I like your thoughts. Can’t wait for my own second brain.
I think your right on with learning the tools and selfhosting- big companies companies are organized around making money, whether it’s good for the users or not.
I see a parallel pathway in terms of work- pursuing and incentivizing self sufficiency. The more robustly self sufficient communities there are, in terms of food, manufacturing, healthcare, infrastructure, the more resistent our society will be to shock.
We are in a very fragile system, and I am afraid that it wont be able to adapt quickly enough to the changes coming.
Plus I like to think about what I’d want to do in a post-scarcity society and work backwards. I would be farming, cooking, caring for people, making stuff.
You’re the first person to comment that I don’t know personally which is a major first milestone for the community so thank you.
Regarding self hosting I think it’s very important that we have the ability to self host, but I don’t necessarily think everyone needs to self host.
I realize that may sound a little counter intuitive but my main concern is control and ownership.
That’s why when i was building websites i always used wordpress.org. I never did host my sites locally but the fact that i always could if I wanted gave me the assurance I needed.
There’s a lot of drama happening there now but anyone who runs on core wordpress can opt out at anytime by forking and run their sites without anyone else involved if they want which is great.
I wanted to take this opportunity to explain this as i also run Open Web UI in the cloud in addition to locally.
And I use my cloud hosted version more because it is more convenient and I still have a very large degree of control.
With all that being said very open to opposing views for discussion.
I share your worries about our current systems being able to adapt to what’s coming and the perverse power dynamics and fragility that are inherent with centrally controlled systems.
I said to another friend the other day if you’re into ai and have ever thought about moving off grid then you’d probably be into what I talk about lol.
Getting to the point where people can work on their true dreams and passions as you’ve outlined below for yourself is one of the huge potential positives of this technology that I am excited to work towards as well.
Are you using Open WebUI or any of the other open source stuff out there yet or still in the learning phase?
I’m not a purist when it comes to self hosting, or much of anything really. Gotta do the best you can with what you got.
I’ve been fiddling away at my own implementation of a second brain, based mostly on the langchain genai stack, but between work and life I’ve been feeling quite stuck lately. I’m really excited to try openwebui. I’ve also used comfyui and other image generators extensively. I’ve been working in tech for a decade so pretty comfortable getting into open source projects.
Actually I found you through your posts on linkedin and elsewhere looking for a tech evangelist. I’m definitely interested as it aligns with what I’ve been doing and what I want to do. Actually me and some friends are putting on a conference next month you might be interested in, New Orleans AI (NOAI). Heres a link https://noai.philosophers.group/. I’m working on a talk for it and hoping to include some hands on stuff. Excited to tap into openwebui and see what it can do, and I hope soon to produce some content about it.
Cool that you are already working on piecing this together.
I hear you on the stuck with work and life point. I think there are a lot of people feeling that way.
One of the things I find is that the more into AI I get and the more ways I see that it can improve the world the more I also notice how broken many things are currently.
I feel the best when i am working on things and with people that are aligned on potential solutions. So that’s a big reason that I’m doing this and my mood always dramatically improves when I’m doing stuff like this.
I haven’t done much yet on the image generation side besides playing with mid journey and grok which i think runs on flux.
I’d be really interested in an option that can help me with youtube thumbnails and featured images for my articles.
The issue i’ve had id the images i get are a bit too creative for my style which i want to be straightforward.
That being said im not advanced with image prompting either.
Any suggestions you have on the image front for me would be appreciated.
I’ll check out the conference link thanks.
Also happy to discuss the evangelist position, I’ll reach out via DM on that now.
I really like what your saying about AI and seeing how we can improve the world. It’s gonna change one way or another, best to think how it can change for the better.
As far as image generation goes I use stable diffusion with either AUTOMATIC1111 or comfyui. I see those are both supported by openwebui so it will be interesting to explore that.
There are lots of little tricks with image generation. I approach it mostly as play. Some things that can really help is specifying the artistic style, using negative prompts, and finding an image model that matches the style you want.
I think thumbnails should be pretty straighforward.