pmontra

I never sell anything, everything finds a new life.

I got a Sony Xperia Compact from maybe 2016. I use it as a bicycle navigator / recorder. Recently it stopped reading SIMs, probably rusty contacts, but all I need is wifi to update OSMAnd and send me the tracks I recorded.

I also have a very old tablet, maybe 2014. My SO uses it to view movies with VLC and TV via TVHeadend.

Interesting but very difficult to perform in the right way. Example: Italy has maybe 45k buildings that could show on that map https://it.wikipedia.org/wiki/Categoria:Castelli_d%27Italia but they can't be collected by looking at the English Wikipedia so it's almost empty compared to what could be. Other countries, same problem.

Somebody suggested in other comments that the data could be from OpenStreetMap. Or let people add their own castles, but then you have all the sort of problems of content moderation.

I've been doing more or less this (from the post) for months:

plan deeply, then capture the plan as a todo list, then start.

I do the plan deeply phase by writing a 50 line specification myself and then asking Claude to evaluate it, find what I missed, ask questions. I answer the question and we iterate until I know that we have what I had in mind to do.

Then I ask it to write the implementation plan, that I might end up reviewing because it always asks some more question. Finally it implements the steps.

If the planning was good the implementation is fast.

Using the cheaper model is also a business decision. We don't know why they are using the Chinese models: openness or money or both or a third one?

Frankly I don't see much difference between Claude and DeepSeek except in the price. I use the Pro plan to work for a customer of mine and I topped up $2 on DeepSeek in late May for personal use. I worked on 3 projects and I still have $0.72 left. The Chinese companies will win on price, not openness.

Possibly, but they already had bullet trains in 2015. I've been on a few of them.

Paris has a metro station everywhere at least in what a tourist can assume to be an enlarged city center. Tokyo is another city with a lot of metro stations. Manhattan too, at least up to Central Park (but 20+ since my last visit.) I don't remember Shanghai to stand out positively or negatively, but 11 years can be a long time.

I've been in China in 2015 and like anywhere else in the world it was very mixed: some urban areas like central NY or central Madrid and Milan (or much shinier) and some rural areas like 200 year ago, but inevitably with electronics.

Basically in every country of the world you can travel one hour from big cities and get in a place deep in the fields or the woods with very different needs and dynamics from the city. They could be different countries and maybe both cities and countryside will be better off if we could have fractally composed states with different laws and regulations.

I break it down in reverse order to let you understand how I get a gist of that definition

wavefunction, for example light or any particle, this can't be argued.

the complex plane, because? But a complex number is only a pair of numbers and I know that in polar notation it's handy for:

rotation! And I remember that the electric and magnetic components are on orthogonal planes so we are back to complex numbers and maybe the components rotate, but how about neutral particles? But they are not neutral inside. I think that I'm already off track, very lost.

Energy = frequency, because I know that higher frequency photons are more energetic.

Finally, usually we learn about kinetic energy, then E = mc^2 and maybe about the relativistic effects, without the formula. Those things can be understood easily and are taught at school. The jump to angular rotation and the complex plane is something that few people are exposed to and, as I hope to have demonstrated, it's not easy to map to common experiences, which could be inevitable.

Northern countries have a lot of light in summer and a lot of darkness in winter anyway. I saw the Sun setting at 23:00 in Iceland at the end of June and rising at 02:00 and pitch black in Denmark at 15:30 in December. What can DST change there?

On the other side, equatorial countries have about the same sunset and dawn all year long. Again, DST is useless.

Where DST is useful is in a place like the one I'm living at, at 45 degrees North. I'd go with permanent DST plus an extra DST in summer. Basically it's what Spain have been doing for almost a century.

It's cloudy and still reasonably bright outside here. It will be dark past 22:00. I'd take an extra hour of light in the evening and darkness until 6:00 in the morning. It matches the time people goes to sleep and wake up here.

Do you mean a real Land Cruiser, like the ones that can cross deserts? I'm with you.

Every car can brake, turn and accelerate much harder than any of us will even think to do especially turning and braking. But you pinpointed why we must not attempt to reach for the limit, not for us but for the others. And anyway a normal car won't lay long if driven like a racing car. Every single component is not designed for that.

The usual computer analogy /s You don't run a LLM on a Core Duo 2, but of course https://yeokhengmeng.com/2025/04/llama2-llm-on-dos/

All of those companies have very compelling reasons to provide those tools for free, as in (most importantly dor adoption) no money asked and (less importantly) open source. They make money because of the number of people using those tools. They are a bridge head into controlling their own share of the market.

So: investment, not good will.

I'm not sure what level of detail those prompts you write about have so I might misunderstand you.

Before AI we had prompts called "requirements" and of course they were the starting point for reviewing code: the customer needs X, the code doing that is Y.

With agents that prompt became much longer, with the original requirement a short note at the beginning of the conversation with the AI, but the resulting code is about the same size. Maybe in YOLO development the original requirement is all the AI needs but I'll be surprised because requirements from customers are too fuzzy and can be turned into very different features, often too different from what the customers had in mind.

Actually I never looked at the detailed prompts of coworkers, only at the original requirements from customers. How they instructed their agents is not particularly interesting except maybe to get a gist of the different approaches.

Why when I read an how to stop Claude from saying X, I grep my saved conversations and I find no occurrences of X? I wonder if I'm using it differently from anybody else. It happens with coworkers too.

RYO is the best way to learn, if you have the time to learn. After building one own solution one is able to assess the strengths and weaknesses of what other more experienced people built. Without the learning phase those other solutions are incomprehensible black boxes. The trick is not to skip the phase in which one researches available solutions.

Which UX problems? When I read

Thunderbird’s robust functionality is its superpower, but a dated interface shouldn’t be a barrier to entry for newer users.

I started preparing for the worst.

Yep, I broke locate when I made my home 700. Its user could not traverse my files anymore. I had to make it run as root. A better design would be to traverse each user home with that user id but apparently it assumes that the home dirs are 755.

This is new to me (thanks, I'll investigate) and it's the kind of things the books of Physics at University (Computer Science) was full of, completely detached from real world experience. Then there was the book of exercises to prepare for the exams. Sometimes finding the right way to go from theory to calculations was a puzzle in its own. I guess that they were almost the very same books they used at the Physics department and teachers are really interested only in students that can master the subject by themselves with no help. I had one of those students at high school. He was reading university books by 16 and he is a professor now.

Luckily the books of Physics at high school started from practical experience and then showed the equations. So, my advice to somebody willing to learn physics, with plenty of time, no ambition to become a researcher is: go through high school books to get the gist of the subject, then go deeper with one of those university books.

Eventually, I left almost all my groups behind, keeping only the ones tied to genuine relationships.

That's why my only social media is WhatsApp, if that's a social media.

This PRIDE? https://cio-wiki.org/wiki/PRIDE_Methodology

Maybe you must harvest data from all the national Wikipedias, easier said than done.

Example the page for https://it.wikipedia.org/wiki/Castello_di_Chignolo_Po exists only in Italian and in the local dialect. I'm sure that there are plenty of castles like that around the world without an English page.

I know how it worked. Buffering (probably complicated on a pure mechanical device) or locking the keys until the dial came back to rest would still be better than rotating and misrotating the dial.

I agree on your general idea but rotary phones were a real PITA.

They were slow to operate. We jumped in a very short time to numberpad phones, which initially only replayed the analog signal of rotary phones. I'm surprised that nobody built an analog phone with a numberpad that operated a hidden rotary dial before electronics got everywhere. It would probably need to be plugged into main but those phones were not mobile so a socket was always available.

On the other side, a 33 or 45 rpm record is only inconvenient to store and transport but it plays back in the same time of any modern music format and seeking is not so bad: you can see where you move the head and with a little of practice one gets quite good at it.

Cassettes were another PITA.

My experience with Claude and Ruby, Python, Javascript is similar: it's pretty good at finding the array of strings that was passed to a method instead of an array of integers. Think about record ids coming from a JSON API call. Or the single value instead of an array. I don't remember which Python XML parser is fond of returning one or the other according to the cardinality of the sub elements. Anyway, not only it writes the code to handle those cases but it traces the code and it finds the bugs. So type checking at coding time and who cares about writing the type annotations. They would be probably good to speed up the code at runtime but none of my customers use them and none of them is concerned about the current response time of their systems.

The issue with formal proofs is that we might end up with correct code that does the wrong thing. I mean, something that the market doesn't care about. At the same time a buggy set of PHP scripts does what people care about and captures the market. Think about 20+ years ago and you find a lot of examples.

It's not clear how the map is built and if it is possible to contribute content. Only in my area there are dozens, possibly hundreds of castles that do not show in the map. Every city and many towns have one. Several hilltops have a castle or fortress too.

If they grow desperate I have GBs of DDR2 AND DDR3 in a drawer.

apps that are safe by design for children

How do we design such apps? Let's rule out age attestation (to allow only some age ranges) or scan of content because they are orthogonal to apps. What are the design patterns that prevent adults to meet kids? No messaging?

It makes sense to use the left hand as phone holder to be free to use the right hand to do some other activity, example: I will have my breakfast while reading HN.

However I usually keep my phone with my right hand and scroll with my right thumb. Furthermore I am writing this comment by swiping with my right thumb. I can swipe with my left thumb but I'm not as accurate.