the_mitsuhiko

You can.

Pretty sure the Mercedes S-Class is also not selling a ton of units.

How can legacy auto refresh models every two years and Tesla cannot?

> Is this once again because electric vehicles don’t get pulled into a mechanics shop every 10,000 miles for their oil to be changed and coincidentally inspected at the same time for mechanical defects that could be caught before they get pulled into a TUV annual inspection?

Last time I saw a TÜV report it was that electric cars show up with a) rather little service checks in between and b) they are too heavy for the axels and that causes wear compared to a regular car.

It used to matter to me a few years ago but somehow I just no longer care enough.

I have three, only one is hard wired (the source). That's because I only have gigabit ethernet at the moment and the wireless backhaul is faster than the cable.

> Right, but where do you live?

A 200 sqm apartment in Vienna, mixture of think walls and drywall. The signal goes everywhere with three well placed mesh things.

I think I told this story plenty of times now, but Wifi got so good, that even though we have network cables everywhere, basically nothing is hardwired in our household any more other than the TV and a few sonos speakers that were close enough to the outlets.

We have about ~100 devices connected to our home network according to my router and other than 6 devices, they are all on Wifi. I would never have expected that, but the reality is that it just got so much better over the years that I cannot be bothered with actually wiring things up any more.

That's in part because the outlets are not necessarily aligned well with the devices that would need to be connected, and then all kinds of other shit that is going on with home ethernet.

In 2020 I wrote about my USB-C adapter breaking ethernet [1]. It is still one of my most read blog posts and I get emails from it still, because apparently even in 2025 actually hooking up a USB-C ethernet adapter will cause quite a few switches to fail.

Long winded way of saying: our Ethernet does Gigabit because I never upgraded and has almost no devices left. Our Wifi does >4Gitabit because it was easy to swap and most devices are Wifi anyways.

[1]: https://lucumr.pocoo.org/2020/7/6/usb-c-network-hubs/

I exclusively responded to a comment about capital controls, which are even less likely. I'm not particularly interested in a discussion about what politicans might or might not do.

That Europe is incapable of doing anything bold is a different topic. You don't have to tell me how fundamentally screwed we are because of the consensus issue. But Europe could, without introducing capital controls, implement something. The US did, there is no fundamental reason why Europe could not either.

It's just a question of political will

By taking inspiration from the US. The US has PFIC for instance and many other reporting requirements that make it more attractive to invest in the US than abroad.

> Unless the EU member states actually impose capital controls, investors will continue to send their capital wherever it can earn the highest returns.

You don't need to introduce capital controls to make it unattractive to invest in the US. There are plenty of options that the EU could pull that would make investments abroad very unpopular quickly.

I can also listen to a notary online in Austria. I just absolutely do not want to have the notary involved in the first place.

> I don't find the wording in the RFC to be that ambiguous actually.

You might not find it ambiguous but it is ambiguous and there were attempts to fix it. You can find a warmed up discussion about this topic here: https://mailarchive.ietf.org/arch/msg/dnsop/2USkYvbnSIQ8s2vf...

The value in open source code was never the code. It was the trust that was created around it that it becomes a place for useful innovation, for trust, for vetting, for keeping dependencies low.

I can build my own curl in a week, but the value that curl gives me is that it's a multi decade old library, by a person that has dedicated his live to keeping the project there, keeping a quality bar etc.

The value of curl is not curl, it's the human behind it.

> will GitHub face the same slop-destiny as mainstream social media

At the very least because it's now human + coding agent, separating out the human input from the machine output in pull requests becomes necessary in my book. There are dramatic differences in prompting styles that can have completely different qualities of output and it's much easier to tell it apart from the prompts than from the outputs given that it's basically an amplification problem.

Not really. less bad than the mcps i used.

At this point I'm fully down the path of the agent just maintaining his own tools. I have a browser skill that continues to evolve as I use it. Beats every alternative I have tried so far.

I find it so incredible disappointing that discrimination by citizenship or country of birth is not just alive, but getting worse. I’m afraid if the US is starting with this, it won’t take long for others to catch up.

> Are you saying the agent says "I need to write a tool", it takes a snapshot, and once it's done, it rewinds to the snapshot but this time, it has the tool it desired? That's actually a really cool idea to do autonomously!

I'm basically saying this except I currently don't give the agent a tool yet to do it automatically because it's not really RL'ed to that extend. So I use the branching and compaction functionality of my harness manually when it should do that.

> If you mean manually, that's still interesting, but that kind of feels like the same thing to me.

It's similar, but it retains the context and feels very naturally. There are many ways to skin the cat :)

> You end up wasting tokens on implementation, debugging, execution, and parsing when you could just use the tool (tool description gets used instead).

The token are not wasted, because I rewind to before it started building the tool. That it can build and manipulate its own tools to me is the benefit, not the downside. The internal work to manipulate the tools does not waste any context because it's a side adventure that does not affect my context.

> MCP allows any client (Claude, Cursor, IDEs) to dynamically discover and interact with any resource (Postgres, Slack) without custom glue code.

My agent writes its own glue code so the benefit does not seem to really exist in practice. Definitely not for coding agents and increasingly less for non coding agents too. Give it a file system and bash in a sandbox and you have a capable system. Give it some skills and it will write itself whatever is neeeded to connect to an API.

Every time I think I have a use case for MCP I discover that when I ask the agent to just write its own skill it works better, particularly because the agent can fix it up itself.

I think the path to dependency on closed publishers was opened wide with the introduction of both attestations and trusted publishing. People now have assigned extra qualities to such releases and it pushes the ecosystem towards more dependency on closed CI systems such as github and gitlab.

It was a good intention, but the ramifications of it I don't think are great.

I would probably not build an actual app with HTMX but I found it to be excellent for just making a completely static page feel more dynamic. I'm using it on my two blogs and it makes the whole experience feel much snappier and allows me to carry through an animation from page to page.

The amount of custom stuff I needed to add was minimal (just mostly ensuring that if network is gone, it falls back to native navigation to error out).

Examples: https://lucumr.pocoo.org/ and https://dark.ronacher.eu/

I also found Claude to be excellent at understanding HTMX so that definitely helps.

They tend to change and when I want to do something that the generator does not do, I either need to hack it in (which might break) or i need to fork the generator.

The AI only changes things when I want it to, and to my command. It's very stable.

I moved to a AI maintained custom site generator and it’s ideal for my uses. I have full control over everything and nothing breaks me.

The widespread deployment of NAT and VPNs has counter acted the market forces that were assumed to make IPv6 appealing.

I find it very hard to pick winners and losers in this environment where everything changes so quickly. Right now a lot of people are using bash as a glue environment for agents, even if they are not for developers.

Thanks for adding pi to it though :)