Aleph Null documentation
Jim Leftwich/Jim Andrews collab
“It is a thrill to see this fresh collaboration between two brilliant artists, Jim Leftwich and Jim Andrews using the most recent version of Aleph Null to generate recombinatory visuals that will surely astonish and delight everyone!”
Karl Jirgens

"aleph null is a tour-de-force piece of work—you can quote me"
Ted Warnell
4 Media types in long visual poem
The collaboration between Jim Leftwich and myself centers around an online, never-the-same-twice recombinatory/generative poem/animation that randomly composits about 240 of Jim's visual poems. The software doing the compositing is Aleph Null, a graphic synthesizer I wrote in JavaScript/HTML/CSS.
There are four types of media in the (long) visual poem:
  1. the never-the-same-twice generative animation in Aleph Null itself;
  2. slideshows of selected screenshots of Aleph Null chewing on Jim;
  3. videos of Aleph Null chewing on Jim;
  4. giclée prints of high-res renderings of the above process.
Selecting 240 Jim Leftwich visual poems
Jim let me select whatever of his visual poetry I wanted to use in Aleph Null. It was exciting to read/view thousands of Jim Leftwich's online images. I selected about 240 ranging temporally from the early 1990's to about 2019. The earlier work consists of shape poems (coherent sentences arranged in various shapes) done on a typewriter or with a word processor, before his work started to look as much visual art as poetry.
When I read/view Jim's work, I enjoy it as visual art amid its literary dimension. I go back and forth between reading visually and literarily until they are one. One of his poems simply says "Action Writing" in blocky, scrawled brushstrokes. Which, of course, is an echo of Jackson Pollock's 'action painting'. To some extent, Jim's work is abstract poetry. Although one might say the same thing of other less visual but distinctly disjunctive poets. Another of Jim's pieces says "language"; stapled over that piece of paper is a plastic, red net, like part of a net you buy, say, avacados or lemons in. Language is indeed a net we throw over reality to 'capture' it (though it usually rolls away). Another piece just has the word "juxtaposition" on it. By now an old word. Still useful. It's part of the language of "collage", for instance or, in poetry, Imagism. From early in the 20th century. Not much new can be done with collage, these days. Not with collage solely. But in combination with other things, possibly. Probably.
What I'm doing with Jim's work
What I'm doing with Jim's work is recombinatory/generative. It's recombinatory in that it's composed of combinations of Jim's works. It's generative in that it's not generating more Jim-like pieces but things, if you know his work, that you will only rarely mistake for Leftwich poems even though you'll recognize they're made of his work.
I'm using Leftwich poems sort of as 'letters' in a language. The circles/brushstrokes Aleph Null draws are 'words'. A screenful or print is a 'sentence' or 'paragraph' or 'stanza'. The recombinatory/generative never-the-same-twice Aleph Null animation itself, using Jim's work, is a 'poem'. The slideshows are more composed and focussed versions of the poem. The prints present much more detailed images than the images for the screen. The prints are more extensively readable.
The 240 Leftwich visual poems used in Aleph Null range from having been written in the 1990's to 2019. The earlier work, as shape-poetry, is closer to poemy poetry. The Andrews/Leftwich work is sometimes quite densely textual and a mixture of the 'asemic/pansemic' Leftwich material plus the earlier, more poemy texts. Part of what I wanted to create is work that is endlessly readable but also visually compelling/interesting.
About Jim Leftwich
Jim and his group are the originators of the term 'asemic writing'—which has blossomed into an international phenomenon. There are several publications and publishers publishing 'asemic writing' by a growing community of visual poets regularly posting on social media groups.
Jim says he now prefers the term 'pansemic'. To me, that does actually describe his own work better. It suggests that the meaning of the writing is all around, or pervasive, in some sense. We ourselves are the constructors of meaning and construct it, to some extent, in the presence of all symbols we find evocative, even if it's solely tone/atmosphere/feeling. Jim's work is rarely 'without meaning'—though it is frequently without the sort of meaning typically associated with poems or even visual poetry. That's part of what makes it interesting, as poetry, of course.
Visual artists might say they've seen Jim Leftwich's sort of semiotic configuration before in collage techniques and found materials; the cubists, dadaists, and their descendants. Art has reached the point where, given any contemporary image, you can say it looks like something else that's already been done. What's interestingly new is not interesting because it simply doesn't look like anything that we've ever seen.
It's a bit like W.S. Burroughs's situation. Burroughs said that he and Gysin were only doing what had already been done in visual art for 50 years, concerning their cut-ups. What he didn't mention was that although, yes, there had been people in visual art doing cut-up-like things, hardly anyone, if anyone else at all, developed the poetics into a whole philosophy of language and media as he did. There have been visual artists doing some of what Jim has been doing—early forms of asemic writing have been traced to Chinese calligraphers during the Tang Dynasty (c. 800 CE), including Zhang Xu, and Huaisu—but Jim has developed the poetics into an approach that has blossomed not only in his own extensive work but in the work of the many other people involved in 'asemic/pansemic' writing.
Leftwich is a serious poet but his techniques and methods also resonate with visual artists. That's a little unusual with visual poets; often, the visual dimension of their work is solely literary. The 'asemic/pansemic' work Jim has been doing is in relation with both visual art as poetry. Also, he's published many books of poetry. He's active as a curator and editor of the work of other poets. His site houses tens of thousands of his own images and also images generated via collaborations with a wide range of other poets and artists.
Previous work with Leftwich's images
Jim's response to seeing Aleph Null 3.1 chewing on his work was this:
"amazing, Jim
i am seeing my life's work 
mixed together 
into one large work
it is stunning for me
thank you very much"
That was very gratifying, given the respect I have for Jim and his work. This was not my first attempt at creating generative work with Jim's poetry. It's been a fascinating journey for me.
Before I wrote Aleph Null I wrote an earlier graphic synthesizer called dbCinema in Director. When you created a brush in dbCinema, you'd type in the concept, and the brush would do a Google image search on the concept and retrieve about a thousand images somehow related to the concept. And then mix them together somewhat randomly.
You see that the dbCinema images also use work from a wide range of other visual poets. That's because Jim is widely collaborative; his name is linked with many other poets. When you do a Google image search for Jim Leftwich you get lots of collaborators also.
The Google image search thing was great but then Google made it impossibly expensive to do a piece exposed to the general public—I would have had to pay (per search) for all the searching the public did with dbCinema.
So then, in 2011, I wrote version 1.0 of Aleph Null in JavaScript/HTML/CSS. Director was becoming obsolete. The HTML5 canvas, touted as a replacement for Flash , had just come on the scene. However, the first couple of versions of Aleph Null didn't deal with sampling from bitmaps; it was all about getting brushes to do their thing, and do their thing simply with color/gradients.
When I wrote version 3.0, around 2017, the central feature of which was its ability to sample from sets of bitmaps and mix them together in unusual ways, I again wanted to work with Jim Leftwich's visual poetry.
I felt that Jim's art/poetry would work well in Aleph Null cuz Jim is always tearing things up and putting them back together with new stuff in new arrangements. That's kind of what Aleph Null does. That's part of why bill bissett's visual poetry works so well in Aleph Null also. Additionally, I really enjoy the work of Leftwich and bissett in so many ways.
Poetics of Signal/Noise
If we take a step back to a larger picture, we see that maybe nothing new can be done in various types of visual art and poetry. To 'push the envelope', now, one way to compose is not at the level of the note or the brushstroke of paint, but combinatorially. In the most ambitious of my interactive audio work, you have a heap of sound/animation icons and can connect them interactively as layers or sequentially. The sounds are loops. Aleph Null is also combinatorial. The 'letters' are Leftwich works. There are 240 of them. The 'words' are the circles/brushstrokes Aleph Null draws/writes which are filled with digital collages of Jim's work. A 'sentence' is one screen of the composition. A poem is a viewing session of successive screens as Aleph Null generates the visual poem.
In version 3.0 of Aleph Null, I did quite a bit of work with Jim's visual poetry. But my conceptual grasp of the sort of work I was doing needed to be stronger. I hadn't come to grips with the poetics of signal/noise, which I've been formulating for a while now. The version 3.0 work was a little washed out and muddy. This is not uncommon in recombinatorial digital art. It's the norm. In version 3.1-4.0, the details of the image are more sharply defined. And more deeply layered.
I did Photoshop work on each of Jim's poems I used. Rather than using jpg's, I created png's where the background is removed—and the text/image is haloed, ie, there's a little distance between the text and the edge of the image—a halo surrounds the text—and this makes the images layer more legibly, makes them combine more eloquently—because part of the aspiration here is to create screens that are interestingly— even endlessly readable. Also, some of the poems have a slight drop shadow, or very slight 3d. Again, this serves to distinguish the poem from the other poems it's mixed with. Also, rather than using a circular brush that changes in opacity from opaque at the center to fully transparent at the edge of the circle, as I did in the 3.0 work, I used a brush that's fully opaque everywhere, and has a black outline to define/distinguish the circles.
Recombinatorial digital work is a 'noisy' process. In mixing different images together, you're creating a lot of noise. You're interupting/destroying the signal and hoping to combine images into new, interesting images. Usually what results is muddy, washed out, indistinct, busy and without focus or interesting compositional structure— which can still be nice if you're new to the game, cuz it can still look pretty interesting—but if you've been at it for a while, not so much. One senses that it's important not just to smush the images together in interesting ways but to also consider, simultaneously, measures that mitigate the smush, measures that create/recover signal.
This issue of the poetics of signal/noise is not limited to visual work. One time the poet Kedrick James asked me why cut ups of cut ups are usually not as interesting as cut ups. He was talking about work with language. I was flabbergasted that he'd asked me a question I'd thought about and actually knew the answer to. And happy that someone else was actually thinking about such things. We kind of bonded over that. In any case, he was talking about cut ups where, when you cut up the language, you cut it into basically phrases and randomly rearrange the phrases. That works pretty well. But when you cut that up, depending on the algorithm you're using, you can end up cutting it into chunks that are half the size of phrases, ie, the chunks are only one or two words long. Putting those sorts of chunks together randomly doesn't work as well. Because the semantic content of the chunks is too small. You just get word salad. It's almost like randomly concatenating single words. Randomly concatenating phrases, on the other hand, gives you more meaning. Again, one can think of it in terms of signal/noise. Cut ups of cut ups are noisier. Cut ups/phrases still contain a great deal of signal.
Jim Leftwich appears in the No New Ideas brush also: he and I had a good conversation about the idea that there are no ideas, so I nicked the conversation and put it in No New Ideas .

Jim Leftwich elsewhere on the web