Pozapola, międzyświaty i (dy)fuzje
I created this project for my—let’s proudly call it this way—IT classes in spring 2023. It was an introductory exercise into the basics of web design (alright, the very basics of web design).
My page was experimental in its nature since the very beginning. I used a free ready theme, Pixl, created by Automattic, and played mostly with the layout. It was to be about experiments in the field, in the media of literature, and some crossrelations with different media and communication systems, where the visual layer and its role are exposed.
Though it might sound ambitious—and is ambitious in fact—it didn’t turn out to be much of so. The topics of posts work more as a to remember list of ones I would like to pursue my research in rather than a trustworthy provider of reliable knowledge. As I planned it to be experimental since the very beginning, I made the content using ChatGPT. Of course have I known how much most of the answers they give are far away from what the current state of research is. Of course have I known how construction of the answers often sounds, by which I mean—like if said by an always-keeping-it-positive business coach, who equals happiness and wellbeing with not letting one ever feel sadness or hesitation. Yet I decided to use the answers ChatGPT generated to my prompts (which were really noncomplicated and rather not pointing into any direction, but it also was planned).
I thought texts generated by machine, basically on the simplest possible prompts, would also be a good idea that would quite fit into the subject of my webpage. Machine-generated text seems to be related to the subject, as the phenomena described in there is often influenced by the changes in technology. (But the machine-generated text, is it literature? Or does literature need a human factor in its creation? But is machine-generated text really devoid of such factor, or is the factor not as much visible in the process as it is when one writes a text wholly by themselves?).
Of course, the fact the content was generated by ChatGPT means there is a lot of substantive mistakes (and language mistakes, too). I decided to not correct them, as it leaves the content in its ‘virgin state’, thus making it more authentic—and, in my opinion, interesting.
Besides posts, there is a page with poetry ChatGPT generated with my prompts. There I tried to teach it some sort of avant-garde techniques and hoped after a long streak of explanations and asking if and how they understood what I wrote, it would finally be able to generate some non–stencil-like pieces of artistic literature. The results varied in quality, yet I think for this moment are quite satisfying.
On the margins, there is some random text generated by, well, Random Text Generator. It is in Chinese and Japanese and works pretty much as if was a piece of asemic writing. Since the page was created in Polish and addressed to Polish audience, who I assumed may know English, German and maybe Russian, French, and Spanish at some level, I thought text written in a language little of Poles would understand and in a system little of Poles would be able to decode, Chinese and Japanese would be the best choice. Why Chinese and Japanese, not Korean, Georgian, Greek, Arabic, Bengali or other languages that use neither Latin nor Cyrillic systems? Well, CN and JP have their glyphs kind of square-shaped and dense with lines, which made these writing systems good as decorative element, too. And in asemic writing, as far as I know, the pretty-visual-and-nothing-beyond aspect of writing plays a crucial role, so the systems and languages I chose seemed to play similar role for not acquainted with them Polish people.
