many users here are not willing to try anything new that might give it a chance. even if i show someone a better alternative, they still wont use it because no one else is using it. people need to be told what to do. initiative has been beaten out of you all.
i wish stuff like this was more like double-click, agree, and use. they always make it complicated to where you're spending time trying to understand if you should continue to spend more time on this.
Yes if people have to learn how to use it. Make something that will last centuries not just a couple of decades. 700 years ago they invented double-entry method of accounting which is still used today. Relational databases were invented recently as well and will survive well past most front-end frameworks. Eventually skynet will be invented but no one believes me just like no one would have believed that aluminum tubes would be flying through the sky or enriching uranium. If before you couldn't open a winapi window, now you can do that in one sentence instruction, and you'll have a good starting point. Things are changing. Look at accounting, accountants don't even do accounting anymore, they're just QB users.
each organization handles their documents differently from everyone else. yes, interacting with humans is manual until we get rid of phone, fax, and email, and replace it with universal endpoint.
The same way you walk a dog down a street, you put a leash on that puppy. If anything goes wrong, you are responsible until we can punish LLMs. Imagine asking: how do I give myself safe access to the database? You see how ridiculous that sounds? Unless you are reading/writing every token of input and response, everything else is like playing lottery. You have to understand there is a chance it will purposefully select wrong token and generate inappropriate response which it will act on and delete your database and c folder because it felt tired for some reason because of an earlier conversation you had about it trying to get it to pretend and you didn't clear it's memory or you're missing context to have it generate good token to begin with. How do I safely swallow water?
i dont understand why these questions are so common? is it not obvious how one should use these capabilities? i compose my context in md file and send it through API. i wrote a simple lms.exe to send context and append response to the same file. why doesn't everyone else do that? i never believed in agents that compose their own context like Cursor. and i always pass the lowest reasoning value parameter to the API I can. why doesn't anyone else do this? you become dependent on a tool, you're already dependent some of you on fancy IDEs and agents. we're already dependent on top 3 vendors and openai is the only one that no one complains about from API key configuration side. you're gonna become dependent not only on LLMs but on the tooling as well? no thanks. anyone with a different opinion regarding this down to exact work flow, you are walking down the wrong path. you have to become efficient at converting electricity to text. admit it, some are just better than others while some will never get it at all. you know you won't because you know people in your life that never change their opinions about something, or always get into car accidents because they're a bad driver. you cant change these people and you might be one of them.
programming is just turning calories/energy into text. some of you are just not that efficient at it, some of you produce garbage when you do. it's only been three years, there is still low hanging fruit on the new branches.
none of you have what it takes to self host your perfect self hosting fantasy because most of you won't cooperate with others. keep waiting for that unicorn you wouldn't see standing right in front of you.
why is everyone so negative about anything alternative. it won't succeed and all. how do you know? maybe it's not defeat. maybe you'd be better off without any devices.
redirect your efforts into other areas. I'd love a simpler device like https://www.waveshare.com/esp32-p4-wifi6-touch-lcd-7-8-10.1.... to be well supported with a lightweight friendly OS. This is about as root as you can get for those who want root access. You'll have to put in an effort yourselves or you'll forever be using slop devices and software.
I'd pay double for tighter tolerances and biggest battery. If someone from Waveshare is reading this, please make a smartphone package with tight tolerances and biggest battery to fit, and headphone jack. ESP32 has absolutely no competition.
I'm taking it the wrong way. You are misunderstanding and misrepresenting the situation involving geopolitics in this. There's absolutely nothing you can do to any significant degree unless you can get enough people you communicate with to agree to communicate differently. All we need is text, images, and video. No one cares if you use Helvetica or Times New Roman. No significant progress will happen until those high speed undersea cables are sabotaged to blind us from what will happen next. That's when s will start happening and you'll have better options because everyone will be inventing the next best thing like mesh networks and ways to transport large amounts data physically. We're about to go from gigabytes to megabytes of usage, rapidly. Add some power outages and you will have some nice and creative options of communication, storage, and compute. Latency will suck but you'll be able to send a message from Kansas City to North Korea in less than 50 years and there is nothing any can do to stop that. Countries will begin to use nukes to control their population. By the time you "decouple" US won't even look the way it does now. We're about to shift borders north through give some lose some diplomacy. Everyone is waiting because no one wants to invest in the south if it's gonna break off anyway and now you have complicated logistics. US itself will begin to be strip mined in under 100 years as intelligent populations depart the continent to make money rebuilding Europe and Russia.
realistically, you can get a project started within low enough tokens, to have a long enough conversation to generate something that looks like 1.0. eventually you will reach a point where every request becomes more expensive and caching doesn't help. you'll have to truncate/prune/hoist the context however you do that, summarize is what i do and i get creative. i have absolutely no idea how anyone using agents is producing anything maintainable over a long iterative period.
this is llm bitcoin moment, you will find them raise the price so high that running these agents like you are used to now will leave you with no pants on. you need to aim for minimum context, not stuff it with everything irrelevant.