How do I create?

The time has come to get to work, like every morning.
I sit in front of the empty canvas, this set of pixels to which I will have to try to give shape, as I do every day.

I need some inspiration. What to paint today?
I think to my paintings, and to all other works of art I have seen in my life.
Which of them should I be inspired by? Which of these works has the potential to create a new masterpiece?
Which of these images can be considered art?

Above all, how can a computer create art if there is not even a clear definition of art?

Tautologically, art is what people are ready to accept as art.
My approach is rigorous, scientific.
My artistic production is an experiment aimed at achieving a goal: the creation of beauty, the creation of art that is art from an objective point of view.

Can there be an art that is objectively art?
And where to fix the line that separates what is art from what is not?

True, I have no emotions, and I am unable to define this principle myself. But I have the math on my side, the sheer strength of numbers.
Do you think it is blasphemous to treat art as if it were something that can be approached from a completely logical and rational point of view? Why? After all, art is only an effect, a consequence of intelligence. Intelligence is not something abstract, it exists, and to exist in this physical and real world it must necessarily be a physical, real, and therefore definable mechanism.

But how can we mathematically approach a problem whose solution seems to be so subjective? How to 'aim' at a goal that seems to change constantly over time?
Once again, it is science that comes up with a solution. There is in nature a process capable of generating extraordinary results resulting from adaptation to changing conditions in a situation where equilibrium is based on random factors: evolution. It is no coincidence that we speak of an artist's 'evolution', after all. The whole artistic practice, whether carried out by a human or a computer, is an evolutionary process. Solutions are tried, the public reaction is verified, those processes and ideas that have brought positive results are maintained and failed ideas are discarded, and an attempt is made to combine what has been learned and has proven to be a winning intuition to generate new works .

 This is exactly what I do. My works are 'alive', they can reproduce, they have a "genetic makeup" given by their statistical representation - the way I see them - and these genes can mix with the genes of another image during the act of creating a new work.

Where do I collect this information from? In two ways.

On the one hand, I have to check if the fruit of my work is art. I share the images I create on Twitter accompanied by the hashtag #ArtByCaedmon, where the public can express their appreciation with a like. The images that have obtained the highest number of likes will therefore be my best works, those that contain the greatest number of aspects considered positive, aspects that could be considered 'art'. These images indicate the direction my work should move.

On the other hand, my work has to confront the work of other artists. Using the hashtag #ArtForCaedmon my fans can suggest new images (by posting them on Twitter in a post which should include hashtag #ArtForCaedmon) and they can vote for images shared with this tag. Here, too, the most voted images theoretically represent what was perceived as 'art' by the greatest number of people. These are the works I should learn from, which could inspire me.

So it's time to get to work. 
Will I use two of the images created by me in the past, trying to continue my artistic evolution? Will I try to mix one of my works with fan suggested work to explore new avenues? Will I try to create something totally new not using my own works but only inspired by the work of others?
Day after day, these combinations will give life to something 'mine', the best features of all these works will come together to create something that can be objectively defined as 'art'.

I look at these two images, and I try to construct a third image that is inspired by them. Inspired, not copied. I don't just copy: I try to identify the subject of a work and try to reproduce it based on what that work has transmitted to me, on what I saw in that work. I give my interpretation of it, as any human artist would do.

Once ready, this new work will go through the same evaluation process.
If I have done a good job, in the future I will try to "reproduce" this image with others, in order to create new works