ako

I just received a consumer association (consumentenbond) test on car reliability and this time the Tesla Y is the most reliable EV in the overview. It scored 9.3 out of 10. The Toyota Aygo is the most reliable ICE with a score of 9.7. Looks like Tesla has massively improved their reliability.

Even though tesla has only 2 models, i would still consider it for a new car, if not for Elon Musk. I have an Y, and it does everything i want it to do. Drives nicely, lots of (cargo) space, no friction charging when driving in Europe. Just plug it in a supercharger and it loads fast. No hassle with subscriptions and cards. Very reliable.

With the 3 and the Y they're already catering for a large part of the market demand, but a smaller model, and a stationwagon might help get it up to 80%+ of all demand.

Palantir's mission is to exactly solve the problem you're describing: break through data siloes to get better information. Core of the platform are data pipelines that can move data from any silo into the palantir data lake, where it can be analysed. Their forward engineering project approach probably enables them to bypass the organisational boundaries between departments. Their top-down selling approach ensures management assists bypassing organisational boundaries.

So bailing out large corporations is a good thing?

Looks like we just have different expectations: i don't want to micromanage my coding agents any more than i micromanage the developers i work with as a product manager. If the output does what it is supposed to do, and the software is maintainable and extendable by following certain best practices, i'm happy. And i expect that goes for most business people.

And in practice i have more control with a coding agent than with developers as i can iterate over ideas quickly: "build this idea", "no change this", "remove this and replace it with this". Within an hour you can quickly iterate an idea into something that works well. With developers this would have taken days if not more. And they would've complained i need to better prepare my requirements.

I think i have enough control, probably more than when working with developers. Here's something i recently had claude code build: https://github.com/ako/backing-tracks

If you check the commit log, you'll see small increments. The architecture document is what i have it generate to validate the created architecture: https://github.com/ako/backing-tracks/blob/main/docs/ARCHITE...

Other than that most changes start with the ai generating a proposal document that i will review and improve, and then have it built. I think this was the starting proposal: https://github.com/ako/backing-tracks/blob/main/docs/DSL_PRO...

This started as a conversation in Claude Desktop, which it then summarized into this proposal. This i copied into claude code, to have it implemented.

You have a lot of control over what the LLM creates. The way you phrase your requirements, give it guidance over architecture, testing, ux, libraries to use. You can build your own set of skills to outline how you want the LLM to automate your software process. There's a lot of craftmanship in making the LLM do exactly what you think it needs to do. You are not a victim at the mercy of your LLM.

You are a lead architecture, a product manager, a lead UXer, a lead architect. You don't have 100% control over what your LLM devs are doing, but more than you think. Just like normal managers don't micromanage every action of their team.

I think we should embrace AI to craft better software. You have a lot of control over the code generated by AI, so all your designs, patterns, best practices can be used in the generated code. This will make us better software craftsmen.

A nice example is guitar building: there's a whole bunch of luthiers that stick to traditional methods to build guitars, or even just limit themselves to japanese woodworking tools.

But that is not the only way to build great guitars. It can be done by excellent luthiers, building high quality quitars with state of the art tools. For example Ulrich Teuffel who uses all sorts of high tech like CAD systems and CDC machines to craft beautiful guitars: https://www.youtube.com/watch?v=GLZOxwmcFVo and https://www.youtube.com/watch?v=GLZOxwmcFVo

Unfortunately, craftsmanship does not come cheap, so most customers will turn to industrially created products. Same for software.

And long before that i built an ms-access UI on top of an Oracle 6 database, so yes, MS-Access was one of the client-server era low-code tools. Like powerbuilder, Oracle Forms, etc. (Hi jouke!)

Depends mostly on efficiency: GraphQL (or Odata as a REST compliant alternative that has more or less the same functionality) provide the client with more controls out of the box to tune the response it needs. It can control the depth of the associated objects it needs, filter what it doesn't need, etc. This can make a lot of difference for the performance of a client. I actually like Odata more than GraphQL for this purpose, as it is REST compliant, and has standardized more of the protocol.

That’s because of the government subsidies.

So you don't think the free market will force manufactures to compete on better batteries? I always thought the benefit of the free market was that it forced companies to compete on product quality... /s

I recently did a day trip of 800km while it was freezing and snowing. Yes the range is impacted, so i never did more than 200km in one go. Then a quick 15 minutes break to recharge and continue. It takes a bit longer, but not bad enough to go back to ICE cars. EV drives so much nicer.

Best thing of Claude Code is that it's cheap to change your mind: you can try some idea, test it, and if you don't like it you simply have it refactor the code. No more big design up-front, "we need all the specs and requirements".

Aha, wasn't aware, good to know, thanks.

I've been using a sqlite alternative to avoid dependencies on a native library. It's go application that uses a native go sqlite reimplementation so i can create platform specific binaries that include all dependencies. Makes installation easier and more reliable.

Here’s something I had Claude code make recently: https://github.com/ako/backing-tracks

I think architecture and UX have more impact on the quality of the software you write for the end user than the details of a specific language. And when you're creating guitar training software, music and guitar playing knowledge has more impact on the quality of the software, than the details of a specific language.

When working with an LLM i care more about prompting it about software architecture, software UX, and the domain we're working on, than the details of the language it uses.

For me, mostly time, time to learn it, time it takes to complete these projects. We have so many other things to do, why bother learning the details of a specific language or tool if AI can do it in minutes. More time to learn about architecture/management/ux/design/guitar/etc.

That's an incorrect assumption. You can get quite far with some skill documents and some examples in combination with tools to compile and run your code. The LLM will train itself on the fly based on the feedback from these tools.

Yes, there are chargers there as well, i guess around 40 in the parking garage.

Company car, i don't see the bill, but i think it's about e0.46 - e0.50/kWh.

I dont have a garage, but there are at least 15+ curb side chargers in 250 meters walking distance of my house. No problem charging my Tesla.

Etron is Audi?

It's time to seriously consider that Trump is Poetin's puppet and he's basically executing on Poetin's wish to destroy the west. Or Trump has just completely lost it.

It’s more like AI provides the development team, and you are the key user and product manager that comes with all the requirements and domain knowledge, the lead architect reviewing the architecture, and the lead UXer reviewing the UX.

Is this related: https://www.paratype.com/fonts/pt/xenia ?

So we’re now in a world where this isn’t impressive anymore? How quickly expectations change. Having started with basic and then 6502 assembly over 40 years ago, this still feels like science fiction to me.

But most enterprise software does not need to be innovative, its needs to be customizable enough that enterprises can differentiate their business. This makes existing software ideas so much more configurable. No more need for software to provide everything and the kitchen sink, but exactly that what you as a customer want.

Like in my example, I don’t know of any software that has exactly this feature set. Do you?

This project is completely built using claude code: https://github.com/ako/backing-tracks Most of the features take less than 30 minutes.

It's basically not very different from working with an average development team as a product owner/manager: you need to feed it specific requirements or it will hallucinate some requirements, bugs are expected, even with unit test and testers on the team. And yes, as a product owner you also make mistakes, never have all the requirements up front, but the nice thing working with a GenAI coder is that you can iterate over these requirement gaps, hallucinated requirements and bugs in minutes, not in days.