Hacker News new | past | comments | ask | show | jobs | submit login
Generative Art Finds Its Prodigy (artnome.com)
340 points by trueduke on Sept 10, 2018 | hide | past | favorite | 106 comments

I like Manolo Naon's art, and I am quite fond of creative coding, but something that strikes me when seeing this article comparing his works to Kandinsky, Ernst et. al is that his works are always harmonious, a single idea in action, whereas all these classic artists have conflicts and discord in their works.

They deliberately avoided doing things that were simply decorative or pleasant. (Probably they were avoiding accusations of just making couch and wallpaper patterns)

There is more to experience then just one decorative process at work.

You can feel the artist's presence and discord in the images. Look at the Max Ernst works and ask yourself why he left it imperfect, fading away like dreams or unconscious processes.

Look at the Kandinskys. There is far more information (different design processes juxtaposed), and even after looking at it for a while you still have to wonder how it is balanced and yet there's barely any repetition.

Manolo is always explicit and readable. There is a lack of mystery. Once you've examined the image then you like it, but nothing stirs in the depths.

But he's young, I like the direction, and he will definitely keep developing.

I've experienced what you're talking about as maybe the fundamental challenge in making computer-based art. Computers love to follow rules. Come up with some interesting rules, and they will spit out interesting pictures/music. It's much much harder to figure out how to tell the computer "okay, but break the rule just a little bit, at the appropriate time". Codifying breaking the rules as its own rule is...well, so far it's mostly eluded me.

Definitely check this artist out: https://twitter.com/quasimondo/status/1035626920090718209

"If I let the machine decide which images to keep, I must admit that I am probably not ready yet for its sense of aesthetics and originality and my heart bleeds seeing which ones it discarded."

Check out "glitch art" and also this: https://giphy.com/explore/tverd

It is debatable if any rules are really broken, but using glitches is at least an attempt.

There's also a pretty active subreddit at https://www.reddit.com/r/glitch_art if you like these sorts of images. I think most of the time you end up with something that's cool, but mostly just "wallpaper". Occasionally you'll get something really cool and evocative that can stand on its own.

I don't think it's really about computers. These 20th Century artists were deeply involved with systems, rationality, the rise of machines, futurism, the jarring new perceptions that technology was hurling everybody into.

The quaint criticisms people are voicing today about generative art (as if this was new) are the exact same ones everybody was having in the early part of the 20th century.

Authorship lol: https://d7hftxdivxxvm.cloudfront.net/?resize_to=width&src=ht...

I feel like that's an interesting balance in generative art: I want to make something that is aesthetically pleasing or at least interesting to look at, but I also want to be surprised. If the process is really heavily stochastic, then you'll have to do a lot of manual curation to produce the final work. If it's heavily rules-based, then you might not find anything novel after the first few images.

Maybe there's an opportunity to combine computers and hand-made choices.

Art using a mixture of several generative tools and manual tweaks by the artist on the final result could achieve that goal of "breaking the rules just a little bit", especially if the generative tools are parametrical and easy to customize, maybe with direct manipulation on the generated canvas.

If you're interested in this. Combining generated art with manual tweaks. Look into http://gif.com.ai

You're in for a pleasant surprise.

I think you can seperate these two aspects of Kandinsky or Ernst's works. On one side the skill and aesthetics of the artwork, and on the other side the reflection of their times and circumstances.

We've established that Naon, Kandinsky and Ernst make beautiful artworks. So it makes sense to compare them on their aesthetics if they have similarities.

When it comes to the 'depth' you are speaking of, the modernists are pushing back against their environments and their history, struggling and competing with their contemporaries. This makes the art 'live' and gives it uniqueness. Never again can someone have the exact same circumstances and history of Kandinsky, those times are simply over.

Naon has his own history and his own circumstances. Maybe now they may seem shallow, but time will tell if his time and experience was unique, and his artwork a reflection of it. As you say he is young, maybe there is still more inside of him to grow and come out. But I have to say I have not seen generative art that good before.

If nothing else, it might be the best we've got to challenge the next generation of generative AI art with.

We can already suggest a simple interpretation: generative art is more of a craft tradition akin to illustration and what is traditionally called the "decorative" or "applied" arts.

One unpopular litmus test for whether a practice might be appropriately categorized as "fine art" or "contemporary art" is to look at the economic relationships and institutions involved. "Art" with a capital A tends to be supported exclusively by wealthy patrons, with some help from governments, foundations, and non-profits. Artists sell their work through galleries, or directly to their patrons, and as they age they either succeed or fail to enter a "mid-career artist" phase, where their work becomes canonized through reviews in particular publications, inclusion in group shows, and solo shows at "important" institutions.

Anything outside of this very specific system is difficult to call "Art" with a capital A.

It's worth noting that this mode of production is overwhelmingly minor in comparison with modes of production post-Internet. Memes, 4chan, YouTube, DeviantArt, art produced in conjunction with the advertising industry, etc. completely eclipse "Art" today, and it's difficult to speak authoritatively of any kind of canon of "art" in general.

However, it's clear that when we compare the "art" of today to the "Art" of the past that we see in the great museums of the past, we are comparing fundamentally different things.

"If works of art were judged democratically--that is, according to how many people like them--kitsch would easily defeat all its competitors" -Thomas Kulka


My thoughts exactly. These "imperfections" are marks of individuality. My prediction is that the next step would be introducing these algorithmically in a way that we find acceptable. For now, Manolo's art seems nice, but very far from the works of famous futurists, surrealists of Bauhaus painters.

It seems that recently there's been interest in procedurally-generated imagery among artists, but almost all completely unaware that the demoscene has been doing the same (and much advanced its techniques) for many years:



Demoscene demos are nice, but to me they feel more like expressions of technical prowess rather than aesthetic art. I don't think I've ever seen a demo where the result transcends the algorithm, and I don't think that's the goal of the demoscene in any case. They want to showcase how much technical brilliance they can fit in a couple of KB's.

That's different from this example, where the goal is more like how much expression (emotion/aesthetics) can I put into a single image. And I must say I agree with the author that his work is astounding in this regard. Each one of them feels like a proper artwork, not just someone that's fiddling with a couple algorithms.

You're right that many demos simply exist to showcase technical brilliance, but many in the demoscene are starting to trend further towards expressing artistic ability. Some examples off the top of my head:

- Number One/Another One by CNCD/FLT: http://www.pouet.net/prod.php?which=77399

- When Silence Dims the Stars Above by Conspiracy: http://www.pouet.net/prod.php?which=75713

- Aurora by Excess: http://www.pouet.net/prod.php?which=75791

- Along for the Ride by Ümlaüt Design: http://www.pouet.net/prod.php?which=78053

Also any ASD demo. They tend to stray away from the general "demo-esthetic" even when it come to the music.

"Lifeforce" was fantastic and really stand out from any other demo, event to this day.

"For your love" use the same kind of particles extravaganza that Fairlight popularized, but really has it's own esthetic.

Smash smashing it again with the first one...

BTW - your realtime synth engine (https://github.com/monadgroup/axiom) looks great. Any ETA for a Mac build of the VST?

I have a super-beta build working, if you'd like to try it out/help bugfix shoot me an email at "copodt at the google email" and I'll send it to you :)

I think Manolo's work looks great, but I also think a lot of demoscene work is an amazing, highly underrated mix of engineering and art.

Like this piece - https://vimeo.com/25284552

It's funny how a lot of the times the perception of art is that it should look like paintings we've seen in the past. In Manolo's case, paintings of Kandinsky. In a lot of "generative AI art" it gets the Art label because of generative brushstrokes/abstractions that are learned from paintings.

In my opinion art has nothing in particular to do with painting.

They want to showcase how much technical brilliance they can fit in a couple of KB's.

You need to look beyond the tiny size-limited ("intro") categories, there are PC demos which are multiple megabytes and they certainly have more of a traditional artistic feel to them, because things like sampled images can then be used easily.

> I don't think I've ever seen a demo where the result transcends the algorithm, and I don't think that's the goal of the demoscene in any case.

Sturgeon's Law: 90% of everything is crap. "Everything" includes all of art, and all sub-genres of art.

You might not have looked hard enough. If you are interested in finding the good ones, you should spend some time poking around; there are definitely demos with beautiful results, from people trying to make art, that aren't about the algorithm.

These two popped into my head, but there are bunches of great examples.



There is a abstract beauty in the implementation of the algorithms especially under the various constraints. I believe these constraints inspire creativity that those mildly practiced in or familiar with these arts can appreciate as an aesthetic or beauty all on their own, separate from and in addition to the audio/visual component.

Well, artists have been making generative art for a bit longer than that. Manfred Mohr was making stuff in the late 60's.

That said, I kinda agree with your point. A lot of the time, it's through ignorance, sometimes, it's strategy, and sometimes, you have people who are genuinely interested and influenced by stuff like this - but it would never come out in an interview, since the interviewer is inevitably coming from the art world.

Another fellow who's been creating generative images for a long time, as part of the fine-arts world, is Jean-Pierre Hébert: http://jeanpierrehebert.com/index.html Had a grant from the Krasner/Pollock Foundation at one point, I believe.

He's got some great pieces displayed there.

And even going back to the 50s with plotters and mainframes -- this book has a lot of cool examples: https://www.ucpress.edu/book/9780520268388/mainframe-experim...

If you want to understand art since Cezanne, and especially after 1945, you need to understand the modernist idea of the meta-narrative. It works like this -- prior to the loosening up of gesture in painting, the narrative was whatever was depicted. Maybe it was Washington Crossing the Delaware, or Napoleon Crossing the Alps. Pictures told stories directly. Beginning with impressionist painting, and following through to non-objective modernist paint, pictures told a second story -- a meta-narrative. This is the narrative of the events that created the painting. It answers the question -- what is the story of this artwork's creation? This leaves the viewer to deal with things like struggle, imperfection, and arbitrary residue (there are cigarette butts within some Jackson Pollocks). Did you ever visit MoMa and see Barnett Newman's Vir Heroicus Sublimimis?


It could be a generative work of art -- but wait -- the vertical stripes where the masking tape was peeled away leave small amount of paint bleeding into the stripes. Why? It tells you something about how it was made. Cold geometric forms made with a human touch. Visit the Joseph Albers, Piet Mondrian and Ad Reinhardt paintings and there's arbitrariness and human touch in them all.

So, I would ask you -- what is the meta-narrative of this generative art? What is the story behind it? Does it inspire you? Do you feel like you are looking at art? Or a modern art-themed screensaver?

This feels too prescriptive. Can you really create such a sweeping narrative (of the meta-narrative) encompassing the entire world of "art" over the past 75 years?

The meta-narrative is great but it's not everything. Personally I'm not very inspired by the Newman piece, and the bit about the masking tape makes it only marginally more interesting to me - it doesn't necessarily tell a story, it just reveals the technique.

Here is a piece of generative art I made [1]. The story is that I saw these tiles on the clearance rack in Ikea, and I figured they could make some nice patterns, and when I laid them out I realized there were many good ones and I couldn't choose just one, so I made a program to simulate changing patterns. However, none of that comes through in the work itself, there aren't really imperfections or human touches, it's based entirely on simple rules. You might easily dump it into the screensaver category, but I find the endless patterns that emerge out of such a system to be compelling. Would be very grateful if you gave it a look and shared your thoughts.

[1] http://jamesrowen.me/tiles/

I actually prefer your art to the art shown in the article. But, I would not call your artwork generative. However, it is programatic and animated. And could be easily created by the ideas of op art. https://en.wikipedia.org/wiki/Op_art Op art, like your art, is created out of a fascination for illusory space and patterns. It's a footnote to modernist and geometric abstract painting.

The background you provide on the piece informs it well. And a viewer could deduce the piece was made by a fascination with simple formal relationships. This to me is the backstory, and the meta-narrative.

Thanks for the comments. My understanding was that generative was synonymous with algorithmic/programmatic. Was not aware of op art but I certainly see the connection.

I played with this a little more. And, it likely is more generative than my first impression. Especially the Spiral example that seems to transform the composition almost autonomously. Not sure how serious you are about it, but I think it's definitely some of the better programatic art I have seen, speaking from a fine art/academic perspective here. It's very reductive and simple, and deals with some very basic relationships in visually powerful ways. Best of luck. Good share!

The spiral mode was discovered entirely by accident, when I introduced a bug into the translation/rotation code. What you see when first entering the mode (or using the o6 preset) is pretty much exactly what I saw then. I sat there giggling uncontrollably; it was so bizarre and unexpected, yet compelling. In that sense I would agree that it is the most generative mode. From there I continued developing the tools to modulate and navigate it to find other interesting parts. I realized it was becoming a kind of fractal viewer.

I don't have a formal background in art, but based on what I have seen in this area, I do think the project is somewhat novel and worthy of interest, and I have ideas about how it could be adapted into an interactive gallery installation (each mode projected onto a different wall, each with a custom physical control panel). I have hesitated to share it so far because of a few issues on the technical side (untenable framerate on slower machines, UI is broken on mobile). I'm also not sure the UX is dialed, I want it to be minimal and somewhat mysterious, inviting the user to play with the controls, but not so cryptic that they give up before seeing what it can do, hence the somewhat inelegant tutorial bubbles (I started with a simple pulsing glow on certain buttons but people didn't even notice it).

If you have any suggestions or thoughts on these matters I would be all ears :) jamesrowen@gmail.com

I’m not an art history major, but why does a work of art require a backstory? Or a meta-story? Why is a screensaver not “real” art? It seems so limiting. That’s a beautiful painting and the painter is very talented, but there is no existential dread evident in the brush strokes, no social commentary on the intersection between class warfare and feminism, so it’s not “real art”. Who makes these rules?

It’s like when you hear a young pianist play Liszt perfectly or a Rachmaninov piece better than Rachmaninov did, and the critics say, “Well there was no emotion there, no struggle—she has a long way to go.” Bullshit! It was great! Why can’t art just be appreciated itself, without all this “story” baggage?

Of course, if you like it, that's really all that matters. But art history, and contemporary art, are not really determined by polling the audience for what's liked most. Instead it's determined by those who have made it their passion to understand what makes great art and drives great artists. And they are able to agree on what makes it into art history (at least the Anglo American version) pretty well, for better or worse.

Why can’t art just be appreciated itself, without all this “story” baggage?

I think art expresses something inside a person. Sometimes, people find it meaningful on its own. Sometimes they don't and understanding can be enhanced by "story" or whatever.

Kind of a case of sometimes words get in the way and sometimes words help bridge the gap between two minds.

/random internet stranger who didn't actually read the article, just your comment

> Why can’t art just be appreciated itself, without all this “story” baggage?

Because that eliminates the in-group, experts, elite, etc. Hence all the "rules" and terms like "outsider art", etc

You can similarly ask, 'how could art exist without "in-group, experts, elite, etc."'? Why in the world would anyone create all these bizarre objects?

Another interesting artist in this genre:



Shameless plug: Inspired by "inconvergent" I have built a plotter and started working on my own generative art realised with pen and paper: https://www.instagram.com/framedbyher/

Where did you get the plotter?

I have built it myself. I was inspired by a couple of tutorials I found on the internet. Most of the components are off the shelf, stepper motors, smooth rods, belt, arduino, etc. The components that hold everything together are either 3d-printed or laser cut. The software that is running on the arduino that translates gcode -> stepper motor impulses is slightly modified grbl [0]

I have a (poor) timelapse video of the assembly process - [1]

[0] https://github.com/grbl/grbl

[1] https://framedbyher.com/the-story/

Inconvergent is my favorite as well. The artist in the article uses Processing; inconvergent builds his own tools which I think is cooler: https://github.com/inconvergent/snek

For anyone interested in the history of generative art and its parent discipline "(new) media art", some historical resources:

- http://generative.net/read/home

- http://www.medienkunstnetz.de/mediaartnet/

- http://userwww.sfsu.edu/swilson/book/infoartsbook.html

- http://archive.aec.at/, and https://s3.amazonaws.com/arena-attachments/139740/8aec5cb9da...

- http://dada.compart-bremen.de/

- http://turbulence.org/

- http://neural.it/ (magazine)

- http://rhizome.org/art/artbase/

- http://www.nettime.org/ (classic net.art mailing list)

- http://we-make-money-not-art.com/ (previous-generation blog)

- http://www.creativeapplications.net/ (current generation blog)

Referring to Naon as a "prodigy" is a deeply out-of-context claim.

http://archive.aec.at/prix/ all the way back to 1987. Naon is a tiny blip in an ocean of creativity and ideas.

- http://recodeproject.com/ (reinterpretations of early generative art)

I think Naon is very distinctive for his embracement of the importance of human touch & influence in generative art. His choice of color, his adjustment of the shape and form in the work (either intentional or accidental) all produce works that have a human element to them even as they are generated by a Processing script. Automation is a tool and a big element of the work, but the vision of the artist and his creation plays a big part in what the end result looks like.

I've often wondered - and maybe this is an actively explored area of art - what it would look like to attempt to remove all human impact from the process of generative art. A human will write the code, so maybe that's impossible, but how far divorced from the influence of the coder can the resulting work be? Do we just end up at random noise, or is it still possible to make something pleasing without a conscious guiding hand? Distribution-mimicking methods like GANs are very "hands-off" like this in principle, but of course are influenced by the training data. Interested to see how generative art continues to play out in the future.

Joshua Davis [0] has been doing generative art (mostly Flash/actionscript) since late 90’s. Took a class of his once and it was amazing (as is his art).

If OP subject is prodigy, Davis is the grandfather.


Came here to say this

In the same tradition: http://mariuswatz.com/

Generative art is a fascinating space - it's in its infancy for sure - I've been getting into it this past year and wrote my first algorithm about a year ago:


Like all art, it comes down to who witnesses it, and how it makes them feel. However - there's a definite gap between seeing something in real life versus digitally which I'd like to see Naon bridge. There's a large community of plotters which are doing a good job of bringing generative art into the real world. Definitely worth checking out the twitter hashtag:


Naon cranks out amazing compositions at an astounding rate, it makes me real happy to see him get some good exposure like this.

One of the cofounders of Etsy, Jared Tarbell, is also a great artist and worth checking out: http://www.complexification.net/gallery/

If you'd like to see more generative works, the subreddit is a good starting place: https://www.reddit.com/r/generative/

Concentric looks really great! It actually has decent variation

Apparently, he published all of his code (I didn't see any mention of where). Regarding the color -- he mentions being careful with the palette, but I wonder how he gets it to turn out correctly (assuming it's RNG). It seems like you would have to keep generating new instances until you find one that looks pleasing.

One algorithm that comes to mind:

  1. Choose a color palette and initial resolution.
  2. Implement generative features at current resolution.
  3. Generate the output and until it "looks right".
  4. Increase feature resolution.
  5. Go to step 2.
In other words, one could keep generating prints until the locations and extents of the sampled colors "looks right", then keep the random seed and continue implementing finer details until they "look right", etc.

This will be the second time I drop this on HN this week, but check out Primitive Pictures[0] if you are interested in this kind of stuff. Also, I highly recommend Schiffman's YouTube channel[1] which goes into depth on how to get started with processing (and p5.js, the browser/canvas version). Note that the artist in the article learned how to do this by reading Schiffman's book, which is covered in a series of tutorials in the linked YouTube channel.

[0] https://github.com/fogleman/primitive

[1] https://www.youtube.com/channel/UCvjgXvBlbQiydffZU7m1_aw

I don't think this algorithm is very common. In particular, the colour palette can usually be chosen last, and a lot of generative art (at least all that shown in the article) will not need multiple resolution passes. I think a more common discovery/refinement process is to go from more random to less random, or random in particular kinds of ways. For example, you can specify the values or distributions (or at least ranges of values) for particular parameters based on things you think will look good or which should end up similar in ways you like to things you've found already.

Generative art is a search in an extremely high dimensional space. The more you go into it aiming for something in particular, the more nailed down it is and the closer to traditional art (and the harder to pull off well IMHO, think how much difference slightly changing the position of one of the elements in these images would make, versus slightly changing the position of someone's eye). If you go looking for "new and interesting" or "looks good" without really aiming for anything in particular, it can be a lot easier to find good things.

I've seen a couple of really interesting talks recently about the surprising successfulness of objective-less search by Ken Stanley:

https://www.youtube.com/watch?v=dXQPL9GooyI (40 min.)

https://www.youtube.com/watch?v=tZBViI8ZaU0 (90 min.)

He started off looking at human-directed machine-evolved generative art, and ended up with a far more general discovery.

It seems to be published here: https://github.com/manoloide/AllSketchs

I don't think we're being radical enough. I think this kind of art is fantastic, but it copies human-designed art. Let's do things that aren't possible with paint or crayons - let's do multi-dimensional time-shifting video art.

I guess we need to do stuff that looks like 'normal' art first, to get people used to it.

In my town we call that stuff "video games."

I think of most generative art as video games larping as "fine art."

Have you heard of the demoscene? The modern scene is, to a large degree, art as a computer-based audiovisual experience rendered in real time.

Doing something that is only possible with machines in art is indeed what I think this "genre"/period needs in order to be seen as something truly unique and new.

As you point out this is merely creating the same kind of art with a new tool. The use of Processing to create familiar works is the same as picking up a piece of charcoal for the first time in human history and using only one of its worn-down edges to create hand print images on a rock wall. It's interesting and fresh, but it's not taking advantage of the medium.

I've been making art from randomly generated numbers for years. It's a realization of Pollock's objective where the viewer's subconscious is the only artist free from bias of any other person, a conduit of the mind.


I believe the purpose of Art is to reveal the soul of the Artist. This is why simply copying (or trying to emulate) the work of another artist is not considered art. However, learning to follow the rules set by others is a helpful stepping stone to striking out on your own and breaking those rules.

One has to remember that "generative art" only characterizes techniques for generating pictures. Art itself is about what the picture makes people feel, or what they have to say. Just making pictures for the only purpose of being "beautiful" is fine, just not very artful...

> Just making pictures for the only purpose of being "beautiful" is fine, just not very artful...

Beauty is the goal of art.

If his works don't resonate as well as a Nikolay Dubovskoy or a Cy Twombly, or any other human artist, it's in beauty that he fails.

I find that generally this confusion around "beauty" comes from using a popular but non-standard definition of beauty.

Beauty has meant, since the ancient Greeks and up through the romantics to today, "a higher emotional response", whether joyful, sorrowful or disgust.

Beauty is often confused for perfection among laymen. This is not a popular definition of beauty with aesthetic theoreticians, and only reached popularity under the regimes of the early 20th century fascists.

Beauty is a revelatory emotion, any revelatory emotion.

In aesthetics, the opposite of beauty isn't ugliness, but literalism and cold intellectualization.

That which isn't beauty in art is mere journalism, politics, or advertising -- these are lesser concerns than beauty.

Confusing that which isn't beauty in art for the core of art is like thinking that fashion and hairstyles is the core of music, when quite the opposite is true. Fashion diminishes music, not strengthens it.

Similarly -- that which is not about beauty in art diminishes it, not strengthens it.

I feel like you're imposing very strict definitions on what is inherently a very ambiguous or 'loose' field.

>Beauty is the goal of art.

I don't think so. Lots of paintings are not beautiful, but are very meaningful and artistic. For example, I find many of Frida Kahlo's works to be extremely artistic but not beautiful.

>Lyrics are what diminishes music

So you think that a song like 'Hurt' (NIN or Cash) is diminished by the very well written lyrics? Or that lyrics with certain rhyming structures subtract from the musicality of a song?

> Lots of paintings are not beautiful, but are very meaningful and artistic.

Like the GP, you're using a definition of beauty which is not at all standard in the field of aesthetics.

If the work is visually meaningful and artistic, while producing a feeling that is in some way sublime, transcendental, revelatory, etc., then that's beauty you're seeing.

> you think that a song like 'Hurt' (NIN or Cash) is diminished by the very well written lyrics

Lyrics will make bad music better, by distracting from the music. Strip away the lyrics, and the underlying music in Hurt is pathetic when stood up against just about any Classical, Jazz, Progressive, et all, piece.

Lyrics != music. Lyrics are a thing added to music.

> I feel like you're imposing very strict definitions on what is inherently a very ambiguous or 'loose' field.

This is a very common sentiment among people who have not read anything from the last 3,000 years of the philosophy of aesthetics. :-P

No, of course I'm not using the definition of beauty from a specific branch of philosophy. In common language, that's simply not what's understood by the word beauty.

"Aesthetics" isn't a branch of philosophy, unless you consider Formal Logic, Physics and Psychology branches of philosophy as well, it's separate a field of study.

Aesthetics is the sum total of all branches that relate to the thing we're all discussing here.

To flippantly suggest that your misguided lay definition of beauty is preferable in aesthetics than the one hashed out for the last 3000 years by those in the field, is silly.

That's like when laymen remind one that: "Evolution is just a theory", and ask why "monkeys didn't evolve as much as people did".

Any one with a science education will explain that they don't know what the technical terms theory and evolve mean.

And then they'll respond as you did: "I'm using the terms theory and evolve in the normal lay sense, like Conspiracy Theory and Pokemon Evolution. Evolution is just like your philosophy, man."

Don't be that guy.

If you're uninterested in a central field of the humanities, with a rich history going back to Plato, Aristotle, and Plotinus in the west, and further in the east, you don't have to go about arguing it on the internet like it's something you're in any way invested in, while spreading blanket falsehoods about the field.

If you think your having looked at Art makes you technically savvy enough to argue aesthetics, ask yourself if Facebook users are thus made qualified to argue software.

What are you even talking about? Aesthetics is literally considered a branch of western philosophy:


I'm not suggesting my definition of beauty is better, but it's definitely more appropriate given the context. Language is meant to communicate, which is why it's context sensitive. If someone tells me they have a theory about why the sink is blocked, I don't expect it to be a scientific theory - given the context.

And since I make a living from art and design, yeah I feel that I am qualified to have a few opinions on art. But maybe I should just leave that to the philosophers eh?

Aesthetics is a philosophy in the sense that all serious studies are. (Hence PhD, Doc·tor of Phi·los·o·phy, being the catch all for masters of most intellectual pursuits.)

That self-same wikipedia page you listed also denotes Logic and Cosmology as branches of Philosophy, and the Wikipedia page for Philosophy also lists Physics, Mathematics, Economics, Linguistics. Calling aesthetics a "branch" of philosophy in the pejorative sense that you have here is absurd, given that just about any serious humanity is technically an area of Philosophy.

Yes, I agree that language is contextual. And when you discovered that you were using a very informal definition of beauty on a board full of extremely well educated pedants, while having yourself only a middling appreciation for the study of aesthetics, you'd think your language would modify to reflect that realization. :D

As to your field -- working in a field related to Aesthetics does not mean you have education in Aesthetics.

I make a living in Art and design as well, but that isn't the foundation for my arguments in Aesthetics. Quite the contrary. People working in Art are usually some of the least educated in aesthetics, since our art education has been destroyed by baby boomer hippies.

Don't confuse your craft of Art, with the science of aesthetics.

That would be as absurd as an engineer arguing that his having built a building which stands up to gravity means he has a strong education in the study of gravity. The one hardly relates to the other.

The ability to create works which stand up to the gravity of Aesthetics does not mean that one has a strong education in Aesthetics itself.

Now, I'm not going to say that every engineer who hasn't studied theoretical physics is in a poor position to be an engineer, but it certainly tells you something about their seriousness if they haven't read any books in the field.

Further, if an engineer went to some length to pejoratively tell me that Physics is just a branch of western philosophy when debating the theoretical foundation of the force of gravity, I'd similarly call them out as seemingly not serious.

I'm curious how your theory of art deals with minimalism and conceptual art. I suspect you'd have to admit that mid-century we collectively expanded the definition of beauty to include modern ideas – structure, process – and expressions of irony.

As for irony, ask any art historian with experience in middle ages or late antiquity art. She'll tell you that it has always been a popular device in the arts.

The problem with "irony", where this common myth about its novelty comes from, is that Irony is invisible to generations separated by time and place.

To see irony in a work of antiquity requires a significant amount of education in the circumstances of that work. Irony never lies of the surface of a work. If enough time passes, our appreciation of that irony will often be lost.

How much irony is hidden there in the vase paintings of ancient Greece which we will never recognize, having not the context to see it? Irony, a Greek term, coined by Aristotle, passed on by the Romans, and more recently popularized in the 1500's by the French -- yet could we recognize Irony in any of their works without help?

If the average art critic finds no instances of irony in the art of the past they could be excused for that, their job isn't art history. Those are separate professions.

Where art critics do deserve blame however is when they've been shown countless examples of irony in the works of antiquity, and they act as if those instances are all unique anomalies. If they continue to bandy about the story of irony's novelty to defend the valuations of the works of their contemporaries, that's not fair play.

That's Art Criticism made Marketing, and it's an ugly but extremely common thing.

Many thanks for the explanation – you've illuminated a path to a deeper understanding of art history and aesthetics, which I hope one day to possess!

Before I was just discussing the Aesthetician’s Orthodox.

But here, I’ll answer for myself:

As far as minimalist art, there has been something called “design” which has been considered beautiful since long before modern artists began their war against representationalism.

Beautiful abstract design was the standard throughout all of the areas conquered by Islam, with the arabesques, as well as extremely popular across both Italic Europe and Northern Europe (with the mixture of Celtic Art and Roman giving rise to 10-Century Romanesque).

Abstraction has been highly prized in Europe for thousands of years, just search google images for "Romanesque capital". The Romanesque Capitals are just as abstract and surreal as anything created in the 20th Century, and this lineage of minimalism had at no point died.

Further, the cubists and so on, for their part, were mostly just copying African tribal art. The central thesis that minimalism was New was not historically accurate.

Conceptual art, for its part, used to just be called “Action”, and some actions were considered meaningful and others not so meaningful. Particularly meaningful Actions have always been considered beautiful, profound, transcendental, etc. The figure Jesus is immortalized in the story that he, who said himself God, washed the feet of a prostitute with his hair. You don’t get more conceptual art than that.

The war in the early 20th century was NOT whether design and action had meaning or value or beauty, that has been well established since the beginning of human civilization. In fact, most of civilization would have said that Action has always been more intrinsically valued than Art since it effects something tangible, and certainly Design was through most of civilization more prized.

The war in the early 20th century was over whether CALLING your designs or actions by the designation Big-A Art transmuted them into another stuff and imbued them with intrinsic value that they didn’t already have. I’m not at all convinced that the modernists have won that argument. That line of thinking has always been profoundly refused by the general public, and as time goes on, their central thesis appears to be in decline in popularity with critics, art historians, and theorists.

The irony is that the 20th century modernist’s claims at novelty and intrinsic value all required that one totally ignore the history and value of folk art and design, artful action, and so on, throughout human history and across human cultures. If one takes a more open-minded view of art and art history, the modernist’s claims at originality are laughable at best.

I agree with you, but I was referring to the "beauty" they talk about in the article (which is why I put it in quotes), which is something along the lines of "that that pleases the eye".

Not just beauty. Also expression, and emotional appeal. Some very ugly paintings and sculptures speak to whole generations about injustice etc.

Maybe generative art fails there too, I don't know.

Ugliness is not the opposite of beauty. Ugliness is a form of beauty, or at least its necessary foil. This has been understood for thousands of years.

See Laocoön and His Sons, one of the most beautiful works of art in the human tradition, because of the tragic ugliness therein.

See Hotel Rwanda, one of the most beautiful films ever made, because of the tragic ugliness therein.

Injustice in art often builds the beauty. If there was only injustice, with nothing transcendent about it, it would not stand up as a work of art. It would be brutality porn.

Take that most famous picture of slavery, the scarred photo of a slave's back. The picture's power derives not just from the scars, a record of his mistreatment, but also from his uprightness and strength. There would be nothing transcendent or beautiful in it were it a photo of a man reduced to nothing. It would not be art.

Similarly, the photo of the Chinese man in front of tanks in Tienanmen square is not just about the brutality of communism, but also the strength of the individual. If the photo was merely of a man crushed under a tank, it would scarcely be art. At best it would be journalism, but more likely it would be mere violence porn.

Discovered him on Twitter last year and was immediately blown away to see how his compositions resembled futurist and constructivist works to an uncanny extant. (Malevich, Kandinsky et al.) There is something about his choice of color palette that makes him really stand apart.

These are some beautiful images. I am relieved to see this article is about a person and not an algorithm. I wish I knew what quality it is about these that are more appealing. I think it is the abstract nature of it. Some look like flowers, some like 80s pop art, some reminiscent of Dali. The images remind me of specific human shared experiences without having to reproduce an dog eyeball in the middle. Human Generative (algorithms + design) vs AI Generative (algorithms only). Maybe I am biased for team human and if someone told me these images were created by a neural network I would be more critical. Although I don't think these could be created by a NN. Certainly not picking different themes for each set.

Interesting that he wants to keep it all digital and prefers it not made tangible.

Seems like it would be much harder to make any sort of living from your work by limiting the format to just digital.

Really cool nonetheless.

I really do love generative art, but this article was so hyperbolic I couldn't make it through. I think this is generating designs, nice patterns for throw pillows, but not art. The only generative artist I know who deserves the article's level of idolatry is Tarn Adams, because Dwarf Fortress actually creates stories and scenarios that provoke thought.

This gives me an idea. What it game textures for 3D models were entirely made up of shaders/generative algorithms. The entire game could be rendered with an equation and never have to load assets.

That's how a lot of the demoscene stuff works I'm sure!


> The secret behind all this is procedural content generation. in a nutshell, instead of storing a movie as-is we're storing a certain number of mathematical formulas for image and audio manipulation as well as the "recipe" how to apply those small bits of code in a way that what you see comes out.

One of the currently-trending productions on Shadertoy happens to be this recreation of the opening of Unreal, all procedurally generated (except for the geometry, though you can certainly find procedurally generated geometry elsewhere on that site).


No Man's Sky is similar. Effectively unlimited procedurally generated planets, with unique flora/fauna/terrain/etc.

Take a look at an old PS1 game called LSD: Dream Emulator that did this to interesting effect. Alt J used a screenshot from it for cover art of one of their albums.

Check out Signed Distance Fields (SDF) on Shadertoy.

Interesting stuff. All works of art tend to build on what came before, this does so in a much more direct way. At what point is something a derivative work rather than original?

Actually if you check out more of his work it looks quite different from what was chosen for this article. I think the article wants to portray a logical progression from modern art to generative art, or just to show that an algorithm is capable of generating images that are similar to those of highly creative human artists.



He also is a young guy with plenty of time ahead to develop his own style. As someone who also experiments with generative art and music I am absolutely impressed by what he's done so far.

You can't say generative art without saying demoscene.

Both generative art and the demoscene were preceded by decades of work with their own histories.

Very interesting pieces he has produced. I wonder how much is generated via rules vs simply coded. Cool find.

I thought this was going to be a profile of Scott Draves of Electric Sheep fame.

I know art is whatever people describe as art but I can't really get behind this specific kind of creation. To me this creates as much emotion as a gravatar and reminds me of my dabbling in fractal art. Just hit random until something nice comes out.

It's cheating because it was done on a computer!

It's a fairly common attitude that creating with a computer is somehow inferior, from Tron missing out on a special effects Oscar because computer graphics was 'cheating', to pretty much every designer being told at some point 'it can't be that hard, it's all done on a computer right?'

Expected better from HN though.

Well at risk of waking a tiger, I think his view was one of intentionality. Whereas he did not say it was because it was made with a computer, that sort of feels like putting words in his post.

I think cool generative art is the space stuff like no man's sky, and spore. I know it's maybe considered low brow but it was interesting to me.

"Just hit random until something nice comes out."

Reformulate as 'have a system for building new bodies, add a little bit of randomness, and have an environment that selects for survival' and you have life.

Life has ecosystems and - eventually - sentience.

This has neither.

To be fair, it's much more interesting than most generative art. Although it's also very influenced by heritage art styles, which makes it less fresh than it might be.

I think the prodigy label is more likely to apply to someone like Raven Kwok. I've been watching him do amazing things with code for a few years now, and he's been consistently inventive, original, and surprising.

It takes a lot of monkeys with a lot of paintbrushes a very, very long time to even complete a paint-by-numbers. This is worlds away from simple randomness; I invite you to try "just hit[ting] random" and see what happens. The art is in finding restrictions on the process, and that has clearly happened in these images. (Incidentally, that's also the case in "traditional" media too.)

Irritating and parochial post.

It doesn't look random at all. There is thought behind it. Hence the code.

At all?

> Just hit random until something nice comes out.

well, congrats, you're an artist ?

This site is blocked by the coffeeshop wifi as inappropriate content. I'm at Tim Hortons in Ottawa.

+1 416 977 8000

The butter donut reminds me of Alaska.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact