I'm probably not the best person to elaborate on why societal anxieties arise every time our place in the universe shrinks. I guess observing reactions to copernican heliocentrism could be a good starting point. We're witnessing something similar, and while we may accept it at the individual level, the objections of the many will stir painful reactions.
This is so well said I can clearly picture it as a paragraph next to this exhibit in a museum or gallery (sorry I keep to exhibit your work but I just can't help it at this point...)
The text-to-image model is an important component, but the current model in use is IMO good enough. My view for this project is that the internal monologue is more important than the output, so my wish is instead for a better open-weight LLM.
I did! I want to switch to Mixtral-8x22B, time permitting. During the development of Stream of Consciousness I already swapped LLMs twice. This space is moving incredibly fast.
Glad you like it! One of the things I tried to emulate is the notion of obsessions, which I believe is an important component of the creative process. It seems like Livia got obsessed with Dutch Proverbs and slowly drifted to "Dutch protest culture".
I have no control over what happens during the live streaming and I'm sure this autonomous agent could become a source of embarrassment, though I hope the few precautions I've built might be enough. :-)
> I am visiting the website: firstamendmentmuseum.org
> I am seeing that in the 18th and 19th centuries, protest music utilized already-popular tunes with altered lyrics, and songs needed to be easy to learn and sing by protestors, picketers, activists, and more because recording technology was not available and songs needed to be passed down orally.
> I am seeing that in the 18th and 19th centuries, protest music utilized already-popular tunes with altered lyrics, and songs needed to be easy to learn and sing by protestors, picketers, activists, and more because recording technology was not available and songs needed to be passed down orally.
> I am seeing that in the 18th and 19th centuries, protest music utilized already-popular tunes with altered lyrics, and songs needed to be easy to learn and sing by protestors, picketers, activists, and more because recording technology was not available and songs needed to be passed down orally.
> I am thinking that I hope I didn't forget to pay a bill that was due today.
> I am visiting the website rateyourmusic.com
> I am seeing that in the 18th and 19th centuries, protest music utilized already-popular tunes with altered lyrics, and songs needed to be easy to learn and sing by protestors, picketers, activists, and more because recording technology was not available and songs needed to be passed down orally.
> I am visiting the website freedomforum.org
> I am seeing that in the 18th and 19th centuries, protest music utilized already-popular tunes with altered lyrics, and songs needed to be easy to learn and sing by protestors, picketers, activists, and more because recording technology was not available and songs needed to be passed down orally.
So part of me feels like the missing component to this nice "obsession" trait you added is the assessment of whether it is finding new information or just reading stale data. I think given known information, humans will either quickly dismiss it (sometimes just a few words into it) or try to appreciate it through a new perspective. The latter seems harder to achieve, but the former seems feasible today—when finding repeated information several iterations in a row, shake everything up and get creative again.
A corollary to that is "obsession" is perhaps balanced by some measure of "boredom"
Well spotted. The agent does in fact assess whether or not the information it's processing is already in its vector DB. I'm still unclear if what you noticed is due to the LLM failing (Mixtral is good, but not great), or something else, but it needs fixing. Thanks for letting me know.
Thank you for sharing your thoughts. I fully agree re: the importance of art and the absurdity of the concept which, I hope it's clear, is embedded in the spirit of this project.
On the topic of the performance and worries that something is not working: the real-time generation is currently bottlenecked by the GPU used for this project. We try to approximate human reading speed for the recorded sessions (which you can find in the archive).
So great to see interest in ePaper screens. Picture frame projects specifically seem to be gathering steam.
Shameless plug: I built a solar ePaper frame that displays Google Photo albums or loads pictures from an sd-card.
https://jamez.it/blog/2023/05/16/version-2-of-my-solar-power...
So far, I've shared the instructions on how to build a custom solar engine, which was the most time-consuming portion of the project. Coming up next: high level directions on how to build one from scratch.