Ask HN: What excites you today (technologically speaking)?
My first exciting thing was my Commodore 64. I loved playing games with friends, I loved typing code from magazines, I loved creating sprites on paper and dreaming of giving them life on my Commodore 64.
My second exciting thing was my first modem and the first connections to the local BBS in my town. I loved chatting with people, I loved downloading JPGs (well, you can imagine what kind).
My third exciting thing was the internet. This was an evolution of my second exciting thing, it was BBS on steroids. I loved visiting my local newsstand and buying the new issue of .Net magazine. The final pages always included a list of cool new websites to try.
My fourth exciting thing was social media. I loved keeping in touch with friends on Twitter, Friendfeed, Facebook, etc.
My fifth exciting thing was Bitcoin. I got lost in the rabbit hole, having fun mining with my computer (when it was too late to make money), spending hours reading posts on btctalk, buying my first Bitcoin from a guy at a coffee shop for cash, mining with my first Butterfly hardware (again, not enough to make a profit), building my Raspiblitz, staring at Blockchain.info waiting for the next block to be mined.
It has now been quite a while since the last time I was excited about something. The top excitement for Bitcoin was maybe around 2015. Now it's just all about money money money on crypto or developments that are way too complex for me to understand.
What is exciting you nowadays? Any new technology, any new website, any new cool thing going on in the tech world?
126 comments
[ 3.8 ms ] story [ 76.0 ms ] threadIn web development in particular I hope this will lead to a reexamination of lost ideas, such as the hypermedia architecture. I think you are seeing that with things like Hotwire and htmx.
So I would look to ideas from the past, perhaps abandoned today, for things to inspire us.
Otherwise, I'm getting pretty stoked about computer vision techniques. If I had more time I would work on projects for controlling basic things within the home. I saw a Youtube video which demonstrated a volume control system, controlled by a camera looking at the distance between your thumb and index finger. I also think many of the techniques in CV have applications in art which has been completely unexplored.
Blockchain was cool and novel, but the hype should be over already. We’re nearing 15 years of blockchain tech and it is still not magic powder that solves society’s problems.
Its not a magic bullet for society's ills, but its going to be a game changer if people really want the games we play to be more equitable and impossible to hijack.
It costs money, but it's now possible to live a first-world lifestyle - sports car and all - without burning any fossil fuels. Houses can be made self-sufficient. The biggest impediment to turning that into a massive distributed green power grid is regulatory.
We'll need that grid, so I'm confident the walls will come down. Electric cars are widely accepted as the future, with many countries attempting to ban ICE cars completely within a relatively short timeframe, but nobody seems to be strongly considering how on earth we'll charge them all. There's going to be a rocky period of power shortages, but that will incentivize the construction of off-grid homes, which will incentivize ways of arbitrating their surplus power. The future of this space looks interesting.
An offshoot of all this that's exciting from a social perspective is the rise of personal electric transport, like e-scooters and e-bikes. We're well on our way to eliminating fossil fuels from our cities, and that makes me happy.
I'm not particularly well informed but I believe Barcelona city has made some steps on this by incorporating a jointly-held power company to add hundreds of charging points around the city. Public infrastructure achieved but via private investment.
More here - https://www.endolla.barcelona/en/
Plus, I really like Hetzner
This fascinates me because people seem to have forgotten that the reason personal computing was such an incredible thing is that it freed people from that old paradigm. I suspect that, over time, people will lean anew why it was that being freed of it was such a great thing.
The most offensive thing of all is that modern dumb terminals don't have keyboards anywhere near as good as the old days.
This whole bite and switch let-move-on-cloud religion must end.
I think “human scale” technologies can help mitigate some of the more nightmarish horizons of the technological society we inhabit, though, obviously, neither completely, nor on their own.
My background is in networks, so I tend to think about things from that perspective (e.g., a private U-LTE network for communication with neighbors, mesh nets of sensors to make home food production more manageable and efficient). It’s a very fruitful area for anyone interested in a more communal and family-oriented future.
Obvious difficulties are: Is the efficiency hit one gets from decentralization practically viable, long term? In which cases? How do you get your silicon? Other materials? Are those suppliers going to let you do this? How do you do this in the existing regulatory and political climate? Can this work for the poor? Does it open, unintentionally, new frontiers of technological domination?
All interesting questions; only some have technical solution.
EDIT: Adding also that I am interested in new or revived applications for “low-tech,” if that’s something anybody else knows about and wants to share.
One is that advances in personalization on the Internet haven't carried through to physical goods. When I want to curate my Facebook or Reddit feed, I can very tightly control the information I consume so it fits my lifestyle perfectly. When I want to buy a piece of furniture or shelf on Amazon, I get stuck in this uncanny valley where there are millions of products available but none is exactly the size, shape, color, and material that I want. Why can't I say "I want a double corner wall shelf, 23" on one side and 29" on the other, 8" deep, filigreed supports, made out of pine and painted to match my walls"?
Another is that manufacturing is increasingly labor-free and computer-controlled anyway. In a factory, there's going to be a bunch of CNC machines, computer-controlled sawmills, maybe some injection molds, 3D-printers for prototyping, pick-n-place machines, etc. Most of these are computer-controlled anyway, with humans only needed to feed & adjust the machines. Could you computer-control a home or neighborhood machine instead, so that people only need to download a blueprint from the Internet or make it themselves? Why do we need such big production runs, if computers can reconfigure the manufacturing without any human labor? Why not have people buy plastic filament, scrap aluminum, scrap stainless steel, OSB or plywood or 2x4s, and then just feed the machine with a pre-built software blueprint?
A third is improved new manufacturing technologies, particularly 3D printing and pick & place machines. It feels like these are still stuck in existing paradigms, trying to fit into the mass-market industrial production system rather than experimenting with novel combinations on their own. For example, what if instead of P&Ping just electronic components on a circuit board, you used it to assemble individual plastic, metal, and wood parts that had previously been 3D-printed, and then 3D-printed joints to hold them in place? Could you use miniaturized CnC lathes to smooth down the surface of a 3D-printed part, which has traditionally been one of the big problems with 3D-printing?
Then there are environmental problems with supply chains being fragile and globalization potentially unwinding. That could provide an extra kick to hyper-localize manufacturing again.
Micro-manufacturing is the real Amazon killer, potentially. I can't see them being displaced in retail now. But if we just stop buying manufactured products and start making them ourselves, all of their advantages in supply chain management, bargaining, product selection, and logistics go away.
The way the marketplace is organizing, most local hospitals are being purchased and re-organized into large regional state-wide networks. At this scale I wonder if it might become economically feasible for a hospital system (or systems) to invest in getting open-source designs through the approval process and then have an in-house engineering department that could manufacture parts for the regional system.
Open source tools could be a boon to civilization though, especially in developing countries - one problem is materials though. Some optics projects are really clever, like that visual microscope malaria diagnostic kit with the glass ball lens. But then again, you think: hey, are real microscopes really that unaffordable? I really have to read up on how that was financed and why that road was taken.
EHR is sort of similar. To me, the roll-out of that was a disaster, and pushed what was in-house in most cases to being managed by EPIC and other EHRs out of the hospital. The mandates were a big mistake in my opinion, as it forced hospitals to scramble to use something being offered by outsiders, instead of collaborating to produce something open-source, or growing EHRs more organically from within the organization.
I personally blame the rise of hostageware in hospital settings partially on this trend.
My broader point is that although I think there's a lot of potential with open source hardware, software, and things like 3D printers, prevailing economic forces are pushing in the opposite direction. Consolidation and mergers, streamlining everything that doesn't contribute to increasingly dense profits as you go up the administrative chain. In this schema, better to outsource everything you can to trim costs. I don't agree with it, as I think it leads to a lot of hidden costs and hidden but lost benefits, but that's the idea.
Unfortunately, the combination of overregulation and profit-driven hierarchical management is creating pressures against in-house, from the bottom up creation of goods and medical services. The talent is there, it's just pushed out from the top.
Sometimes I feel like healthcare and the biomedical area is today driven more by the interests of profiteers than patients/clients/customers.
That gives you a design that the FDA is happy with.
Now you have to build, distribute, service and support that device and no matter how you slice it, you're looking at substantial costs to comply with 21CFR across all these tasks. So this is where the creativity really has to come in: can we spread those costs across a "community" to make it worthwhile, or will we just end up right back at Square One with single use, $1,000 devices?
I don't think there's a path forward (at least in the US) without change to regulations.
The present day headsets are still bulky, but am sure they will become lighter and even more better.
The interface will also improve. In oculus we can already use fingers to navigate and interface with the OS. This will only improve.
I bought an Oculus Rift CV1 and it was an insulting experience. Huge amounts of enforced updates, internet connection abuse, and general Facebook shenanigans. Fool me once... I will buy one when it's an open platform or standalone peripheral. Not before.
As for open standalone platform, Lynx will soon become available. https://www.kickstarter.com/projects/stanlarroque/lynx/
Would you also pay double that amount? That's how much it costs where I live and probably in many other places too. And you can say what you want, even $300 is not an amount I want to risk going down the drain because Facebook had yet another wave of "oops you're banned now too bad" events.
I will not even touch anything Facebook for free nowadays, to be honest. It's only a matter of time until great alternatives exist.
For my entire life (born in 1972, same timescale) Intel has managed to keep a leash on what personal computers could be. The few times it's slipped have led to exciting advances, but they got them under control again quickly.
The explosion of held up potential that will happen when they shatter should (I hope) make the AT&T breakup look small.
I can't do justice to the examples of bad faith marketing they've indulged in, anti-competitive acquisitions of promising companies that then are buried, architecture choices made for marketing gains that saddle programmers with years of boneheaded bullshit, delightful tricks like "optimized" compilers with "oopsie" slow code paths for non-intel cpus...
Perhaps there's a good intel bashing thread someone could reference?
As to how things will change, we're seeing some of it already; we've got multiple micro-controllers on the market and cheap single board computers that do everything a desktop needs. I'm expecting some backplane bus to take over soon and the definition of what a computer is to become even harder to nail down.
I'm down for open source architectures, but our manufacturing technology is not ready to support DIY processors yet, even at low speeds. Solving the manufacturing process could go a long way to spurring competitive processor architectures.
The Creatures Evolution Engine source code (if anybody remembers the Creatures games / Cyberlife) getting released. Now we'll know exactly what makes those Norns and their world tick, which is fascinating: https://archive.org/details/lc2e-sun-16-jan-2000.tar01
All of the Infocom source code and information about ZIL and the stories from the founders.
The LambdaMOO server getting forked and modernized with Stunt and ToastStunt. Now there is even talk of the original server getting updated again after 20 years of the maintainer doing nothing with it.
Fun stuff!
(Yes, they excite me too! Have an upvote!)
Edit: Oh, and having 2,916 front-end frameworks to choose from. Yay.
AR - I was blown away by the AR comic book reader developed by VeVe for their NFT comic books. I still think that NFTs are lame (at least their current format), but being able to see through your iPhone an old comic book sitting right there on your table and flipping its pages was something magical. Never before I wanted so badly to have AR glasses with me and just being able to read a comic book collection in this way, without the need of my iPhone.
Robots - I was initially excited about the Amazon Echo, but I ended up using it only to set a timer when I cook. The new Amazon Astro is exciting. Way too expensive right now and I am just scared it might be a huge disappointment, but the idea of having a robot in my house that follows me, helps me, etc. sounds like super fun and very exciting.
You may have enjoyed mining Bitcoin, but there’s more to blockchain technology than that. I’m surprised you didn’t mention NFTs or dapps. Search for Web3 if you’re into playing buzzword bingo.
A part of Oculus may have sewn the seeds of fascism that we recently witnessed and are experiencing the effects of as we attempt to squash it back under the rock it crawled out from, but their corporate overlords still feel strongly enough about it to announce they’re dropping a considerable chunk of cash to swindle everyone into thinking they’ve created a “metaverse” instead of another walled garden. There’s a ton of work happening in the VR/AR/XR space that is approachable. Look beyond gaming to AEC and manufacturing applications.
The new 5G phone books are coming! Get ready for that.
Solar and alternative power is only getting bigger. We’ve experienced some technical difficulties getting a certain segment of the population to transition away from fossil fuels, but if we can get an electric car into space, we can hopefully get one in NASCAR.
And I’m sure you’ve seen all the robots, right? The Roomba folks have a nice platform, but there are many others out there. Add a camera, a laser, and a bit of deep learning into the mix and you’ve got yourself hours of entertainment.
Go back and watch Douglas Engelbart’s mother of all demos and you’ll see that some things have progressed further than others. Why is that?
Find what interests you and dig deeper. Make something that works for you and show others the way.
In fact, in general, make it a habit to revisit 60’s, 70’s, and 80’s technology. Look up old computer magazines, read some articles.
There are TONS of ideas that were pioneered back then that have either been forgotten, have not fully evolved to reach their potential yet, or have evolved in negative ways such that they “lost the thread” (and therefore, provide fertile ground for revisiting and iterating upon).
One of the most monumental mindset shifts I’ve had in my career was looking into the whole Xerox Parc Smalltalk stuff. The concept of a live environment that you can change anything about it, right then and there, was a magical lightbulb that got me thinking about the UIs I build and the data behind them in new ways. It shifted my mind towards user empowerment rather than tight controls and structures.
Rediscovering some of these core ideas in their pure, early forms can be really inspiring.
I saw a comment that said it feels like a game from 5 years in the future and I agree.
VR/AR in general feels like a young field with lots of room for innovation.
People may say the same about the computer platforms of the old days, but i think that we're still seeing a pretty good period of time before the larger walled gardens have taken over. You can have a server in the cloud up and running in minutes. There are even managed services or PaaS offerings, if you'd prefer to use those and don't fall in the SaaSS trap: https://www.gnu.org/philosophy/who-does-that-server-really-s...
Want to do embedded development? Arduino and the compatible platforms are lovely! Need something more powerful? Raspberry Pi and the compatibles have got your back! Even more so? Just get a low power x86 CPU like 200GE and it'll be more than you need for the near future! Of course, you can also do your part in decreasing e-waste and use cheap refurbished hardware as well, perhaps even buying professional grade stuff, like the people over at https://www.reddit.com/r/homelab/ community often do.
Need a software library or a language to help you with a particular task? It probably not only exists, but also has documentation and even tutorials available at a whim! You don't need to read magazines or manually copy code, you can download ready to run examples from GitHub, or even view them without leaving your browser! And the variety of languages is also lovely, anything from Python with its rich ecosystem, Ruby and PHP for simple webapp development, to Java, .NET for more serious platforms, Go and even Rust for working at a lower abstraction layer.
Need a larger piece of software? There are self-hosted platforms for blogging, sharing files, even e-commerce stores that you can host on your own. Most of those are also open source and free. While the licenses vary, in most cases you can modify the software to suit your needs, or to even offer fixes that may benefit thousands or millions across the globe. And on the opposite side, you also benefit from the work of others as well! As for the complicated domains, there are businesses to address your needs. Want to take payments? Stripe or Paypal has got your back.
And even within these walled gardens, things are mostly acceptable for now: you can host videos on YouTube, reach your audiences on social media or even use platforms for app delivery. That's not to say that they're perfect, but 20 years ago you simply didn't have anything like that. I recall one of the GDC talks about how back in the day people had to order video games from a small company by phone, which nowadays seems as curious, as it does unnecessary.
Oh, and also the FOSS and open source movements in general are amazing to behold. You get entire production ready operating systems like Debian or FreeBSD for free. They even probably run on the hardware that is in your home! And they can scale from a laptop to a server farm with few to no issues! Even driver support is improving and you also get a whole bunch of amazing free software: everything from LibreOffice, Firefox, GIMP, Krita, Blender, Audacity, kdenlive, VSCodium/NetBeans/Eclipse/IntelliJ (though personally i pay for the package of all JetBrains tools) to even game engines like Godot.
On a technical level: containers and software that compiles to small static binaries. In my eyes, both of those approaches are good for achieving higher levels of environment independence and if software is written with 12 Factor App principles (https://12factor.net/) in mind, or just follows some of the UNIX best practices, then it's likely that it'll be reasonably easy to configure and run.
For example, in my homelab, i run about 90% of the software in containers, givi...
Two things:
1) The new space age, really aiming far to the Moon and Mars.
2) The altered state of mind that wide acceptance of Psychedelics is going to bring to the masses.
Xlights: Sequencing and Scheduling software. This is what tells your lights what to do. (https://xlights.org/)
Falcon Pi Player (FPP): This software runs on your controller - A raspberry Pi or a Beaglebone are quite popular. This is what you run your sequences on. In my case, I drive my lights directly from my Pi. (https://github.com/FalconChristmas/fpp and https://falconchristmas.com) Also worth noting: A lot of people use dedicated hardware for this, such as the Falcon line of controllers (https://www.pixelcontroller.com/)
WS2811 Strands: These are probably the most popular lights - they're individually addressable, and pretty dang cheap for what they do. Even cheaper if you plan far ahead and order from China. This link is to my preferred brand, but lots of companies make these. The trick is to choose one company and stick with it, as they'll use the same LEDs across their products and you get good color matching (https://www.amazon.com/BTF-LIGHTING-Diffused-Individually-Ad...)
AusChristmasLighting: I live in the States, but for me this is the best of the online forums for lighting. If for no other reason, the UI is substantially nicer thna others I've come across. (https://auschristmaslighting.com/)
My own blog post about getting started and setting stuff up: (https://aaroneiche.com/2020/12/27/holiday-lights-display/)
Good luck, and if you have questions, feel to ping me.
https://github.com/WebAssembly/design/issues/1397
https://github.com/pannous/wasp/wiki/
As you pointed out true GC is currently postponed, so no long running processes yet. The hope is to get the language finalized just in time when all wasm 2 features become standardized