Eliminating programmers will be possible when we figure out how to eliminate engineers in designing buildings.
Only a true AGI will be able to do that, and while LLMs feel like a step towards AGI, they are still missing the critical ongoing learning component that needs to happen for an AGI to exist. The way the current systems are trained simply doesn’t allow for accepting and adopting new information continuously.
A lot of ai hallucinations can be resolved by simply running the results through additional prompts automatically, then checking the various results against each other or against reference material.
Many agentic systems already do that with a limited number of follow up/check steps, but they’re often restricted by acceptable response times or just sheer costs.
I managed to get copilot in excel to run a 43 prompt chain in just a little under 10 minutes the other day. The result was exactly what I needed.
If you have 73 times the output, you can potentially afford to do that kind of processing in an acceptable time frame and cost level.
That was a server side compromise, but the update function could have detected it with more security features so you’re still right.
Nothing is stopping it, it’s just not particularly convenient because it’s designed around the limitations of the phone system.
SIP could handle it all if you wanted though.
Then then answer to your question is yes, you interpreted it incorrectly. The original (and common) definition of that word is vastly different than what you assumed.
Do you know what a semaphore is?
Quite literally designed for long distance communication.
Used extensively in many different variations before wire/electricity based systems became available, then even used after that for communication in places where you couldn’t have a wire (like between boats)
The number one google result is Owncast.
Looks like it doesn’t save the video by default but there are workarounds to record it.
To add to this a bit, if the site is just a single page that will be fine. If your site has a bunch of pages that all need to share the same theme and headers and footers and such it may be worth learning a simple templating system that runs before the site is sent for hosting.
That way if you need to update something in the design or theme you can do it once, rather than on dozens of pages.
Personally I use eleventy for my little site.
You can absolutely use your external IP rather than a domain, your ISP likely keeps your IP the same for long stretches of time. Mine only ever really changes during long power outages.
You can get around that the IP issue by broadcasting your ip from your local machine to a known service, there are some free ones but they aren’t the most stable things in the world if it’s critical. Google Dynamic DNS
I just have it running in a docker container…
Improving your input, and the system message can also be part of that. There are multiple optimizations available for these systems that people aren’t really good at yet.
It’s like watching Grandma google “Hi, I’d like a new shirt” back in the day and then having her complain that she’s getting absolutely terrible search results.
I’d argue home assistant with some smart LEDs and a few sensors would be great.
Having a bulb that let’s you know the outside temperature/weather when you’re getting dressed in the morning is neat. Having a dimming pattern for sleeping time. Tons of other really simple stuff available too.
I mean, the text on a website isn’t the problem for not being able to use 56k.
It’s only images and video that take up space, the libraries used on websites are all cached at this point so that’s hardly relevant to ongoing usage of a website.
I hear theres a few neo Mennonite colonies starting up that only use tech before the turn of the millenia. Maybe you could look into them.
And the vast majority of my automations don’t use it because it isn’t required.
I will stand by what I said, the tool is great even if you don’t touch AI at all with it.
The comment was wrong. It’s an automation tool. It can use ai, but its not the primary focus.
“There is zero chance any of these investments are going to turn a profit.”
This isn’t accurate. There’s a zero chance ALL of them turn a profit, but there’s actually good chance that one or more of them will return a huge profit.
If I invest $100 in a hundred companies, and 99 of those companies fail, you may think I’m a terrible investor. If that 1 company returns $50,000 on my investment though, I’m actually a fucking genius.
That’s how venture capital works.
It really just helps in cases where you get hacked, but the hacker doesn’t have continued access. Say someone physically penetrates into your building, grabs the key through an unlocked station, and leaves.
That being said, like you mentioned, if someone is going through this effort, 45 days vs 90 days likely won’t matter. They’ll probably have the data they need after a week anyways.
Encryption key theft really requires a secondary attack afterwards to get the encrypted data by getting into the middle and either decrypting or redirecting traffic. It’s very much a state level/high-corporate attack, not some random group trying to make a few bucks.
Rag was never supposed to be about learning over time. It was supposed to provide better context at inference. It could never scale to handle new learning beyond focused concepts.