I think the keywords are "trusted" and "citizens' rights". US burned a lot of trust in the last few years, and what's happening with ICE doesn't really scream "citizens' rights" either. I can see why many "trusted foreign partners" would now think twice rather than help out the US compared to just 5 years ago.

As a EU person, I'd really like to not have ties with US when possible, and I'd really like to foster the economy of non-US alternatives.

Apparently not so much, the restyle and a consolidated assistance and charging network make people prefer it over the homemade products lately, which keep innovating but you essentially need to re-learn from scratch every time (my original source article is in Italian - https://www.vaielettrico.it/troppa-scelta-poca-fiducia-i-cin... ; this is the closest I could find in English: https://paultan.org/2026/01/21/tesla-model-y-again-best-sell... )

> They’re stagnant.

To some degree, I agree. Esp. on the 800V part which could make charges a bit faster. But I think they're overall stagnant: they pushed a ton of boundaries in the first 10-15y because they had to make something people could buy and use, but now that innovation push has diminished greatly: the cars work for most people, and the range is acceptable and not such a big inconvenience that pushes people to other brands. So no sense in going axial flux motors (that everyone is experimenting with) or bigger battery capacities, or even 48V + Ethernet instead of 12V + CANbus (the CyberTruck has these, but I don't think it has trickled down to the other models).

> Meanwhile, in Europe, the Volkswagen Group is outselling Tesla in EVs. The #1 EV in Europe is a Skoda, and on top of that Tesla doesn’t make a European style city car. Small sedans like the Model 3 are really not popular in Europe, you need a hatch.

Yes, but the Model Y remains the single unique best-selling car. VW needs to put together a few vehicle models to get to the same numbers. And I think the reason is the same as with China: Tesla has always been EV only, and built assistance and infrastructure for it, while VW doesn't have a very clear direction and makes products which are perhaps slightly mechanically superior, but feel inferior on the software and UX side. A car with hardware equivalent to Tesla (which now has 2 trims and 2 battery picks) costs anyway more than the base Tesla (you need to add a few optionals to the VWs). What Tesla is lacking in Europe are smaller models with ~450-500km range.

I've been very happy with www.madetoorderjeans.com (there are others) : get your measurements as they explain, pick the fabric you want (normal, stretch/fit, elastic; light and thin, thick, ...), make your stile, wait 2-3 weeks - for the price a brand jeans (so overall reasonable). Jeans last 3x times longer (wear patterns) and they fit me perfectly, instead of needing to have them adjusted after (expensive) purchase.

Opus can give better results on architectural/concept tasks and I use it sparingly, but it still costs more than Sonnet 5. Opus 5 seems to achieve results very close to Fable 5 while costing less (keeps Opus 4.8 pricing IIUC), but still more than Sonnet 5 then.

Not just the title bar, but often also e.g. half of the settings and the Cancel/Save/Ok buttons.

Basically the screen was 600 pixels in height, but the modal was designed after 2005 so it was 768 pixels tall, and you would get a cropped modal missing maybe 10% top and 10% bottom pixels that you couldn't interact with via mouse, and you couldn't resize nor move (either because it was fixed size, or because all corners that would allow you to do so were off-screen).

Uh? I can literally count the times my Model 3 2019 software broke something on one hand:

- when they redesigned the AC controls to make them more visually appealing but less functional (no button borders and no fill)

- when they decided to put air recirculation under auto-control and ignore the user's settings

- when they optimized the cellular connectivity and it took them a while to get back proper reconnect on loss of signal (garages etc)

- when they tuned sentry's sensitivity and there was some back-and-forth for a couple cycles between "record everything" and "record nothing"

None of this made the car undrivable or totally useless. I did hear of reports of early HW4 cars bricking their FSD computer, and Tesla replaced it.

In my opinion it's still a much better experience than the absolute guesswork of "what will my screen display today when I connect the phone? and where will I find Maps again?", based on software updates on the car AND the phone.

EDIT: also agreeing with the sibling comments: my 7 year old car got a lot of extra features since release, and most of them working very well at the first try.

I had it last year in the Canary Islands (where I assume subscription is way below max capacity - no priority plan) before my current provider placed fiber, and I was getting up to 430Mbps down/100Mbps up with a consistent no-bufferbloat ~70ms ping, which is really not bad.

I guess in populated areas it can only reduce in speed as subscriptions increase, but in sparsely populated ones where fiber is not an option and 5G is crappy and priced similarly to Starlink, it's a valid alternative with no contenders for now. Not to mention that the ISP infrastructure it relies on does not depend on ground infrastructure, so if you're off-grid, you also don't depend on cabling/5G towers/etc for internet access.

This is most visibile for me on Windows/MacOS and complex web apps (e.g. GitHub and GitLab, including consoles of hyperscalers), where 80% of "normal" things work, then you need the last 20% and it's always not working as documented, only half-working, or just outright broken, and you need to find "temporary" workarounds that stay in place for years.

I feel this is being amplified by AI: tests come last (if at all) and are still written by LLMs, nobody really looks at them anymore, green pipeline checkmarks mean less than they did in the past.

It stops making any difference if everyone stops trying. You compare this to subtracting $100 from a billion, but a billion dollars is composed of 10 million $100, and there's only a few rich people in those 10 million.

For me the takeaway is:

- the libraries we use for chip design are small and reproducible, and we can reason about and account for their properties properly, so a designer can minimize design time by reusing small bits that are well understood and ship better architectures faster

- the optimal structures that would allow a design to minimize power use/maximize efficiency and at the same time guarantee highest performance are basically spread-out structures that humans could hardly conceive and reason about, because they look more like roots in a natural ecosystem rather than nice blocks of cut stone that you can stack (not to mention that developing a way to fabricate these chips can be harder)

Template libraries pushed chip design this far, and perhaps there some more to come, but we're hitting a wall where structures are becoming small enough that using libraries is starting to hinder the ability to scale further. These spread-out structures minimize interference and hot-spot creation by separating components that are normally "on" together, allowing better thermal distribution (and thus budget) and reducing interference by physical distance or "off" components, thus unlocking higher clocks and better efficiency.

I once read this article and I find the diagram at the top conveys the point perfectly: https://insideevs.com/news/332584/efficiency-compared-batter...

EDIT: I once replied to the occasional "but that's optimistic for EV and pessimistic for ICE" poster: IIRC, making every EV step 10% worse (e.g. 15% energy losses in transport, storage and distribution) still has EV come out on top by a decent margin.

That chart also partially tells the reason why hydrogen is not going to be a solution for cars: you'd be swapping gasoline for a much greener but only slightly more efficient fuel, while carrying around bulky tanks and having to redo the entire transport network (oil pipes won't do). Maybe for things were work/mass works in favor (naval shipping would likely be one of those) and/or for storage it may make economic sense.

Technology is evolving, so perhaps some day we'll be able to produce and store hydrogen with losses similar to the EV chain, but I doubt it: hydrogen is the simplest gas of all and it's extremely reactive, plus its molecules are tiny (it can leak through metals), so tanks and pipes need to be done with different materials, and I doubt we'll find better containment materials given the physics.

Fairphone 2 came out in December 2015 and saw the last release in March 2023. That's 8 years of software updates when Android devices from that era barely got 3, and you had to pick the flagship.

Kernel updates used to be bound to Android versions because of how the kernel modules development was handled, not really limited by the company or the hardware. I owned a OnePlus 5 and the custom ROM was stuck with older kernels for a while too, until the community stepped up to port it, because there just wasn't an easy way to build kernels with updated custom vendor modules for the hardware. Google addressed at least that part thanks to the Kernel Module Interface (read e.g. https://arstechnica.com/gadgets/2021/09/android-to-take-an-u... ). So newer kernels may still lag a bit based on the manpower available and priorities, but they should be easier to do.

Now that replacement cycles have become longer because hardware is good enough that you don't need to change phone every year, and Android supports KMI and other features that make maintenance easier, more vendors decided to extend support, so at least there's some more choice.

From the environment point of view, I can now finally easily repair my phone. A friend of mine always had bad luck with pixels, so bad that he went through 4 phones in ~5 years (2 bought, 2 replacements). I am at my third phone in 12 years (OP1, OP5, Fairphone).

I get that Google doesn't want to be sued for failing to protect its users and indirect users of the mobile phones sold by other companies, but for advanced users there should be an option to update the signing keys used by the bootloader, so that you can unlock, flash your custom ROM, update keys, and relock bootloader. Such a phone should still be considered "trusted" by Google Integrity APIs. But currently there's no way to do this, so basically you don't really own your hardware.

I gave up on custom ROMs trying to extend my devices' lives and bought a Fairphone instead, so I have the assurance from the vendor that I will have software updates for a very long time.

Emissions scale with performance, and inversely of fuel efficiency. So the environment may not be the most important point, but I'm pretty sure fuel efficiency is high on the list when you're picking a compact or long-range car that is supposed to be fuel efficient.

Also, by advertising as compliant to green specs something that wasn't, means people may have been swayed to purchase irregular cars despite them not being really green, only due to the fact that they may have received rebates and contributions for the purchase, regardless of whether "being greener" ranked high on their decision metrics.

It does one job and does it well, and you have additional features if you want/need to configure them or tweak the results to your liking.

AI is an add-on that you can elect to use or not, and I enjoy just adding a "?" at the end and getting a proper AI response instead of an hallucinated AI summary by default that covers the entire page.

I normally have 5-50 tabs open (so perhaps on the lower end), but I can't recall the last time I crashed a tab in the last 3 years. I also use persistent/pinned tabs and never noticed issues.

My parents put up solar in 2005 when it was very expensive. The original panels, inverters, cables are all still there, nothing had to be replaced in the last 21 years. The only maintenance task is a recommended (not mandatory) yearly cleaning that is probably in the 300€ range and takes maybe 1h, but since 5 years they have 2x EVs and 2x Powerwalls and are self-sufficient for 80% of the year, so that cost is negligible. The area where they live also has enough rain that if they skip cleaning for 1 or 2 years, the average total output is still close to the max they can get.