Discussion of article "Automata-Based Programming as a New Approach to Creating Automated Trading Systems" - page 5

New comment
 
FAQ:  Wagon of emotions
I'll try to explain - Article: A loud title, accusation of local programmers in ignorance, claims of genius of the method, and finality of the solution. In fact, it is nonsense and primitivism. I don't want to consider the advantages and disadvantages of both branching methods (if/case) - enough has been said about it here. I would like to ask the author one question: What is the automatism of his proposed programming style (exactly the style)? In any case, all input conditions (states), relations (solution methods), and actions will have to be programmed manually from and to, and this system will be rigidly bound to them just like any other solution methods. That is, there is no flexibility of approach. And there is no possibility to change the reaction of the automaton on the fly.
 
FAQ:
I would like to ask the author one question ...
I'll make an assumption. The author of the article is not a programmer. Perhaps he has some memories of his former participation in some projects. But at this stage he does not program unambiguously. The content of the article shows that it is nothing more than reasoning. And the reasoning is thoroughly detached from practical experience. If the author was a programmer, this article would contain at least one example of an EA/indicator code on mql5. There is no example.
 
abolk:
I'll make an assumption. The author of the article is not a programmer. Perhaps he has some memories of his former participation in some projects. But at this stage he does not program unambiguously. You can see from the content of the article: it is nothing more than reasoning. If the author was a programmer, this article would contain at least one example of an EA/indicator code on mql5. There is no example.
Argument style Stop arguing about grain loader variants. Down with disputes around technical issues.

We are mastering a higher style of argument. Arguing without facts. Arguing on temperament. An argument that moves from unsubstantiated assertion to the personality of the partner.

What can a lame man say about the art of Herbert von Karajan? If you tell him he's lame, he'll be defeated.

What can a man who has not changed his passport argue about? What views on architecture can a man without a propiska express? Caught red-handed, he confesses and admits defeat.

And anyway, how can we be interested in the opinion of a bald man with a nose like that? Let him first fix his nose, grow his hair, and then speak out.

Behaviour in a dispute should be simple: do not listen to the interlocutor, but look at him or hum, looking into his eyes. In the most acute moment to ask for a document, to verify the propiska, to ask for a characteristic from the place of work, it is easy to switch to "you", to say: "And this is not your dog's mind business", and your partner will soften, as scalded.

Nowadays, when pests are destroyed by sterilising males, we have to raise the level of argument to abstract heights. Let's quibble about the collapse and rise of Hollywood without seeing a single film. Let's clash philosophers without reading their works. Let us argue about the taste of oysters and coconuts with those who have eaten them, to the point of hoarseness, to the point of fighting, perceiving the taste of food by ear, the colour by tooth, the stench by eye, imagining a film by its title, a painting by its surname, a country by the "Cinema Travel Club", the sharpness of opinions by the textbook.

Bringing products to the level of world standards, which no one has seen, we will develop all seven senses plus intuition to the limit, which successfully replaces information. Which I have to congratulate myself on. Come to the table, it's boiling!

M.M.Zhvanetsky.

 
I honestly expected at least a matrix method, or even something close to the I.I. ... and here it is. That's what I'm saying, now it's clear why Mars is now"Curiosity" and not"Curiosity".
 
FAQ:
I would like to ask the author one question: What is the automatism of his proposed style (exactly the style) of programming? In any case everything ... will have to be programmed manually. That is, there is no flexibility of approach. And there is no possibility to change the reaction of the automaton on the fly.

I'll try to answer instead of the author, since he hasn't appeared here.

"Automativity" comes from the term "finite automaton" and has nothing to do with the term "automatic".

And "finite automaton" is in turn a way of implementing algorithms, an architectural design. If you like, it is a robot, a device, an apparatus that executes your algorithm step by step. You can load any algorithm, but the executing structure remains the same. It can be implemented in any language, from Assembler to MQL, and even hardware (on chips),

so it has nothing to do with "programming style" either. It's not a programming style, it's a design (development) style.

FAQ:
I honestly expected at least a matrix method, or even something close to I.I.... and here it is ... That's what I'm saying, now it's clear why Mars is now "Curiosity" and not "Curiosity".

You shouldn't be so dismissive. Automatic programming is a powerful, flexible and reliable method widely used in programming industrial computers and controllers (PLCs). Another thing is that there is no novelty and genius in it.

And by the way, 40 years before Curiosity there was Lunokhod-1. And rockets in Soviet times were controlled by finite automata :).

 
I agree with you on one thing, I was wrong to call it programming. Yes, more like design. Maybe it is good for highly specialised systems with rigid algorithms, but to exalt it as a new word? And don't tell me about "Lunokhod" and how spaceships are travelling .... I was there, as they say, "I was there, I drank honey-beer." Well, of course not specifically in those times, but the end of 80 captured, and what was there and how I know not from hearsay. My thesis about "Curiosity" refers to the fact that domestic technologies are firmly stuck in those times. And about PLMs and so on, and down through the architecture too. Not by hearsay I'm familiar with it. And now take any computer ( imported) starting from 2000 and calmly re-fill it with bios, and think about our advanced "design style" .... It's the same, or as they say in our country, "otozh, bo kume".
 
It is a very interesting approach to designing an EA. Reminded me of my university years and several courses on Mealy/Moore machines. I see how it can be used in expert advisors, but I doubt it will become "base standard for all traders."
 
Good article. I use state machines and this approach as a design pattern in some of my EAs since a long time ago, and probably many other systems developers and engineers.
 
MetaQuotes:

Published article Automata programming as a new way to create automatic trading systems:

Author: SO_ CHE NULLA NON SO

Automata passed in the university. I program complex tasks only in this way. Nothing new and revolutionary in this is not.

It is really a very good method. It allows you to keep even a large number of branches under control and find errors very quickly. It is especially useful for 4 because there is no OOP there.

And, of course, it is necessary to do design work beforehand. Which, in any case, is useful. This approach allows you to see and process those branches that are not always visible when designing "at a glance".

 

This article is great. You explain the model in a very simple and clear way;

I'm new to Metatrader and MQL5 but highly experienced in software development, C, C++, OOP and State Machines and Process Management as enivid and figurelli said,  and in fact I'm applying these in my EAs as it is a very good way to manage trading rules.

I do not know if this will become a standard as so many times a great product fails, but I'm sure it could.

1234567
New comment