My approach. The core is the engine. - page 20

 
Реter Konow:

Well, that's the kind of answer I assumed. However, why haven't you created a markup language? You've been doing graphics for a long time, and you haven't made a language in a weekend).

It's not relevant to me, and I have both compilers and interpreters implemented in assemblers under my belt. But I'm afraid that doesn't tell you anything.

 
Реter Konow:

As far as I understand, your windows use the standard graphics library (judging by its appearance).

How long do you think it would take you to create your own graphics library from scratch?

I use my own library, the base was made in about a month. And then slowly evolved as new needs arose. Note that new functionality was usually added in less than a day's work.

 
Yury Kulikov:

It's not relevant to me, and I have both compilers and interpreters implemented in assemblers under my belt. But I'm afraid that doesn't tell you anything.

I'm not trying to belittle your achievements (unlike you). It's just, - it's a different experience.

The first topic I started on the forum was about creating a visual studio on MT4. Strangely enough, the goal has not changed over the years.

No matter how cool your compilers and interpreters are, they don't solve the problems of algotrading.

I set a goal - to expand the capabilities of algotraders. Been walking towards that goal all these years.

And I no longer deny OOP. I have agreed that it is necessary and useful.

I just want to show what I have achieved with my approach.

 
Реter Konow:

I am not trying to belittle your achievements (unlike you). It's just, - it's a different experience.

The first thread I started on the forum was about creating a visual studio on MT4. Strangely enough, the goal has not changed over the years.

No matter how cool your compilers and interpreters are, they don't solve the problems of algotrading.

I set a goal - to expand the capabilities of algotraders. Been walking towards that goal all these years.

And I no longer deny OOP. I have agreed that it is necessary and useful.

I just want to show what I have achieved with my approach.

Here it is not clear in what way your development will solve the problem of algotrading? And what is the essence of this problem? I wrote earlier in this chapter that the most important thing for traders is to have profits. The question is how to take profit from the market using your methodology.

 

And so, it took Anatoly a year and a half to set up his library. (It only took Yuri Kulikov a month).

It took me three years to create my graphic environment. Created it completely from scratch. Using only my own codes. Without any help from outside.

Question: What is the difference between a graphics library and a markup language?

The difference is this:

The markup language lowers the required user level.


It is this property that enables mass distribution. The Visual Studio lowers the required level of user training even further.

To go from a graphical library to a markup language is a long and difficult road.

But I never created a library. I created Visual Studio from the beginning. And the markup language came about by accident).

The approach itself also came about by accident. It was created and forged by the need to solve the problem.

That is, my approach is the result of endless persistence and determination, regardless of any dogmas and standards (even if they are correct).

The approach absorbed only what was needed to develop the programme quickly.

And in three years, I created a markup language and engine with the approach. And also came close to creating Wiz.Studio.

Therefore, the effectiveness of the approach is beyond question. After all, it was created and polished by trying to solve an unrealistic task for one person.

 
Vitalii Ananev:

It is not clear how your development will solve the algotrading problem? And what is the essence of this problem? I have already written in this thread that the most important thing for traders is to make a profit. This brings me to my question - how to take profit from the market using your methodology.

The problem of algotrading is not about traders' profit. It's the enthusiasm for algotrading.

 
Vitalii Ananev:

It is not clear how your development will solve the algotrading problem? And what is the essence of the problem?

How about this. Here it is:

Retag Konow:

...I want to outline the scale of the problem on which I had to test the approach.

That is, you have to come up with some "scale" problem (exactly come up with it) and then heroically solve it for years to come:

Retag Kon ow:

No matter how cool your compilers and assembler interpreters are, they don't solve algotrading problems.

I set a goal of empowering algotraders. Went towards this goal all these years.

And it is not really important whether the problem exists in reality or only in the imagination. The main thing is to solve it, senselessly and ruthlessly for many years. Well, why not, if you have plenty of time and someone else brings home the food.

p.s. Sorry, Peter. You really are a good man, I don't want to offend you. But you just need some criticism from the outside. Made similar mistakes myself at one time or another.

 
Реter Konow:

There is one particular reason:

PROGRAM DEVELOPMENT.

....

So that new features can be added with a few lines of code.

My approach is superior to OOP in solving this particular problem.

Hmmm...

It would be interesting to see how you can get DEVELOPMENT "with a few lines of code" ?

With a few lines of code you can add a new window from what you already have. But, that's equivalent to me adding a TC League system, out of the 500-plus that I have. I, too, by adding just one line of code - I am adding a fully working TS, already debugged, tested on the history, and working for some time on the demo. But is it a "development"?

In my opinion, "development" - is, in my case, adding a new TS. For example, two months ago I added to two types of TS with entries by crossing of price and moving average and by touching of channel boundary the third type, entries by pending orders on zigzag tops.

In your case - I understand "development" as adding a new TYPE of windows or control. Something I'm not sure you can do adding a new type of control with just one line. Moreover - just adding new complex control - will add you a lot of headache, which would be very difficult to solve with your approach, while with OOP it would be much easier. Say, is there such a control as "grid" in your engine? Something like Excel table? With possibilities to click on buttons above columns or rows to sort? How much effort would it take to add such a control to your library?

 
Реter Konow:

The problem with algotrading is not the traders' profits. It's the passion for algotrading itself.

By the way, Peter, this topic is exactly what I call the "dramatisation of an idea". True, the dramatization is mostly based on conversations rather than vivid examples, but nevertheless, as you can see, the topic is in demand.

Moreover, if the proof is presented that your system allows to DEVELOP and HARDEN the product easier than when using OOP (recall about grid control) - it will be a very strong argument for the use of your library by users.

 
Wouldn't want to be disappointed after many years. You overestimate the importance of gui. Peter, if they made a console, I might use it myself (although I don't suffer without it either), but I don't care about gui.