tazjin

> where use of the GPL isn't merely inconvenient, but legally impossible

What sort of ecosystems are these?

> Unfortunately nowadays traveling with DB has become a game of Russian roulette.

Ironically, Russian trains (even over distances of thousands of kilometres) are usually almost perfectly on time.

Germany's DB seems to fill the same niche as other companies there, like Telekom: semi-private companies living off old state-built infrastructure that they're now incapable of (or unwilling to?) maintain.

> Or, we can expect better from software. Maybe someone can fork Firefox and make it run better, hard cap how much a browser window can use.

You can already do this. For example, I use `systemd-run` to run browsers with CPU quotas applied. Firefox gets 400% CPU (i.e. up to 4 cores), and no more.

Example command: systemd-run --user --scope -p CPUQuota=400% firefox

> Is anyone else worried that a lot of people coming from the Rust world contribute to free software and mindlessly slap on it MIT license

Yeah; I actually used to do that to (use the "default license"), but eventually came to the same realisation and have been moving all my projects to full copyleft.

(I know Erlang well, but haven't used Gleam)

The library the parent links to says this:

> Not all Erlang/OTP functionality is included in this library. Some is not possible to represent in a type safe way, so it is not included.

Does this mean in practice that you can use all parts of OTP, but you might lose type checking for the parts the library doesn't cover?

That is literally how a large fraction of airlines work. It's called Amadeus, and it did have a big global outage not too long ago.

One particular cynical reading of that could be "the day I'm held responsible for my code is the day I quit".

In one of the Hitchhiker's Guide to the Galaxy books there was a bit about the company that makes Marvin, the depressed robot, also making sentient elevators. These elevators had issues with getting depression, and the companies using them in their office buildings would hire psychology students to talk to them on the basement levels and convince them to go up again.

Assuming they're being economical (and considering the level of thinking TigerBeetle seems to put into stuff - they probably are) this might be only a few beefy physical servers.

For them it seems safety and QA is a large part of the sales pitch, so that seems worth it.

Your intuition is correct and the sibling comments are wrong. Modern LLM inference servers support hierarchical caches (where data moves to slower storage tiers), often with pluggable backends. A popular open-source backend for the "slow" tier is Mooncake: https://github.com/kvcache-ai/Mooncake

This used to be fairly common, back in the old days. Programs like Pidgin unified many messengers into a single app.

For a while, many messengers actually shared underlying protocols (e.g. Google Talk & Facebook were both using XMPP at some point, and you could even cross-message).

Nowadays this is much harder. There's some exceptions (Telegram) with open client protocols, but I wouldn't wanna try and implement something like Discord, it'll be a never-ending tarpit.

At this early time (this is not long after the invention of the phone itself) - none, really. Stockholm had a much higher number of telephones per home, but not very long after this operators put the cables underground and that was that.

Personally I like Babbel. It looks a bit dated (or did the last time I used it), but its content is really good and it helped me bootstrap 3 out of the 5 languages I speak fluently.

There's no gamification like in Duolingo, you have to bring your own motivation and endure the UI, but it really does get you to the level where you can continue on your own.

(GrapheneOS user, no Google services)

My bank provides the APK of their app directly on their website, and it supports updating itself after that. Actually a surprising amount of apps do this!

Other proprietary stuff I either get from RuStore (Russia-specific), or occasionally from APK mirrors / Aurora. At the moment I have no such apps (they're usually for some specific thing, e.g. an airline app that I need for a day or two).

Serve even statical pages with direct DB access on every hit, using some slow and bloated JS/Python backend, and voila.

Would be cool to get an actual desktop screen with something like your screen tech.

Telegram's clients are open-source, and there's plenty of non-official ones, but for other proprietary messengers you're SOL.

Hard to believe at this point that these messengers used to use open standard protocols, and you could send messages from Google Talk to Facebook once.

Of course they do in Moscow too, the metropolitan/commuter areas exist in most big cities. Seems to be about 8 million less than Moscow for Paris, 6 million less for London, based on what Wikipedia currently says about the respective metropolitan areas.

For what it's worth, I lived in London for some time and its commuter rail system is much less integrated and much more confusing (with different companies, incompatible tickets and so on). In the city it's fine, though the tube isn't very comfortable (except for, ironically, its oldest lines which have luxuries such as ventilation).

Moscow has been integrating many of the commuter rail lines that run to various train stations in the city into essentially the metro network.

This means building new stations, connecting them to existing metro stations where possible, unifying the payment system, upgrading trains and so on. Since 2019 five lines like this that all run overground have been opened: https://ru.m.wikipedia.org/wiki/%D0%9C%D0%BE%D1%81%D0%BA%D0%...

You can see how they connect to the larger metro network on the general metro map: https://www.mosmetro.ru/metro-map

There have been a few similar projects, like reporpusing an unused underground rail cargo tunnel network as another metro line (the turquoise ring around the centre in the map).

In much smaller cities like London or Paris you could probably also find some more lines like this and integrate them, but it needs the political will of course.

There's a few points to make here, actually, based on discussion of the article in the weeks before it hit HN:

Two major and immediate reasons why the pulling is slow are clear and fixable:

1. Guix will try to substitute from unreachable caches for a LONG time before giving up, and Guix infrastructure is less developed than Nix. When I kept experimenting after this post I set up https://mirror.yandex.ru/mirrors/guix/ eventually which fixes the substitute problems for me (Guix mirror on bigcorp infra).

2. Guix by default pulls its channel from GNU Savannah which, for some reason, is incredibly slow in serving git. The project moved over to Codeberg, and setting the URL to there significantly speeds up pulling.

In NixOS, channels (esp. the unstable ones) advance when CI finishes building a certain subset of packages successfully (this is what https://status.nixos.org/ shows). CI populates the binary cache, so this means that a channel HEAD always has a large number of packages available in the cache. Guix has no such guarantees, so updating to any latest commit might even yield one where you end up rebuilding all of its Scheme from scratch or whatever. Picking a commit that's a few days old helps with this.

I eventually switched to a laptop with a "normal" Intel CPU, and pulling did not magically turn fast. In fact, some of the slowest stages (there's one where it says "Computing Guix derivation") don't seem to exercise the CPU much at all. I haven't spent time looking at why (yet).

Guix System does a lot more per-user stuff than NixOS by default, which means that you'll likely end up pulling multiple times.

Pulling remains a much slower operation than Nix evaluation in any case.

There's a few ways to make this all more palatable (e.g. the `guix time-machine -C $channels` setup), but those might be for a follow-up post in the future.

No, not at all. Huawei is targeting markets like China (obviously) and Russia, where the Google App Store is irrelevant.

I've been using phones without Google Play for years.

My point is that YMMV based on where you are.

Do you use the DeepSeek hosted R1, or a custom one?

The published model has a note strongly recommending that you should not use system prompts at all, and that all instructions should be sent as user messages, so I'm just curious about whether you use system prompts and what your experience with them is.

Maybe the hosted service rewrites them into user ones transparently ...

> depending on the probability you assign to a conflict with China in the next five years

And on who you would support in such a conflict! ;)

US-aligned IT specialists are uniquely propagandized (they're one of the main targets of Western propaganda for good reason - they have outsized influence!), so don't expect many reality-compatible takes on this website.

Their resolver resolves sites that should be censored in the EU, so it doesn't look like a government-aligned project.

This is a new public dataset that can be used for training (music) recommender systems, based on anonymized user track choices (for things like "users who like $X also like $Y" and so on).

There's more technical info in the associated paper: https://arxiv.org/abs/2505.22238

Events are timestamped, too, so this could potentially be used to train for things like recommending stuff based on the current mood (I run into this all the time where I might get a recommendation that, in another mood, would hit the spot, but it's not the thing right now).

(Disclaimer: I work at Yandex, but not on anything related to Yandex Music or this particular project. I'm an ex-Spotifier though and music recommendation tech is always interesting!)

They're pretty common in Russia, too. In Muslim countries they're basically everywhere.

When you're used to them it's very annoying to end up in a country that doesn't have them. Especially the US, which in addition has strange toilets that fill up with a lot of water during normal operation and bathroom stall privacy that begins 30cm above the ground.

Checked out when it turned into bad geopolitics fiction.

It's not like anything critical is affected by this. For ordinary people it's things like buying Western games, but those all have workarounds (e.g. through crypto).

It's useful for evading illegal US sanctions, and sometimes also works just as a payment method. I paid a hosting provider with it earlier this year, and that actually worked well.