Conrad Dreyer

Otl Aicher A-Z + AI

“It took the philosophy behind grids to open my eyes to just how greatly Aicher’s life’s work is still providing stimuli to the world of today.”

Basic questions

Conrad Dreyer

Taking Aicher’s slant on things and perception of himself as a designer (“I’m no corporate decorator”) as the basis for publishing a guide for young designers that poses the following questions:

  • How do I present myself to customers?
  • How can I externalise my expertise?
  • How can I supplant Aicher’s attitude into the present context of fast-moving, cheap design and make my aspirations count?
  • How do I get customers to regard me as an equal?

A question of grids

This soon led to the idea of devising a kind of branding concept that would simultaneously answer the above questions as to how to arrive at a good product. But that’s already been done, it is an incredibly involved undertaking and compiling such a work would exceed the remit of the course and, indeed, of my studies as a whole.

Thereafter, ideas materialised as to whether to create an interactive website with facts from Aicher’s life but, to be honest, reading Wikipedia would be enough there. This did, however, give rise to a thought worth pursuing: a brief portrait of Otl Aicher that maps the key points in his life anecdotally. The brief portrait ought to take the form of a booklet arranged alphabetically: a kind of mini-encyclopaedia on Aicher, as it were, with just one item per letter. That’s how Aicher A-Z came about.

I briefly entertained the notion of issuing entries as cards that could be randomly shuffled and serve as a source of inspiration. I decided in the end, however, that a brief joined-up portrait was what was needed: one or two weighty sentences in each case on the word entered for a given letter. I also had the idea of issuing each letter as a zine with several entries, but that didn’t have a sufficiently joined-up, compact feel about it when dealing with Aicher either

It proved particularly difficult to find words for the letters J and Q, but I’m very satisfied with what I finally came up with: “Jahre” (years), as a means of rendering Aicher’s life on a year-to-year basis, and “Quadrat” (square), the shape underpinning his pictogram grid. I have to admit, though, that the relevant words only ever came to me at the very last moment. Each letter was also to be accompanied by images.

To this end, I initially thought of using Illustrator Adjustments to alter generally available pictorial material in such a way that it would no longer be subject to copyright. This idea soon proved to be very boring, however, and not particularly innovative or visionary. It was simply too paltry for a work on a man of Aicher’s calibre.

In search of alternatives I hit upon the idea of using grids to generate images. But which grid should one use for this task? Should each letter be accorded a grid of its own? I eventually reached the point of subjecting the concept of the grid to philosophical re-assessment. What is a grid anyway? How is it possible to “fall through the grid” as a German saying goes that could perhaps be rendered as “not making the grade”? Is a grid not merely a means of interpreting specified forms of content?

Results of the AI Midjourney for various queries: for each search term, a total of four drafts are proposed, one of which can be selected for further processing (1).
© Conrad Dreyer

Results of the AI Midjourney for various queries: for each search term, a total of four drafts are proposed, one of which can be selected for further processing (2).
© Conrad Dreyer

Results of the AI Midjourney for various queries: for each search term, a total of four drafts are proposed, one of which can be selected for further processing (3).
© Conrad Dreyer

Results of the AI Midjourney for various queries: for each search term, a total of four drafts are proposed, one of which can be selected for further processing (4).
© Conrad Dreyer

AI – grid for the future

I had a brainwave following this last question: artificial intelligences are fed with data from the real world and go on to interpret their users’ inquiry on the base of how they have been trained. An AI exploits the knowledge in its stock of training data as a grid with which to interpret its tasks. An AI maps grids or, rather, is the grid for the present and – given further broadening of the technology – also for the future!

There seemed to be a case, therefore, for getting an AI capable of viewing the entire web as its stock of training data to generate the pictorial material. Luckily, “Midjourney” is just such an AI that can be comfortably operated with the Discord bot. Discord bots function to a never-changing pattern: the AI can be assigned to imagine an image using the /imagine command. The values thereupon entered in the prompt are then interpreted and the AI ideally generates an applicable image. By refining one’s enquiry and selecting good draft ideas by iterative means, it is possible to obtain images on almost any topic.

I discovered, however, that some terms are prohibited. I was, for instance, unable to generate images relating to the letter “N” and the words “National Socialism”. I accordingly designed that page in the booklet differently. My enquiries did not always yield what I had expected. But some of the results were so arty and outlandish that I set myself the restriction of only conducting a maximum of one iteration or other enquiry per letter.

I was not intent, after all, on imposing my own view of things upon the AI but solely upon having my questions re-interpreted through the grid of digitised human knowledge. The enquiries and images are set out below. Important note: Midjourney always issues 4 proposals one of which can be selected for further iteration. Applicable links lead to images intended to serve the AI as inspiration.

The booklet "Otl Aicher A-Z + KI" invites you to leaf through it.
© Conrad Dreyer