Berlin Brown and Software Development

Big Claims, Real Limits There’s a lot of hype about AI right now. Some CEOs say AI has already surpassed human intelligence. Others claim it can replace all office work. Take this quote from OpenAI’s Sam Altman: “We are past the event horizon… The takeoff has started.” – Sam Altman, MarketWatch Interview (2025) He also wrote: “Robots are not yet walking the streets… we still can’t easily go to space, and there is a lot about the universe we don’t understand.” – Sam Altman’s Blog: The Gentle Singularity Bold stuff. But here’s my everyday, boots-on-the-ground experience. I’ve experimented with GitHub Copilot, DuckDuckGo AI, Agentic AI, Google’s AI tools, and I'm also building my own local LLM setup. I’d consider myself a layman when it comes to the deep internals of LLMs, but I use them regularly in a hands-on way. Good Uses of AI (So Far) AI tools work great for quick, well-scoped questions: “What’s 45 million divided by 100,000?” “How do I fix an SSL error in a Java HTTP...

 Here are some top courses and the impact to AI, so far not a big amount of change.  A couple of additions. Kind of reminds when I was school at UA, they added a Java source. See Georgia Tech and UA. https://catalog.gatech.edu/coursesaz/cs/ https://catalog.ua.edu/undergraduate/engineering/computer-science/courses/ CS 7545. Theoretical Foundations of Machine Learning. 3 Credit Hours. CS 7751. Probabilistic Graphical Models in Machine Learning. 3 Credit Hours. CS 7651. Human and Machine Learning. 3 Credit Hours. CS 7643. Deep Learning. 3 Credit Hours. Reinforcement Learning Computational Foundations of Machine Learning

 Elon Musk and Donald Trump feuding right now.

 I asked this into Chat GPT (source chatgpt question) "So when you look at current AI technology like LLM technology, it is currently, it is highly based on digital prompt inputs and analyzing large datasets with billions of parameters from text data.   The human brain does not necessarily operate that way.   What are current approaches to bridge the gap between low tech human brain processes to the machine learning algorithms with LLMs.  Projects like the numenta project and thousand brain project" Here is the response from chatgpt - source chatgpt model as of Jun5. "Recent AI research increasingly looks to the brain for inspiration. Neuroscience-derived ideas—such as sensorimotor learning, modular cortical columns, and memory consolidation—are being used to overcome limitations of today’s deep learning systems. For example, Numenta’s new Thousand Brains Project is an open-source initiative to build AI on cortical principles (sensorimotor input, grid-c...

 I remember some of these companies back in the day, now they are larger. I am going off memory: Google to Alphabet Yahoo to Yahoo Netscape - Didn't morph into Mozilla and Mozilla Foundation SmartBear - of Smartbear UI Fiddler to Fiddler AI 3dfx - bought out by Nvidia

 Changelog : changelog- May 13 - gamedev - updated and added characters, still just a little opengl demo.  See PR.  Changelog 1 opengl, java, libgdx https://github.com/berlinbrown/mechumbra-samples

 We are going to share that we’re currently working on a series of updates to our basic game development tutorials. These improvements are part of our ongoing commitment to helping developers—both beginners and experienced—build engaging and technically sound games. The upcoming changes will place a strong focus on OpenGL development, with special attention to both Java-based OpenGL and the LibGDX framework. If you’ve been following along with our earlier content, you’ll notice that the foundation has already been laid with examples covering core concepts such as camera movement, object movement, and other essential techniques. These examples are just the beginning. What to Expect Here’s a quick preview of what’s in store: Deep Dives into OpenGL: We’ll be expanding our coverage of OpenGL, especially from the perspective of Java developers. Expect walkthroughs on rendering pipelines, shader programming, and scene management using modern OpenGL principles. LibGDX Tutorials: As one of...

 https://chatgpt.com/s/dr_681654bc445c8191a77a94e2184172a4 ... Brain-Inspired Life-Simulation Projects The following projects use brain-like architectures (spiking networks, neuromorphic hardware, artificial neural tissues, or even living neural cultures) to model or reproduce life-like behaviors. Each section details a project’s goals, methods, modeled behaviors, implementation (hardware vs. software), available tools/code, and key publications or results. OpenWorm (Virtual C. elegans simulation) openworm.org pmc.ncbi.nlm.nih.gov Objective: Build a full virtual Caenorhabditis elegans worm so its behavior “emerges from its underlying physiological processes” pmc.ncbi.nlm.nih.gov . It is an open-source collaboration to create “the first virtual organism in a computer” openworm.org

 Still update C/C++ opengl setup for simulations. See updates: https://github.com/berlinbrown/OctaneMechOpenGLGame Also posted Mom book: Book: " "Carolyn J Brown Life on Land" Life on the Land: Memoir of a Farmer's Daughter Hardcover – March 17, And dad's "The Emerson Street Story: Race, Class, Quality of Life and Faith: In Business, Money, Politics, School, and More" ISBN-10 ‏ : ‎ 1665519428 ISBN-10 ‏ : ‎ 1728370787"

 Built with mac: ... https://github.com/berlinbrown/OctaneMechOpenGLGame/tree/master/retutorials ....

 Biohacking test - one, April 14, 2025 I testing various teas to calmness. Valerian Root from US grocery store, packaged.  Probably best for sleep but awkward feelings in stomach.   Will try again sparingly Chamomile Tea - US grocery store, packaged. I feel something going on, a certain calmness.  I dont' have much experience here Yerba Mate Tea - not really calmness but trying it out.  Ehh, can't tell.

 So making progress through my open projects. Updating mech umbra and octane umbra mech https://github.com/berlinbrown/OctaneMechOpenGLGame

 Looking at this source, looks fine to me.  Lot of comments, maybe added later. msword seems to have 200 files, 180,000 lines Looks well written, a lot of comments.  I think everyline is almost commented actually 31 ASM (assembly files), comments might have been added later? 24 C files https://www.gatesnotes.com/meet-bill/meet-bill-home-topic/reader/my-first-memoir-source-code https://github.com/michaelmsonne/MS-WORD https://github.com/microsoft/MS-DOS/tree/main/v4.0

From my libgdx examples https://github.com/berlinbrown/mechumbra-samples LibGDX is an open-source, cross-platform game development framework that lets you write your game once in Java and deploy it across desktop, Android, iOS, and web platforms. It has components for graphics, audio, input, and physics, LibGDX gives you everything you need to build a complete game.

 Just basic opengl Also, here are benchmarks for recent computer runs with Blender benchmark https://opendata.blender.org/users/benchmarks/ Fastest to Slowest Nvidia GeForce RTX 4060 - Windows - (score) 3105 Nvidia Geforce RTX 3060 - Windows - 2229 Nvidia Geforce GTX 1050 Ti - 197 AMD Ryzen 5 5600x 6-Core - 143 Intel Core i9-8950HK CPU @ 2.90GHz - 80 

 Go my personal site up, have fun. https://myberlinaustin.com/home/blog/index.html

 Check out the book here, From github, copyright, me. https://github.com/berlinbrown/all-berlin-blog-book/blob/main/v3/BerlinBrownBookLookSoftwareFinal_Refined_Fixed.pdf

"They use a Mixture-of-Experts (MoE) architecture, where only 37B parameters are activated for each token out of the total 671B. This sparse" https://composio.dev/blog/notes-on-new-deepseek-v3/  

 Just random acts of coding. Uploading my book, wanted chat gpt to clean up it up some. https://github.com/berlinbrown/all-berlin-blog-book/tree/main/v2 And RSS reader data, you can see the output here https://github.com/berlinbrown/umbra-corp-pub-web-extract/tree/main/experimental/umbra-crawler/output/samplerun/data/jan23/001563918924405/data

Our hearts go out to everyone affected by the recent LA fires. Your resilience inspires us all, and we want you to know that people across the country are thinking of you, hoping for your safety, and standing by your side. Stay strong—we're with you every step of the way.

  I asked this to Chat GPT, this is the response. "What would Aaron Swartz feel about the current state of the web, paywalls, advertising and cookie tracking?" Before I post, he actually was one of the creators of RSS, I am looking at now.  A machine readable web.  Wanted to post some links. Even RSS board has ads. https://www.rssboard.org/news/15/rss-advisory-board-goes-public https://www.openweb.com/resources This is straight from Chat GPT - openai "Given Aaron Swartz’s passionate advocacy for open access, freedom on the internet, and privacy rights, it's very likely that he would be deeply disappointed with several current trends on the web, including paywalls, advertising models, and cookie tracking. Here’s a breakdown of what he might think: 1. Paywalls and Information Access Swartz was a vocal proponent of open access to information, famously co-founding Infogami (which became Reddit ) and later being involved with initiatives like Open Library and The Public...