How do you programmatically distinguish between a PROFESSIONAL fetus and a DILIETAN fetus?

 

This is more a question for professional programmers than for beginners. But beginners' opinions will also be interesting. We will not talk about traders, but about programmers who write Expert Advisors, indicators and scripts in Mql4.

Imagine that you have such a task:

You have a folder with 100,000 files in the mq4 format. You need to determine which of these files belong to the creations of professional programmers, and which belong to the brushes of dilettantes.
But it must not be done visually. The tester cannot be used either - it is too long and painful.
The task is to write a program in any language (for example: C+, Delfi, VB, etc.), which would itself analyze the texts of the Mq4 files and select from them those which are written by professional programmers.


Now the question is: what criteria must be followed by such a program?

To write a program, the criteria must be clearly formalized, there is no

- "la, la, truffle";
- "plus or minus a foot on my bare feet";
- "nine inches plus/minus a mad cucumber;
- "you can't do this kind of thing overnight, come back in a week."
- "the size of the salary will be good";
- "a flat in a tall house in the natsa district of Enns";
- "yes no can be likely, but it's possible";
- "God willing", etc. etc. etc.

_ will not do! The program analyses the text file. What should be in it, what shouldn't be, etc.?

The criteria must be simple and clear. Only clear, concise and comprehensible formalisation!
If I have clear criteria, I will order a free and available to all program - code quality analyzer.

So, what are the criteria? Please speak up. There's no rush, just think about it...

 

This is bullshit, Dima.

You downloaded a big archive with lots of EAs and you don't know where to start?

 
Mathemat:

This is bullshit, Dima.

You downloaded a big archive with lots of EAs and you don't know where to start?

No, I haven't. But, now we are busy with one of the programmers writing a program which will by itself browse the Internet (when the Internet traffic is not busy) and download mq4 files from different sites. And, what we have written above is the second phase. There will be a third.
 
DmitriyN:
No, I haven't downloaded it yet. But, now we are busy with one of the programmers writing a program which will by itself browse the Internet (when the Internet stream is not busy) and download mq4 files from different sites. And, what we have written above is the second phase. There will be a third.

Collect a list of pro's coordinates and check the code for coordinates from the list.


ZS: where no coordinates are found in a separate folder, then take a piece of code in the folder with coordinates and compare it with the code in the folder without coordinates.

Each pro should have a different style of match

ZZZY: the rest of it to manual overflow

 
DmitriyN:
No, I haven't downloaded it yet. But, now we are busy with one of the programmers writing a program which will by itself browse the Internet (when the Internet stream is not busy) and download mq4 files from different sites. And, what we have written above is the second phase. There will be a third.

The point is that the robot will do meaningless work...????
 
forte928:
The point of a robot doing meaningless work...????
It has to make sense, that's what the question is for.
 
DmitriyN:
It has to make sense, that's what the question is for.

It makes no sense to collect a bunch of empty information to dig through it looking for a grain of truth - the same thing is on Google but as they say ("I'm a robot") ask the right questions to get the right answers...
 
forte928:
No sense to gather a bunch of idle information that then rummage through it looking for a grain of truth - _

Almost everyone on this forum is (has been) doing it.

sanyooooook:

ZZZY: the remaining part in the manual overkill.

Nothing manual. Only automatic. 100% automatic. You can automatically select ten to twenty pieces out of 100,000, but no more.
 

Let him put in a multi-currency tester and MTF, trade, optics, build fans and read reports and magazines and search the bush on the internet.

And let him remember to draw conclusions. Pages or so.

 

A professional programmer's code will not always generate income (unless, of course, you sell it). It is also the other way round.

And non-professional code can bring income (when trading on the market). How to distinguish them - I don't know. To try each in the tester.All that drains in rubbish. Doesn't leak or doesn't leak much, to the judgment of a higher authority - a person

 
Dersu:

Let him put in a multi-currency tester and MTF, trade and read reports and magazines and search the bush on the internet. And let him remember to draw conclusions.

Testing is a very long process. Testing individual Expert Advisors sometimes even takes days. Therefore, it is not realistic to test hundreds of thousands of them.

Vinin:

The code of a professional programmer will not always bring profit (unless you sell it, of course). It is the other way round as well. And a non-professional code may bring you income (when trading on the market). I don't know how to distinguish them. To try each in a tester. All that plummet into the trash. Doesn't leak or doesn't leak much, to the judgment of a higher authority - the man.

I agree with you. But, the notion of a "professional trader" is even more blurred than that of a "professional programmer". On the other hand, the probability that an amateur will accidentally write a good EA or indicator is negligible.

Reason: