Brian Jackson
@axnjaxn.bsky.social
Scientist on weekdays, PICO-8 demoscene enthusiast over the weekend, and Principal of the Excluded Middle School the rest of the time (He/Him)
Research interests: computer graphics, image processing, programming languages, virtual computing
Research interests: computer graphics, image processing, programming languages, virtual computing
Refraining consumption as art is deeply self-deprecating - submitting to spiritual domestication underneath a host of commercial masters. It is not only rejecting the brush as extension of one's hand, but the grafting of a billboard onto the stump of one's wrist in its place.
June 4, 2025 at 7:43 AM
Refraining consumption as art is deeply self-deprecating - submitting to spiritual domestication underneath a host of commercial masters. It is not only rejecting the brush as extension of one's hand, but the grafting of a billboard onto the stump of one's wrist in its place.
The point of this whole rant is that communication is central to the human experience, but that framing another channel for consuming content as artistic creativity is incredibly insulting, and not only to those who "do it the hard way," but to oneself.
June 4, 2025 at 7:43 AM
The point of this whole rant is that communication is central to the human experience, but that framing another channel for consuming content as artistic creativity is incredibly insulting, and not only to those who "do it the hard way," but to oneself.
This polemic is not aimed at the use of large models to overcome an otherwise-impossible scale, timeliness, logistics, or scientific problem, either. Or else I'd be an enormous hypocrite: some problems can't be solved practically without complex statistics applied to huge datasets.
June 4, 2025 at 7:43 AM
This polemic is not aimed at the use of large models to overcome an otherwise-impossible scale, timeliness, logistics, or scientific problem, either. Or else I'd be an enormous hypocrite: some problems can't be solved practically without complex statistics applied to huge datasets.
Artists' tools inherently extend or replace a function of human bodies and behaviors. A brush is a prosthetic finger, a lens is a prosthetic cornea, a notebook is prosthetic memory. So what is a prompting system replacing? The capacity to imagine? The skills of others, suitably packaged?
June 4, 2025 at 7:43 AM
Artists' tools inherently extend or replace a function of human bodies and behaviors. A brush is a prosthetic finger, a lens is a prosthetic cornea, a notebook is prosthetic memory. So what is a prompting system replacing? The capacity to imagine? The skills of others, suitably packaged?
Many, many artists have found creative ways to imitate. Even forgery can be art! What it communicates is negative, but the only communication available interacting with conversational AI agents is "I want this, no, more like that than like this."
June 4, 2025 at 7:43 AM
Many, many artists have found creative ways to imitate. Even forgery can be art! What it communicates is negative, but the only communication available interacting with conversational AI agents is "I want this, no, more like that than like this."
Becoming a user of software you didn't write, that summarizes an enormous amount of everyone's content but yours, and responds to queries is the pattern of a user - and users aren't inherently bad, there's just a difference between audience and musician.
June 4, 2025 at 7:43 AM
Becoming a user of software you didn't write, that summarizes an enormous amount of everyone's content but yours, and responds to queries is the pattern of a user - and users aren't inherently bad, there's just a difference between audience and musician.
You can write a program, even with assistive tools, and _that_ can be art. Art communicates about you through the use of technique, craft, creativity, experience, education, stubbornness.
June 4, 2025 at 7:43 AM
You can write a program, even with assistive tools, and _that_ can be art. Art communicates about you through the use of technique, craft, creativity, experience, education, stubbornness.
It's an incredibly thoughtful, well-executed game - I was shocked at how deliberate every design decision was.
February 12, 2025 at 3:49 PM
It's an incredibly thoughtful, well-executed game - I was shocked at how deliberate every design decision was.
Love it! Looks like a ruleset not too dissimilar to Puzzle Fighter? The sheer responsiveness of hard drops - no wasted space, no wasted friction - is incredibly satisfying.
February 11, 2025 at 6:57 AM
Love it! Looks like a ruleset not too dissimilar to Puzzle Fighter? The sheer responsiveness of hard drops - no wasted space, no wasted friction - is incredibly satisfying.
Oh! Also coincidentally: the Moire in this screenshot is one of the open items in the to do list for which AG's lecture in Denver gave me an idea, and eventually I'll climb back up out of compiler-land and properly deal with the Fourier-domain work to anti-alias the barrel distortion shown here.
February 11, 2025 at 6:54 AM
Oh! Also coincidentally: the Moire in this screenshot is one of the open items in the to do list for which AG's lecture in Denver gave me an idea, and eventually I'll climb back up out of compiler-land and properly deal with the Fourier-domain work to anti-alias the barrel distortion shown here.
Here's a screenshot of some of the onboard editors I was working on around the time SIGGRAPH came around - a font editor for the BFX-101 and a tiny shader I cooked up as an experiment after reading "Reconsideration of CRT Monitor Characteristics" (Katoh 1997)
February 11, 2025 at 6:49 AM
Here's a screenshot of some of the onboard editors I was working on around the time SIGGRAPH came around - a font editor for the BFX-101 and a tiny shader I cooked up as an experiment after reading "Reconsideration of CRT Monitor Characteristics" (Katoh 1997)
This has been: Brian Still Isn't Releasing Blua to the Public But He Really Likes Talking about It
February 11, 2025 at 6:49 AM
This has been: Brian Still Isn't Releasing Blua to the Public But He Really Likes Talking about It
So I have continued building out a lot of Blua's features since, and while it went on ice any time my day job got to be too heavy, it's become a circular labyrinth of interesting technical problems to play with, and there is an entrance in any room in any building I may presently occupy.
February 11, 2025 at 6:49 AM
So I have continued building out a lot of Blua's features since, and while it went on ice any time my day job got to be too heavy, it's become a circular labyrinth of interesting technical problems to play with, and there is an entrance in any room in any building I may presently occupy.
Their curiosity and freedom - doing things because it was fun, satisfying, puzzling, even obvious - helped me identify that there was a resonant frequency to which I was not the only vibrating instrument.
February 11, 2025 at 6:49 AM
Their curiosity and freedom - doing things because it was fun, satisfying, puzzling, even obvious - helped me identify that there was a resonant frequency to which I was not the only vibrating instrument.
Needless to say, @glassner.com and the others I got to meet in Denver held a much greater significance to me than any previous instance of the conference after all the work with Blua. These, at last, were creatures of the same species!
February 11, 2025 at 6:49 AM
Needless to say, @glassner.com and the others I got to meet in Denver held a much greater significance to me than any previous instance of the conference after all the work with Blua. These, at last, were creatures of the same species!
Ironically enough, I didn't connect that it was the same person that I'd had a lovely conversation with after his lectures at SIGGRAPH last summer until I'd gotten all the way home and Graphics Gems on top of my stack of books on the countertop.
February 11, 2025 at 6:49 AM
Ironically enough, I didn't connect that it was the same person that I'd had a lovely conversation with after his lectures at SIGGRAPH last summer until I'd gotten all the way home and Graphics Gems on top of my stack of books on the countertop.
I read a lot of Andrew Glassner's work in the form of his papers from the late 80s, and from Graphics Gems, of which he was editor (at least of a couple of the volumes I consulted).
February 11, 2025 at 6:16 AM
I read a lot of Andrew Glassner's work in the form of his papers from the late 80s, and from Graphics Gems, of which he was editor (at least of a couple of the volumes I consulted).
This whole exercise was incredibly educational, especially about cool tricks I thought I might have invented but inevitably showed up in a 1968 paper somewhere. GLR parsers, for instance, or the Graph-Structured Stack, or this or that integer method for rasterization.
February 11, 2025 at 6:16 AM
This whole exercise was incredibly educational, especially about cool tricks I thought I might have invented but inevitably showed up in a 1968 paper somewhere. GLR parsers, for instance, or the Graph-Structured Stack, or this or that integer method for rasterization.
Number 2 has been a slow crawl, and in the meantime every time I'd think of a cool feature I'd stick it in the notebook for the next revision to "hardware."
February 11, 2025 at 6:16 AM
Number 2 has been a slow crawl, and in the meantime every time I'd think of a cool feature I'd stick it in the notebook for the next revision to "hardware."
I finished number 1 last year - building a compiler-compiler in Blua (BCC), reimplementing the assembler, compiler, and BCC's own syntax as BCC grammars. And I had a huge battery of tests to verify behaviors would be equivalent and meet spec.
February 11, 2025 at 6:16 AM
I finished number 1 last year - building a compiler-compiler in Blua (BCC), reimplementing the assembler, compiler, and BCC's own syntax as BCC grammars. And I had a huge battery of tests to verify behaviors would be equivalent and meet spec.