I remember .NET fails when there were dramas about not open sourced debugger, planning to remove hot reload (make it only available in Visual Studio because of licensing) or poor support in Visual Studio Code compared to other languages. AFAIK many people were angry that VS Code is not first class editor for C#/.NET. I am not sure how it is these days.

C# Dev Kit is also receiving huge performance improvements with .NET 11. It's available now on the pre-release version of the extension, if you want to check it out. Some details in the .NET Unboxed video here https://m.youtube.com/watch?v=5fvi7m1QxIY

The situation about debugger didn't change really. https://github.com/dotnet/core/issues/505 At least they didn't close the issue...

https://isdotnetopen.com/ has a good list of "dramas" questioning if .NET is truly open. At least there are not many new dramas from Microsoft...

For me it's not that big of a deal that some parts are not open. At least we have good competition (Microsoft (Visual Studio, VsCode c# Dev Kit) vs Jetbrais (Rider, VsCode + recent "Resharper for VsCode" extension.) that prevents one party to mess up something big.

Debugger was there, hot reloading was the issue.

Still I take debugger in Visual studio or Rider any day instead of VSCode debugger.

> Debugger was there, hot reloading was the issue.

Was it? There was only Samsung debugger.

it wasn't. Jetbrains can't legally use .NET's debugger, only Visual Studio and Vs Code can.

Yes, there is a funny PR in open source project about refactoring source code to assembler code or even machine code 0s and 1s. I do not remember exactly but I remember I LOLed hard when I saw it.

I think yes because there are people who prefer picking up something rather than building their tool. Recreating just everything from scratch or make custom solutions are not the one way to go. I do not want to create editor/IDE because it is not in my interests. I choose VS Code, Kate, Notepad++, Nano instead. If I have paid LLMs I will use them different way than reinventing editor lol.

Also keep in mind that it is software which requires a little bit maintenance from time to time. And access to LLM (mostly paid) and (or - depends on if we want to understand something) domain knowledge (in this case how to build editor, how to do parsing etc.).

However if somebody wants to do custom things then let him/her do it. And yeah first editor, then programming language, then... OS and so on.

I think similar way - CSV should be abandoned and we should use different, better format.

I mean no one is arguing that it is the best or even an OK format for serious data storage. Other solutions exist for a reason. But it’s evergreen and won’t go away precisely because of its simplicity. You will be able to use a CSV more or less identically in 2050, no questions asked.

It’s inherently universal because it’s text-based, and I think you underrate the occasional need to inspect it yourself. And when you do, it’s nicer to view than JSON to throw out an example (arguably JSON and variants need their own bespoke file viewer!)

It’s like saying markdown should be abandoned because text editors do all the formatting for you anyways and different markdown parsers exist and don’t always play nice. Or like saying txt should be abandoned in favor of rtf, even. Okay, sure, not invalid points. But in practice? Simplicity is often a virtue.

Firefox displays .json files decently if you ever need something in a pinch.

Eating 1 GB of RAM and with 3D acceleration. We live in a sick world.

Yeah. If he really wants (but really, not really) to prevent inequality then LLM/AI should be free, open sourced and available to anyone.

Yeah, that's right. But you know they can say whatever they want :D

I think Spec-Driven Development is similar to waterfall but it is not like that. It is more agile oriented and for fast iterations. The idea is simple and can be vibecoded too - quick chatting, planning and so on.

The main things I don't like about LLM is that frontier models are paid (so no unlimited "tokens"), subscribtion based payments, local models are "not ready yet".

Can you understand that there is a lot of DIFFERENT people than YOU who can use AI tools? My team leader (he is also director of department) can use Claude/Codex (or other LLM/AI tools) and create faster, better things than doing it manually. You would not say that he is going down - he is even better with it.

Privacy, no need to pay for tokens, offline usage, endless possibilities - you do not have to pay for tokens (subscribtion payment), so you can do more. The big issue is that local models are not "ready yet" compared to frontier and paid services. It is hard to run decent model without decent hardware. And to be honest even if you can buy hq hardware and spend a lot of money then it is not the same quality.

So start creating quality, convenient and easy to use tools for native development and then create that quality software. You can use free/paid LLMs/agents/AI or don't use any AI. If something feels wrong then change it via next agent iteration on manually by hand. People complain but do opposite things or nothing at all :(