see in daily live that personal data is being used (without any consent) and be fed to models that the user that uses them has no idea om what data the used model was trained so that the outcome is yet again a roll of the dice. And this already happens in healthcare, legal system, economic reports etc.

It is already so that models are restrictive on purpose about certain topics because "society" demands such restrictions but that also poses a thread to the (actual) truth. So the outcome is biased by default (not to mention what data it is trained on which is per definition is also biased).

Whether or not you agree to that is another matter as old traditional research can be manipulated in a similar matter but at least that is done in a more controlled environment (there you need to know the parameters upfront to be able to manipulate the results so that these match the required outcome).

When you talk to people not knowing the technical details it is easy to spot the indifference on opinions because most people (including those that decide on the usage of such things) do not have a clue whatsoever and just go with the buzz. It saves time and personnel therefor money so it is gong to be used no matter what. It is scary without even touching the topic of f.e. how the energy peak will impact society).

[Joanna from IRC]

@VisualLab
If you can, stop the babbling and show some code that makes your point.
I put my code on the line.  No more lame excuses, do the same. 

What code? I don’t see any code from you and neither has anyone else. I think you just like to play games with people for fun or your code doesn’t exist.  Demanding that we rewrite 86 files for you is a ridiculous request. 

For the record I think those other languages which make everything a class are ridiculous. Fpc has gotten it right.

[VisualLab]

For the record I think those other languages which make everything a class are ridiculous. Fpc has gotten it right.

Not necessarily. If a new concept appears, its several implementations will usually differ from each other. Each implementer will have a slightly different approach to details. Some will treat such ideas quite loosely, while others will treat them as an untouchable foundation.

Although in my opinion, the requirement that everything be an object (even simple types such as integers or boolean values) is a bit exaggerated (or even: too drastic). As a result, it is impossible to write a procedure/function that is not part of a class. So programmers of languages ​​such as C# or Java create various classes (various "JUtils") grouping together such subroutines that in C++ or Object Pascal would be placed outside any class. To me, this is an example of a useful idea reduced to absurdity. But it doesn't seem to bother programmers of the above-mentioned languages. Some of them probably consider it completely normal (and in extreme cases there are those who consider it a better solution than loose functions).

This type of thinking is not an exception. Just remember that half a century ago the concept of "everything is a file" was being vigorously implemented. And when it wasn't, dummy or imitations of such files were created. To some extent, this allowed for the introduction of certain abstractions in the operation of the Unix system. Computers have changed (become more complicated) and the old idea that was useful back then is no longer sufficient today (to put it mildly). Linux and *BSD are still based on this. Half a century has passed (sic!) and many people treat it as sacred, as if it were an order from the IT God himself.

On the other end of the spectrum, there are "lovers" of Python who believe that OOP should be treated very, but really very loosely (also an exaggeration, but in the other direction). So there is no encapsulation. Besides, Python doesn't have many other important (useful) features, but that's a topic for a separate discussion.

For me, in technical fields it's really about not overdoing it and not going to extremes. Otherwise, living with such solutions will be a pain. Apart from a handful of steadfast believers (who strictly adhere to this strangely understood techno-kosherness), the rest of the people will struggle with an overly complicated or overly unfinished solution, in any case not adapted to real needs.

[440bx]

What code? I don’t see any code from you and neither has anyone else.

If you were capable of downloading an attachment you'd see plenty of code.  I'm so sorry to make things so hard for you but I'm one of those unreasonable people who expect programmers to be able to download a file, extract a file from an archive  and be able to write code.

In my favor, I should mention that I don't expect A.I to do those things competently either but, your inability gives a clue why A.I is replacing some people in the performance of some activities.  A.I unlike you (and "others") would have presented some code instead of never ending defensive babble.  At least A.I has that going for it.

[LV]

@440bx, I have reviewed your sources. Your work is impressive, and I have much to learn from it.

[440bx]

@440bx, I have reviewed your sources. Your work is impressive, and I have much to learn from it.

Thank you LV.  I appreciate the kind words.

[Joanna from IRC]

LV if you’ve seen his code would you mind sharing the best parts with us?
If you don’t I will not believe you’ve seen it.

Visual lab the everything being a file doesn’t seem like a good idea. Some things only need to exist only in memory at runtime. For instance I have classes  that store information about the database only for use while the program is running. 

440bx I wanted to see if I could help you with your request to convert your code to oop. I guess you’re not serious. I am no longer interested in your “secret code” . I will have a lot less patience in future for people playing games I think.

[Bogen85]

Visual lab the everything being a file doesn’t seem like a good idea. Some things only need to exist only in memory at runtime. For instance I have classes  that store information about the database only for use while the program is running. 

Yes, and in *nix everything in a file means how you say it needs to be. You access things as if they were a file, but they only exist in runtime in memory when you need them.

It is the way things on the computer are accessed, via file like interfaces, rather than API type interfaces via libraries. The file like interfaces are the API. They are not actual files in the sense of a text file that gets loaded into a text editor. What is being referred to here is what is being presented in *nix type systems in /proc, /sys, and /dev.

Linux, MacOS, and the BSDs still use /dev to access devices, Only Linux really relies on /sys and /proc, the BSDs don't really use /sys, /proc is optional.

As far as "not being a good idea"... Well, Linux won the for the cloud and most other Internet servers, and those use those file-like interfaces very efficiently. And MacOS since around 2000 is based on their own OS adoption of BSD. And Netflix uses FreeBSD for all their video streaming, because of the efficient performant way FreeBSD does networking. The "file like interfaces" are not a bottleneck in performance and they don't use excessive memory.

Everything being a file does make them hard to access from programs for many things, but as VisualLab pointed out, some like sticking with the past and how things were done decades ago. But the cumbersome nature (not the memory inefficiency) in using them is likely why MacOS and BSD moved away from them for /proc and /sys like how Linux uses those.

[440bx]

440bx I wanted to see if I could help you with your request to convert your code to oop.

Apparently honesty isn't one of your character traits.

As is painfully obvious, I neither need nor want your "help", nor do I need nor want a second-rate OOP version of my code.

You don't even know trivial things such as that advanced records don't initialize themselves and when/why they should be used. Never mind more advanced hardware and software knowledge.

I've answered your questions in the past because I enjoy sharing/giving knowledge but because of your blatant dishonesty, I will no longer share my knowledge with you.  Thank you for helping me stop wasting my time on you.

Additionally, you imply that @LV isn't being truthful.  Where do you get off making such completely unwarranted implication ? You owe him an apology.