I think svelte 5 is an actual innovation, it does not behave or feel like a “framework” to me. It feels like a natural extension of building an front end with raw html, css, js, solving the headaches you eventually hit when you do so.
Optimist take, the tech is shit and the people like Trump not JD Vance.
Cryptocurrency = Not being used as a currency, doesn’t even meet the criteria for a true medium of exchange.
AI = Can’t actually be used to automate these federal worker’s jobs, it’s snake oil.
Most importantly, JD Vance, Elon Musk, Peter Thiel, although quite powerful and influential right now, are NOTHING without Trump. And as michael Bloomberg showed us, money alone doesn’t win elections.
Seriously this is a very well written article and the concerns are real, especially around the degradation of our institutions over the past 50+ years. But if you put JD Vance, Elon, or Peter thiel in front of your average American, the response will be a more derogatory form of “neurodivergent freak”.
In 1-2 years I’m predicting democrats take back the senate and house, 4 years we get a younger more normal democratic president. Hopefully most of the old neo lib democrats die off and we get more smart charged up dems like AOC, Buttigieg, and the fella from Georgia who will actually tell these rich bastards to stuff it.
Svelte 5 does a great job of raising warnings for this. You try to use an onclick on a div you get warning to make it tab-able and give it an aria role, and a reccomendation that you should probably be using a button.
I am saying that all people, including neurodivergent people, have personal agency in what kind of work they choose to do. If you work a job that is primarily meaningless work, that’s your choice. My original comment was to say that you have a choice to not work dull jobs, and it is imperative as a nd person to not do it as you will lose. If instead you take a little personal responsibility for your life’s direction, and do work that plays to your strengths, work is easy. Probably true for all people, but absolutely essential for thriving as a neurodivergent person.
Ive got a pretty intense ADHD. This article is vaguely offputting to me. I think it makes it out to be that neurodivergent people need to be baby’d in some way.
From my perspective, I think neurodivergent people can’t stand doing things without purpose. A lot of jobs are truly meaningless, dull, uninteresting, especially in the office space. People with ADHD and ASD should not be doing these jobs. You need to find your strengths and interests and work towards that. You should never work a job that you deep down think is bullshit as a ND individual, this is the equivalent of a death sentence. Find something you are passionate about and you can do better work then any normie no matter how loud or fluorescent the office is.
This is actually a major issue in the LLM wrapper space. Building things like agents (which I think are insanely overhyped and I am so out on but won’t elaborate on), usually in Python, where you are making requests that might take 1-5 seconds to complete, with dependencies between responses, you basically need to have expert level async knowledge to build anything interesting. For example, say you want two agents talking to eachother and “thinking” independently in the same single threaded Python process. You need to write your code in such a way that one agent thinking (making a multi second call to an llm) does not block the other from thinking, but at the same time when the agents talk to each other they shouldn’t talk over eachother. Now imagine you have n number of these agents in the same program, say behind an async endpoint on a FastAPI server. It gets complicated quick.
I think database modeling is a bit of a lost art, it’s so important though for building applications that do anything important. Love your distinction about the database not needing to be 1 and the same with your domain model. I usually do my first iteration with a few json columns then split them out into tables as the domain model comes into focus … it’s kind of an art.
And he just secured a massive salary increase for his constituents, in short time. As a member of the labor class of society, I can’t help but cheer him and the union on.
It’s not wrong to use dicts, it’s just bad practice when you could use something like a dataclass or pydantic model instead.
Dicts are useful for looking things up, like if you have a list bunch of objects that you need to access and modify, you should use a dict.
If you are using the dict as a container like car={“make”:”honda”,”color”:”red”}, you should use a proper object like a class, dataclass, or pydantic model based on whether you need validation, type safety, etc. This drastically reduces bugs and code complexity, helps others reason about your code, gives you access to better tooling etc.
This looks great, actually using streaming for responses is a nice touch.
Having built a few front end ai chat features, one of the key things is making sure the llm provider is swappable, and that chat history is stored in your own database.
The architecture I typically use is chat frontend (usually react) -> custom llm wrapper api -> openai/anthropic api | my own db for chat history .
Would this be able to support calling my custom llm wrapper out of the box?
Exactly, things like changing the signature of the api for chat completions are an example. OpenAI is looking for any kind of moat, so they make the api for completions more complicated by including “roles”, which are really just dumb templates for prompts that they try to force you to build around in your program. It’s a race to the bottom and they aren’t going to win because they already got greedy and they don’t have any true advantage in IP.