carpo

Yeah, I think we should try to separate his business acumen from his nut-case personality. He can be a successful businessman AND a fucking prick.

Use /context <prompt> to have aider automatically add the files based on the prompt. It's been working well for me.

I was thinking the same while was walking my dog.

Yeah, I've found it great at XML too. All sorts of boilerplate stuff works so well. As it the project gets bigger Ive had to think for longer about things and step in more, but as you said, that's usually the fun, meaty bit that you want to work on anyway. My app is about 15k lines of code now, and it's a bit more work than at the beginning.

I agree, but only to an extent. For me, the passion changed over time. I used to love getting an O'Reilly tome and learning something new, but now I don't really want to learn the latest UI framework, library/API or figure out how a client configures their DI container. If the AI can do most of that stuff, and I just leverage my knowledge of all the frameworks I've had to use, it's a huge timesaver and means I can work on more things at once. I want to work on the core solution, not the cruft that surrounds it.

I agree though that the Studio Ghibli trend feels off. To me, art like this feels different to code. I know that's probably heresy around these parts of the internet, and I probably would have said something different 15-20 years ago. I know that coding is creative and fulfilling. I think I've just had the fun of coding beat out of me over 25 years :) AI seems to be helping bring the fun back.

That's awesome! That's a lot of data and a great speed increase. I think that as long as you test and don't just accept exactly what it outputs without a little thought, it can be really useful.

I take it as an opportunity to learn too. I'm working on a video library app that runs locally and wanted to extract images when the scene changed enough. I had no idea how to do this, and previously would have searched StackOverflow to find a way and then struggled for hours or days to implement it. This time I just asked Aider right in the IDE terminal what options I had, and it came back with 7 different methods. I researched those a little and then asked it to implement 3 of them. It created an interface, 3 implementations and a factory to easily get the different analyzers. I could then play around with each one and see what worked the best. It took like an hour. I wrote a test script to loop over multiple videos and run each analyzer on them. I then visually checked the results to see which worked the best. I ended up jumping into the code it had written to understand what was going on, and after a few tweaks the results are pretty good. This was all done in one afternoon - and a good chunk of that time was just me comparing images visually to see what worked best and tweaking thresholds and re-running to get it just right.

I've been thinking about this a lot and agree. I think the UI will change drastically, maybe making voice central and you just describe what you want done. When language, image and voice models can be run locally things will get crazy.

If you do want to get more into it, I'd suggest something that plugs into your IDE instead of Copy/Paste with ChatGPT. Try Aider or Roo code. I've only used Aider, and run it in the VS terminal. It's much nicer to be able to leave comments to the AI and have it make the changes to discrete parts of the app.

I'm not the OP, but on your other point about completing sprint goals fast - I'm building a video library app for myself, and wanted to add tagging of videos. I was out dropping the kids at classes and waiting for them. Had 20 minutes and said to Aider/Claude - "Give me an implementation for tagging videos." It came back with the changes it would make across multiple files: Creating a new model, a service, configuring the DI container, updating the DB context, updating the UI to add tags to videos and created a basic search form to click on tags and filter the videos. I hit build before the kids had finished and it all worked. Later, I found a small bug - but it saved me a fair bit of time. I've never been a fast coder - I stare at the screen and think way too much (function and variable names are my doom ... and the hardest problem in programming, and AI fixes this for me).

Some developers may be able to do all this in 20 minutes, but I know that I never could have. I've programmed for 25 years across many languages and frameworks, and know my limitations. A terrible memory is one of them. I would normally spend a good chunk of time on StackOverflow and the documentation sites for whatever frameworks/libraries I'm using. The AI has reduced that reliance and keeps me in the zone for longer.

I guess it depends why you're writing the code. I'm writing a local video library desktop app to categorise my home and work videos. I'm implementing only the features I need. No one else will use it, I'll be finished the first version after about 4 weeks of weekend and night coding, and it's got some pretty awesome features I never would have thought possible (for me). Without AI I probably never would have done this. I'm sold, even just for the reduction of friction in getting a project off the ground. The first 80% was 80% AI developed and the last 20% has flipped to 80% coded by me. Which is great, because this part is the meat of the app and where I want most input.

I'm the complete opposite. After being burnt out and feeling an almost physical repulsion to starting anything new, using AI has renewed my passion. I've almost finished a side project I started 4 weeks ago and it's been awesome. Used AI from the beginning for a Desktop app with a framework I'd never heard of before and the learning curve is almost non-existent. To be able to get the boring things done in minutes is amazing.

"... it's just a matter of time until someone else steps in to exploit that feeling for their own gain." Someone just did ...

I'm listening to Brain Energy by Christopher Palmer and he talks about this.

This is exactly how it use it too

aider looks amazing - I'm going to give it a try soon. Just had a question on API costs to see if i can afford it. Your FAQ says you used about 850k tokens for Claude, and their API pricing says output tokens are $15/MTok. Does that mean it cost you under $15 for your Claude 3.5 usage or am I totally off-base? (Sorry if this is has an obvious answer ... I don't know much about LLM API pricing.)

Why does it matter how someone achieves this? If it helps them and doesn't hurt anyone else, your view on the "correct" way to achieve it is really only you virtue signalling.

E a d

Even before WFH, meetings with people in different offices occurred. There's a much better range of software options now too.

Am I working on a Wordpress site ? Yes -> MySql No -> Postgres

You sell it for money

There are many things in science that scientists don't understand but still believe exist. The difference is the scientist won't then ascribe that thing to a god just because it's not understood.

I can guarantee that all of those people at some point in their lives hadn't heard of it. People have to learn about it somewhere, why not here?

I don't think you need to understand what suffering is to feel it.

Well, the Government could still be the one who writes the smart contract (or commissions one or uses a global standard) and has oracles with all the data required. I know many purists wouldn't like that, but I'm just bullish on the technology, and see it going further than NFT images and the FUD that people keep arguing about, i.e right click save and ponzi schemes. There's a use-case for this tech, and it's in discovery mode at the moment. Writing it off now feels a little premature.

Thanks for enlightening me with your knowledge of the blockchain. but please explain how would you "register" it? Why with my "favourite" blockchain? Why can't a centralised state use it, but get the benefits of decentralisation and proof of ownership? Your arguments and lightly veiled ad-hominem attack are actually the point of my initial comment.

Imagine that Country A decides to use Blockchain A (which has been around for decades and has proven itself secure) to store its property titles and allow the purchase of property digitally, with full transferal of title and ownership. The government writes a smart contract on that blockchain that facilitates payment, title searches, hooks into other government databases or blockchains for easement information, liens etc. They then, obviously, migrate their existing title information onto the blockchain and distribute the NFT's to the current owners. How do you "register" your fake one?

If you can't see how this could change an old and plodding industry like real estate which is surrounded by mountains of paper-shuffling, then yes, you're being unimaginative.

Buying a house is the same. It just adds your name onto a title document that is stored in some custom government database. Imagine if there was a global way to register property title and ownership. Imagine if multiple countries used it. How do you think that could change the property market - or any other market that is currently geo-locked or database locked or company locked? We're talking about a decentralized, trustless and immutable proof of ownership for any good. I really don't understand why people focus so much on the images, gaming etc part of it. This tech is just starting with images - it's finding it's feet, perhaps frivolously, but it will end up changing the world when it's used for other things. It's a little shocking actually to see the lack of imagination on HN about a technology that could revolutionize multiple industries.

Last time I bought a house we needed a solicitor or conveyancers to do a title search. That info could be on a blockchain as an nft amd a smart contract could handle it.

Poor people.

That's a bizarre comparison considering the prevalence of Object Oriented programming languages and the ubiquity of that style of programming.

You could have reduced the risk substantially by transferring off their wallet to yours right after purchase. You still could have purchased your weed too.

For some people giving up an addiction is an ongoing battle, not a one time and done thing.