At the moment the deer population has to be controlled by humans because otherwise they’ll completely destroy entire forests, when lynx are introduced they’ll prey on the deer so humans don’t have to and the forests can recover.
We solved the Ein Stein Problem. And when I say we I mean people way smarter than me and when i say ein stein i actually mean ein Stein as in german for one stone. It's a shape that can tile the plane infinitely without producing a repeating pattern.
In maths, we are excited about new things even if they seem to have absolutely no practical value or application. Sometimes, things become important later on, like prime numbers, which have been studied just for fun for centuries, and are now the backbone of encrypted communication.
So the only reason why this exciting is because nobody did it before.
I'm only a hobbyist in this space, but FOSS AI and its potential for individualized learning is a potential human existence game changer.
I'm exploring the potential to make a student companion that only knows the same information as myself as I explore a subject, and I'm looking for ways to essentially make a professor by uploading a large database.
Peripherally, I'm really curious what would happen if a LLM is trained on the Forth language. Forth builds exponentially with the programmer essentially building tokens. This seems very close to what a LLM is doing internally. It seems Forth is just obscure enough that it hasn't been considered by academia. Based on the little bit that I understand about both subjects, Forth could maybe take LLMs to a whole new level. If an LLM had access to a threaded interpreted language that self compiles tokens, it creates dynamic memory, a way to compress and expanded context, and the potential for adaptation and access. Forth is about as linear of a language as is possible. There are very few rules, and little arbitrary syntax in the language. Ultimately everything in Forth is possible as a single word, from machine assembly and register states all the way to entire operating systems, everything can be turned into a single word/token in Forth, and all words can be combined as needed at any point in the program. I really wish I was knowledgeable enough to explore this academically.
I don’t want to keep harping on about AI, but seriously. AI. I work in the creative sphere, and Adobe are (finally) earning their subscriptions. Lightroom’s new denoise DNG tool is massively impressive and not all that resource heavy. Photoshops generative fill solves so many problems with photo manipulation, my team is saving hours a week. On the video production side, crumplePop’s suite of audio correction tools is a game changer, and davinci resolves 18.5 beta has so much that we haven’t explored but I’m excited by.
Also, 32-bit float. Holy shit. It’s like recording audio in RAW. I no longer need to consider resourcing for audio production on smaller jobs.
consumer-grade field recorders can now capture in 32-bit float (I used Zoom F2s and have a Zoom F6), which reduces/removes our need to set levels beforehand or during. It effectively just has a ridiculous dynamic range.
I guess the technology has just recently made its way onto the scene for these sorts of products.
Rapid prototyping in general, and additive manufacturing (3D printing) specifically. I used to spend a lot of time prototyping by hand with foam, paper, tape, wood, clay, etc. We would have "arts and crafts week" at work when it was time to step out of CAD and make concept models of some new design. Occasionally some VP would walk by and wonder why the company was paying a bunch of salaried engineers to stick cardboard pieces together with hot glue. Getting decent parts often meant spending thousands of dollars on molding tools to make cast urethane prototypes.
Now I can walk down the hall to one of my 3D printers and get beautiful parts overnight. And there are a ton of companies offering rapid machining and sheet metal forming services. Getting design feedback and finding mistakes is so much faster and easier when I can go from 3D on a screen to real parts in my hand in a few days.
Additive manufacturing in the healthcare space is super interesting and has been really exciting to watch. The big players (Stryker, Medtronic, Smith and Nephew, etc) are rapidly approaching the intersection of multiple technologies that will make surgery outcomes wildly more successful simply by being able to customize the solution to the patient.
Consider how far just joint replacement has come in the last decade. Starting in the late 70s early 80s and the patients joint was cut to fit a standard implant regardless of any other factor. In the 2000s they started creating cutting blocks to form the patient's anatomy to an implant system selected to best fit the patient. By the 2010s they started using 3D printed, patient specific cutting block and jigs. Now they are printing the implants themselves. All of these advancements paired with robotics, real time computer assisted navigation, and a host of other diagnostic tools...mmm mmm mmm.
We may be living through the fall of humanity but at least our failing joints won't be the reason we die during the Water Wars of the 2050s.
I am watching some amazing things happen with social media decentralization that doesn't so much effect my work but my personal studies. Seeing this happening and wanting to be able to contribute to it, has encouraged me to try to learn web development again. This time I am going to go about it the smart way with thanks to The Odin Project. I am also keeping a keen eye on the burgeoning offerings of SD-WAN to the consumer market. Some really neat stuff is being contributed back to the community by Slack through their Nebula project.
It's an exciting time to study and experiment in FOSS right now. The internet is taking a dramatic shift back towards its decentralized roots. It's fun to be able to participate in ad-free discussions and without being shown content we largely do not want to see because an algorithm is steering us towards it to boost ad revenue.
I work in audio engineering, and while this is more of a slow burn, I think we're about to see a new generation of digitally enabled audio equipment hit the market. Audio-over-IP has been in use for a while now, but has been quite expensive, and adoption has been somewhat spotty. We're heading out of the chip shortage and shipping crisis, and that excites me.
My mixer for sure! An Allen & Heath SQ5, with two dx168 stageboxes. It runs a proprietary AoIP protocol, but it's SO easy to connect! The flexibility in routing is huge.
This mixer isn't new, but it seems indicative of what might be to come. It's a little expensive, (for what it is or honestly isn't though) but I truly believe we're overdue for a mixer with these capabilities for less. Or one for the same that does more. In price vs performance the X32 outperforms this, and nothing has touched that spot since its release, but that can't last forever.
The school i attend has some more cutting edge stuff that's more indicative of the future: stageboxes and monitor mixes run on the same network as the rest of the building, so I can connect them at any Ethernet plug in the building and get the same signal, it's so cool!
Yea we just placed an order for the new Yamaha consoles, everything on our is transitioning to Dante and with the release of DanteAV we could easily see all of our audio and Video networks streamlined and simplified.
I'm in the data centre industry, so the latest arms race for us is obviously AI/ML. Although it doesn't really directly impact what we do in the physical data centre space, the speed and scale of what we're expected to deliver definitely does.
I'm really interested in seeing what new and wonderful things people do with AI/ML. I don't really GAF about kids using it to cheat on their university assignments - I'm more excited to see how it gets used for some really cool shit that could benefit humanity. Medical science, and things like that.
Oh hey, I design substations and my major client is a data center powerhouse company. Nothing to say other than that.very cool about the AI thing though.
Kylie Jenner has a new tequila brand. The blanco is gross but the añejo is oddly tasty.
Oh, and the state is mandating that we get (discretionary) paid time off, but nobody's sure how that's going to work in an industry where most compensation doesn't come from base wages.
Fun fact: the vast majority of the millions of alcoholic products in the US are generated from the same few corps' distillates. To put it simply: "distillate" (the intended end-result of distillation) is not a saleable "product" in the US without first being "proofed", which in turn is little more than adding a certain volume of water to said distillate. The simple act of which satisfies the requirements for using the term "produced", as the distillate is now a product that can be marketed and sold.
In no uncertain terms, "produced and bottled in [place]" 99.99% of the time means "water added to, and bottles filled in [place]", and the entire implication that it was crafted locally, let alone in the same time zone, is a complete and baldfaced lie.
tl;dr: your favorite craft cocktails are actually Jim Beam, etc., and the only difference is what you tell yourself it is. 🤷🏼♂️
People using AI to rapid prototype concept art and reference material for 3D modelling on-top of using AI to generate meshes.
Chat GPT is almost exclusively used when it comes to code problems since it has similar hit and miss rate compared to searching solutions online yourself.