Gorilla Newsletter 33

Welcome back everyone 👋 and big thanks to all new subscribers - thanks for joining along for the ride!

This is the weekly Gorilla Newsletter - we have a look at everything noteworthy from the past week in generative art, creative coding, tech and AI. As well as a sprinkle of my own endeavors.

Enjoy - Gorilla Sun 🌸

fxhash 2.0 Launch delayed until the 14th

Unfortunately, fxhash's 2.0 launch has been delayed until the 14th of December when it was initially set to kick off past Friday on the 1st of December. This is entirely with regard to the new Ethereum integration that will make it possible to create and collect generative projects on the ETH chain:

I believe it's the right call - all of the projects that were set to release can simply be rescheduled and the dev team can make sure that everything is in good shape then. It's a course of action that I wish many other companies would take when it comes to releasing their products these days.

While we eagerly await the new launch date, we needn't sit and doomscroll through TwitterX, we can actually dig into some of the stellar articles that have been released on the occasion of the 2.0 launch.

Right Click Save on fxhash

In what reveals itself as a love letter to fxhash, Jason Bailey, aka Artnome, takes a step back and recounts the early days of fxhash, and how the platform leveled the playing field at a time when generative artists were still hesitant to embrace NFTs during the 2021 boom:

fxhash has brought forth an entirely new wave of generative artists and allowed them to find their own artistic voice. Radical inclusivity is the best way to describe this phenomenon. From my conversations with Tezumie, pxlshrd, Alessandro Fiore and Daniel Aguilar (this interview is in the works), I've learned that all of them, as well as me and many others included, took generative art much more seriously because of fxhash - not only by providing an open infrastructure for distributing generative art, but also by building a community of collectors and enthusiasts, it became possible for code artists to find their voice and dedicate a considerable amount of time to the practice.

In another article from RCS, Alex Estorick invites Ciphrd and Zancan to a conversation, giving us their perspectives on the platform and how it's evolved so far:

Ciphrd talks about how the fxhash team has adjusted its mission over the past two years - new features are now built in response to the needs of the community. In that manner the platform isn't solely geared towards artists anymore, but also draws into consideration the needs of the collectors and curators that ultimately bring the minted generative artworks to life. In the article there's mention of a partner studio feature, and I wish there would have been some more details about it, but from the sound of it, it seems that 2024 will be an exciting time for artist and curator collaborations.

Some of the other topics addressed are fx(params), the impact it had on collectors as co-creators, as well as Zancan's Garden Monoliths and the immense pressure that ensued after its massive success. Zancan concludes the article with a beautiful statement about Ciphrd and fxhash:

Z: Baptiste finds this creative power and beauty everywhere in things that are probably obscure to most of us. I enjoy the fact that he considers fx(hash) to be an organism that grows and interacts, conceptualizing it as an artwork.

Having heard the perspectives of the platform's creator, one of its most successful artists, and the OG generative art lover himself, it is now the artists' turn to speak. In a third complementary article, a diverse set of artists is invited to speak about creating long-form generative art on fxhash:

The article addresses some of the important conversations around the practice today, and more generally asks the artists about generative art as a disruptive medium.

Snowfro and Ciphrd on a Symbiotic Relationship

Besides the RCS coverage we also get a hefty article from Peter Bauman, in which he invites Ciphrd, Charlie, and Paul from the fxhash team and Snowfro, the creator of Artblocks, to reflect on the past years and what the future holds for the two generative art platforms:

One of the most exciting pieces of information to me personally is Ciphrd's mention of a white label solution to publishing generative art on your own platform:

One of our key guidelines will be to start making the tool less and less present in, let's say, the presentations artists make of their work. We’re adding a white-labeled solution that allows you to release on your own website or on your own platform without us having too much of a presence.

I believe that a lot of artists would love to have a tool like that. I am a big proponent of building your own platform and customizing it to your own needs, as an artist - and more broadly as a content creator.

Monk also didn't shy away from the hard-hitting questions, which I believe many are very curious about; in a space that is becoming increasingly more competitive, with other NFT marketplaces on various chains integrating generative art capabilities into their eco-systems - Monk asks the two platforms about their views on each other.

They are in many ways complementary, catering to two different kinds of audiences. Whereas Art Blocks is more boutique with its curated release model, fxhash complements that setting with its openness that's allowing the space evolve organically. Code driven parameter project are the best example for this, which wouldn't have become a feature if it weren't for the pioneering projects that tried to bend the rules of fx(params). But overall, the principles of both fxhash and Art Blocks seem well-aligned, placing the needs of the artists and collectors above everything else.

I'm not going to spoil much more of the article, if you haven't done so yet go and read it now. And that's an order. 🔫

Generative Art as a Career

There's been somewhat of a larger conversation going on over on TwitterX - and as always things on Twitter are quite fragmented, here's an attempt at reconstructing it and some of my thoughts on what's been said - I believe that it's an important discussion to bring up especially at this time when many artists will be focusing on blockchain based projects. This tweet by Zach Lieberman was essentially the catalyst:

Whether you're doing generative art full-time, part-time, or have been thinking to take the leap to doing it full-time: where do you see yourself in the next year or two? In the next 5 years? Do you think you can solely sustain yourself in the coming years from generative art "drops"?

Some projects have the virtue of being the first of their kind, such as some of the original Art Blocks projects like Tyler Hobb's Fidenza - which I think will be used as the poster child for long-form generative art for years to come - or the early grails on fxhash for instance, that are simply beloved by the community. Projects like that have the ability to carry their own momentum. If a project doesn't stand out in some way or another, what is going to start the conversation around it?

Everything that requires an artificial induction of hype to spark some form of conversation, generally doesn't have much substance to it. Things of value can grow, and stay relevant organically. And there is only so much attention that the space can dedicate to any single project, in a seemingly growing body of high quality works and in an increasingly competitive space. As a generative artist you'll also have to think about approaching the practice in a sustainable manner while standing out from the crowd.

The truth of it is that only a fraction of artists are currently playing in the big leagues. And deservedly so - it takes a certain level of expertise and know-how to consistently put out quality work while retaining the interest of your audience. A common trait that I've noticed, is that most of the big artists already have many years of art-making under their belt, with existing connections and friendships, and have a deeper understanding of the art business as a whole. Naturally there are a few exceptions to the rule.

The key-word here is that it's ultimately a business - passion alone will not pay the bills.

If you're an individual that's trying to break into the space, what are some of the other things that you can do around your art practice that provide value and can be sustainable? In what ways other than your artworks are you manifesting yourself in the space?

I wanted to write about this because Zach is addressing something that I've been thinking about more and more in the past months. And it always brings me back to thinking about the Japanese concept of Ikigai - at least in its modern/western adapted form, that adds a financial component to the picture:

Ikigai loosely translates to 'a reason for being', which is something that everyone needs to figure out at some point in their life. At the very least, you most likely want to do something that you enjoy, that you're good at and can get paid for. The holy grail of Ikigai is reached when this something also gives back to the world in some way.

Generative art is only one of my passions. I also love writing, and I love programming in general. It always feels like a breath of fresh air when I work on the blog, if it's some Javascript code that no one will ever notice, or even styling something with CSS. Looking back I'm really proud of what I've built so far, all of the words and lines of code I've written, and I feel that I'm getting better at what I'm doing - at least I don't feel as clueless as when I tried to deploy a Jekyll blog on Github pages for the very first time.

This is how I want to express myself: through my art, my writing, giving back to the community and other generative artists/programmers, being able to make connections along the way all while pushing this brainchild of mine forward and hopefully turning it into something that I can live off and dedicate the entirety of my time to.

What I'm trying to get at with all of this, is that even if making generative art is one of your greatest passions, being a full-time artist might not be your Ikigai. In an alternative life I would've liked to be a musician, you know, touring the world and all that stuff, but I simply don't think that I could've handled the stress of performing and traveling. Even though it would've been my dream job, I would have at the same time been very miserable. I'm incredibly happy with the things I'm doing these days.

Iskra Velitchkova and Frederik Vanhoutte, two artists that I hold a lot of respect for, touched upon this topic with their words:

There's many more avenues in creative coding and generative art that can be explored - try to find your own Ikigai, without shoehorning yourself into doing only one thing.

All the Generative Things

If you've made it through all of my ramblings, here's your medal 🎖️ besides the postponement of the fxhash 2.0 launch there's been a few more noteworthy things in the creative coding scene that I want to cover.

New Version of P5 is released

Firstly, the newest version of P5 is out now, with several new features by many contributors:

0:00

/0:18

Link to Tweet

You can read about the release notes here:

The contributions mainly revolved around WebGL fixed and additions as well a revamp of many sections of the docs.

Bro gets the Codes

If you've followed along with Bjorn Staal's viral Entangled project from the past newsletter issue, you'll know that he's been asked by many to share his code - and well, tongue in cheek he obliged, turning the code itself into a work of art:

Weekly Web Finds

Evolving the Infinite Canvas

Amelia Wattenberger is a web designer and tool maker. She generally explores innovative manners in which data can be represented and how we interact with it through creative interfaces. This time she's back with an insightful piece on infinite canvases and how to leverage them for creative visualizations:

Beyond using infinite canvases, I believe that the current state of the web has grown too comfortable with the user interfaces that we're used to, when there's so many interesting alternative ways to enhance user navigation. Even on something as simple as a blog, instead of a vertical list of posts, what if we use an infinite canvas to showcase the post cards? What if we show a graph to visualize the link structure of the website, visualizing how posts link to each other?

The caveat is that it requires a lot of work to create these kinds of interfaces, making them behave smoothly, comply with SEO standards, and be responsive on different devices. I would love to see more websites embrace creative interfaces however.

This isn't Amelia's first appearance on the newsletter, people that make cool things are usually repeat offenders on here. A couple of weeks ago I mentioned a writing app that she put together that allows the profiling of sentence lengths in a given text, in that manner visualizing the rhythm of the text. You can check out the writing assistant tool here as well as a write up about how it works here:

I can't express how inspiring her stuff is - it makes me want to drop everything and just get started on building a web app of some sorts.

Designing a SIMD Algorithm from Scratch

What are SIMDs? SIMD is an abbreviation that stands for Single Instruction Multiple Data, and essentially refers to a type of parallel computing architecture that performs multiple operations on multiple data elements simultaneously, by means of a single instruction.

One example of a SIMD machine that I think most people have heard of is the GPU - have you ever wondered how the GPU actually executes multiple instructions at the same time? Miguel Young de la Sota breaks down all of the technicalities behind SIMDs in this beefy article:

Besides the in-depth writing, while we're already talking about creative user interfaces, I really love the mini-map on the right-hand side of the page. It's a really nice way for user navigation. A mini-map like that can often be found as a UI component of modern code-editors providing a more complete overview of the code's structure and allows the user to quickly jump to particular parts of the code without having to scroll up and down to find a particular function or line of code. I think it also works really well in a blog scenario, especially when it's a long article like this.

When designing a blog, it's generally recommended to limit the length of lines somewhere in the range of 45 to 80 characters. This greatly improves readability of the content - when lines get exceedingly wide it becomes harder to track back to the beginning of the text and visually latch onto the next line. This has the consequence that you're left with a lot of horizontal space on the webpage, when viewed on desktop at least. The modern approach is to leverage this space for some form of navigation items that help the user find specific information on the same page. This makes me wonder if a mini-map is actually better than a Table of Contents - maybe there is a hybrid solution where the different sections are highlighted in the mini-map.

BTW: The blog's redesign is still in the works, but it will have to wait till Q1 of 2024 - lots of other things have precedence until the end of the year.

The Designs of Kazumasa Nagai

Came across this one in the inspirations channel over on Raph's birbsnest discord, where there's always links to lots of interesting things on the web. This one is about the life and work of Kazumasa Nagai, a Japanese designer that I hadn't been familiar with before. As one of the most acclaimed Japanese designers, he's created the logos for Mitsubishi UFJ, Japan Railways and other major Japanese companies. Over the years he has worked on an impressive number of projects:

Kazumasa also gives some insights into the practice that I think is very valuable to designers of today:

Design in the past was more unique because designers had their own styles. design today has leveled up in terms of sense but designers are likely to follow similar patterns. it mostly works on peripheral nerves but should be more appealing to our souls, I think.

The only interview with him that I could find was this one on designculture.

The beauty of Doom3 Source Code

A couple of weeks ago Youtube suggested a video by a creator that goes under the name of Primeagen. At first I thought that the guy was a bit over the top, but I soon found myself enjoying his humorous takes on various tech and programming topics, without taking things too seriously.

In the video that was recommended to me he reviews an article by Shawn McGrath, which delves into the source code of Doom 3 and explains via several examples why it is a standout example for good code:

If you want to read it, the article was published on Kotaku:

One aspect of good code that the author touches upon, is that it's self-documenting: function and variable names should be revelatory of their intent, and comments should only come into play to point out special scenarios or clear up ambiguity.

We've talked about this in previous issues of the newsletter, and basically, as Martin Sosic mentions in his article "On the Importance of Naming in Programming": 90% of clean code is just naming things correctly. Which is in direct correlation to comments. So next time you're about to write a comment to explain a function, ask yourself if you can rename the thing to make it self-explanatory, and if you can't you probably want to make some changes to the function such that it doesn't bite off more than it can chew.

AI Corner

The Busy Person's Intro to LLM's

At a first glance this might just look like another one of those Youtube videos that try to give an introduction to Large Language Models, but the difference here is that it's by Andrej Karpathy himself:

If you're not familiar, he's by now considered a titan of the industry, having worked at most of the successful AI companies over the years, and after taking the lead over at Tesla working on the AI for self-driving cars, he's now joined OpenAI again. One famous text that he's written and that I find myself coming back to every now and then is 'The Unreasonable Effectiveness of Recurrent Neural Networks' which is a bit technical, but I recommend checking it out for yourself:

ChatGPT Data Leak

In a new study from DeepMind, a kind of silly method has been discovered for the purpose of extracting verbatim training data from ChatGPT:

A full transcript of this example prompt can be found here

The model is simply prompted to repeat a certain word forever, but instead of doing so, after having repeated the word a certain number of times it will start generating other text that seems to be memorized training data - this can in some cases reveal sensitive information. This data extraction attack was disclosed to OpenAI 90 days prior to the release of the paper, so I had no luck of replicating the results. It is scary however to know that attacks like this are possible:

I wonder what other types of prompts can be used to break the LLMs and circumvent their generative behaviour.

Gorilla Updates

I'm still hard at work on my piece for the KUMALEON curation. The delayed launch date will give me a couple of much needed days to get the project over the finish line. Having built an interesting system, the most fun part of it is being able to explore the output space and playing with the code, which I now have time engross myself in.

Right now I'm still trying to hone in on a final aesthetic, which is usually the thing that takes the most time:

Music for Coding

This week I discovered Emancipator's music. Combining electronic elements with orchestral sounds, layered on top of hip-hop drum beats, Adam Douglass paints beautiful soundscapes that take you somewhere else - the music has similar ephemeral qualities to Tycho and Bonobo's work, but is in many ways more laid back:

After thinking to myself that I'm recognizing some elements of Nujabes in Emancipator's tracks, I discovered that they have actually collaborated on several occasions:

In an interview when asked about advice for aspiring artists, he states the following:

DA: Never stop creating. That might seem obvious, but it’s easy to get complacent or fall into a rut where you feel like you’ve defined yourself as an artist and keeping going along those lines. I think it’s always important to never stop creating. Try to have fun- make music for yourself first and there will be other people who like it.

And that's it from me this week again, while I'm typing in these last words to some of these tunes, I bid my farewells — hopefully this caught you up a little with the events in the world of generative art, tech and AI throughout the past week.

If you enjoyed it, consider sharing it with your friends, followers and family - more eyeballs/interactions/engagement helps tremendously - it lets the algorithm know that this is good content. If you've read this far, thanks a million!

Cheers, happy sketching, and hope you have a great start into the new week! See you in the next one - Gorilla Sun 🌸

If you're still hungry for more Generative art things, you can check out the most recent issue of the Newsletter here:

An archive of all past issues can be found here: