I would love to see an agreement on the supposed number of (unarmed) civilians killed. Over the course of the past few months, I have heard claims of thousands up to 50k.

You would think the traffic and surveillance cams hacked by the Israelis would’ve shown the extent of this bloodbath.

https://apnews.com/article/iran-war-security-cameras-surveil...

No one’s putting a public traffic cam in the regime’s secret detention sites.

Ah, so now none of the protesters were gunned down in the streets? How convenient.

> As many as 30,000 people could have been killed in the streets of Iran on Jan. 8 and 9 alone, two senior officials of the country’s Ministry of Health told TIME—indicating a dramatic surge in the death toll.

https://time.com/7357635/more-than-30000-killed-in-iran-say-...

Imagine infiltrating the Iranian surveillance camera network and being unable to produce footage of 30k people massacred across two days.

I do not like Iran because of its actions in Syria and Yemen, but even with my bias, I could hear the bullshit Western elitist consent manufacturing engine starting up from miles away.

Yeah, the 30k number is hogwash, but HR NGOs and OSINT volunteers worked up 7k dead in protest over 50 days, including 200 police/military forces, and a maximum of 18k death if you count the fights against separatist/freedom fighter/terrorists (depending on who you are aligned with, choose the description you like more)

Hogwash? More like state-backed propaganda disseminated by so-called objective and professional media organizations in order to justify an offensive war against Iran; a war that has achieved virtually none of its stated aims.

I personally trust OSINT sources more than NGOs these days. I would wager that the security forces numbers are higher. I would also wager that the majority of the deaths were CIA and Mossad backed insurgents operating in the context of a wider, legitimate, civilian-led protest movement.

> I would also wager that the majority of the deaths were CIA and Mossad backed insurgents operating in the context of a wider, legitimate, civilian-led protest movement.

This seems far less likely than the most plausible scenarios, which is that most deaths were the result of IRGC terrorists opening fire into crowds of protesters for the purposes of ensuring they remain in power.

> the 30k number is hogwash

30k is one estimate of actual deaths, it's expected to be higher than any verified number of deaths.

Most estimates fall into the range of 20k to 40k from my understanding so 30k is certainly plausible.

The absolute maximum number of deaths imputable to the IRGC during the winter revolt is 18k. Of that, only 7k have been verified, and of that, only 6k have been from protesters. The reason the 11k have been harder to verify is that most of them were in the fringe of the country, far from hospital, in rural area, and the fighting there was intense. A good part of the unverified 11k might have been civilians caught in the crossfire (an element of propaganda from the IRGC in Baluchistan is that separatists are terrorists targeting civilians, which is 'funny' (very relatively) because it looks a lot like western usual propaganda)

> The absolute maximum number of deaths imputable to the IRGC during the winter revolt is 18k.

Source? I'm curious how you could even verify an absolute maximum given the IRGC/Regime has heavily suppressed information relating to deaths.

HRANA and other watchdogs who actually have people in Iran.

[Edit] https://www.en-hrana.org/the-crimson-winter-a-50-day-record-...

It's even better documentation that i thought, they added the methodology now.

If I remember correctly, the methodology from the NYT and other western MSM was 'the last time, IRGC said 80 death and HRANA said 800, so now since the IRGC recognise 3k death, it should be 30k!' which was then amplified to 40, 50 and even 60k from an Israeli outlet, in 3 days, when the protests and insurgency lasted over 50 days. Honestly I don't trust any numbers if it's published in an American outlet anymore. I now trust 'house of Saoud' more than WaPo or NYT.

Some of the people the theocratic Islamic Regime murdered for wanting a little bit of freedom.

https://www.bbc.com/persian/resources/idt-c005edd8-7204-4c74...

Do you blame them? If a global ring of elite pedophiles turned out to be true in spite of all the gaslighting and denial, then why couldn’t the moon landing also be a conspiracy?

Keep in mind that the elite class couldn’t give two shits what the peasant class thinks. In fact, having us believe in false conspiracies helps distract the masses from the true conspiracies :)

Astounding to watch the mental gymnastics at play.

It’s like how Israeli lobbying orgs state that “claiming Zionist orgs control the media is antisemitism”, and then the solution is literally “we should use our contacts & supporters in the media to stop this kind of rhetoric”. Beautiful.

So BYD for cars and Samsung for phones and consumer electronics more generally (from fab upwards).

In fact, I believe Samsung is the only company on the planet that can design & build a state of the art smartphone from scratch - silicon/fabrication, SoC, battery, baseband, camera sensor, memory, and display.

What other high tech vertically integrated producers exist in this group?

While I could have sworn RIM put out their own modems (which Qualcomm used to make life difficult for them, especially as the world transitioned from 3G to 4G), and did their own hardware and software, I can't currently find a source

Qualcomm makes life difficult for anyone who has to deal with them.

And now it appears Microsoft and Nvidia are gonna make life difficult for them in the computer market.

They are "more" vertical, but they too have vital suppliers that they could not do without. The semiconductor supply chain is deep. Everybody knows ASML, but there are countless others that produce raw wafers, etching machines, special chemicals and so forth.

IBM

IBM no longer has fabs (spun off as Globalfoundries and later sold), and no longer manufactures PCs (sold to Lenovo?), but it does make mainframes I guess?

I am still amazed that IBM is pushing their POWER processors forwards for things like their System Z Mainframes. From what I have heard they are still really fast with I/O and general shifting of data but I'm not sure how much better than is than the alternatives.

Wouldn't be surprised if that finally gets sunset in the next few years.

POWER still exists? That's kind of neat. I had a POWER 1 rs6k way back. Almost wish I'd had room to keep it just as a sort of museum piece. The processor was several chips on one or two large PCBs IIRC.

https://en.wikipedia.org/wiki/POWER1 IRC.

I know, they are up to the POWER 11 spec now. It is the definition of a brilliant architecture that was simply out run by others (x86) that could push their mediocre setups far better than they could.

I thought GlobalFoundries was AMD's fab division spun off.

Precisely. IBM paid them to take their fabs away (yes, they paid them, not the opposite, they were so obsolete that it was basically a waste disposal operation)

was there some regulation preventing them from just shutting the plants down and selling off the real estate and equipment? severance payments too high?

IIRC they had an agreement that Globalfoundries were to take the fabs and further develop them to shrink to better nodes. I think it was less than a "take this crap off of us" and more of a "we still need fabs, so here's some money and please continue running them for a while". GloFo then utterly failed at competing with TSMC, Samsung and Intel which was quite a problem for IBM. In general if you have a certain design you can't immediately port it to a different fab, you need to engineer it around the constraints of the fab node and process

Yes, more specifically IBM -> AMD -> spin off.

They are mainly in services now, no?

In 2021, the majority of IBM's "Global Technology Services" was spun-off as an independent company named Kyndryl.

I define all units in Nix because:

a) It is way nicer and you get decent validation at build time

b) A LLM can port units over if the need arises; it’s a very light abstraction around systemd syntax

c) I personally don’t see how I would ever move to another distro :)

b) Or you could take the compiled units from /etc/systemd and copy them wherever

I believe that probably won’t work unfortunately; the generated unit files have a bunch of hardcoded Nix store paths.

Yes, but I believe there is a symlink from /etc to the Nix store, so systemd can find them.

+1, NixOS makes working with systemd a breeze. Defining units in Nix beats wrangling INI files.

  systemd.services.sync-recyclarr = {
    serviceConfig.Type = "oneshot";
    path = [ pkgs.podman ];
    script = ''
      podman exec -it recyclarr recyclarr sync radarr
      podman exec -it recyclarr recyclarr sync sonarr
    '';
  };
  systemd.timers.sync-recyclarr = {
    timerConfig = {
      OnCalendar = "daily";
      Persistent = true;
      Unit = "sync-recyclarr.service";
    };
    partOf = [ "sync-recyclarr.service" ];
    requires = [ "podman-recyclarr.service" ];
    wantedBy = [ "timers.target" ];
  };

I don't currently have a personal use-case for container services, but Quadlets are another example of systemd (and podman) beauty. It looks like someone has gone through the trouble of making the OS+home-manager modules: https://github.com/SEIAROTg/quadlet-nix

I never had the opportunity to try out quadlets, but they seem powerful.

I maintain one of the competitors listed in that README (compose2nix), so I am a bit biased haha.

For now, I prefer the ability to interop with Compose.

is this irony?

No. Is that not readable to you lol? I think anyone with even a passing familiarity with systemd would understand what that chunk of Nix is doing.

Compare it to the alternative of using plain systemd (including command(s) required to enable units).

Also, consider what build-time validation you get prior to starting the unit/timer. Hint: zero.

It's significantly uglier and it also skips the helpful headers / sections in the systemd INI files. `[Unit]` and `[Service]` and `[Timer]` represent different layers of execution. Many Nix people got used to the horrible syntax of Nix, I guess? I still find Bitbake significantly more palatable than Nix.

I do appreciate build-time checking but I think this can be solved at systemd side as a separate tool just as effectively.

Ugliness is subjective :)

But my question was around readability: were you able to understand what the snippet I shared is doing?

Re: build-time checks - but systemd hasn’t done it, and I am also unsure where exactly this verification would even take place given systemd’s configuration model. Unless you’re talking about some kind of language server or IDE integration.

Uh, downvoting is definitely a useful tool to signal that you don’t agree with an opinion or subjective statement.

But yes, downvoting a factual statement makes no sense.

That's just silly. The fact that you disagree with an opinion does not mean that other people should not get the chance to be exposed to it. That's how echo-chambers form.

At the same time, the amount of disagreement an opinion gathers is an extremely important channel of information for determining whether you agree with someone's position. Silencing the disagreement with it gives an outsized benefit to harmful and malicious statements.

That's fine when a disagreement (or downvote) is just a signal on the post, but when it's used as a way to silence an opinion (e.g downvotes will hide it) that's even more harmful and malicious. Especially when the guideline is to downvote posts that are low-quality or don't conform to the rules, not posts you just merely disagree with or are against your belief system. Popularity should not be confused with truth.

No, that is not a “complex” position at all. On the contrary, it’s a fairly simple position where you take no strong stance but still want to claim the positive aspects from each side.

Are unions universally good? No, because humans are in the loop, and humans can do bad things.

Does that change the fact that the concept of a union is one of the greatest innovations in all of human history? No.

Can unions today help disparate human workers collectively improve their working conditions? Yes, because this is what unions were designed for, and I think is the key outcome the Rockstar folks are betting on.

For a recent example, read up on the Samsung union bargaining for company wide bonuses in the wake of the huge profits made off the surge in demand for memory.

The Samsung deal is exactly what I'm talking about. It is not all entirely good. Everyone at Samsung not working in chips got screwed. And those lucky few union members used their power to extract an unfair amount of money from the company. This will cause the company to lean towards other avenues in the future, potentially harming everyone else.

Why are those people getting a huge check? Not because they worked harder. Because AI came along and made their product more valuable.

https://www.nytimes.com/2026/05/27/world/asia/samsung-ai-pro...

> But a smaller union associated with workers in the consumer electronics division — which boycotted the negotiations and whose 15,000 members were excluded from the vote — accused the lead union of neglecting their interests, and decried the deal as “discriminatory.” Under the agreement, workers in the consumer electronics division are expected to get payouts that are a fraction of those of their semiconductor division peers.

And how exactly is this situation worse than the unfair allocation of salaries and bonuses in companies today? Even within the same company, people can get paid more based on the org/division they work in (e.g. core AI teams), or even based on their (team or individual) perceived value to higher ups.

At least with the Samsung union, the decision is being made bottom up vs. top down.

> And those lucky few union members used their power to extract an unfair amount of money from the company.

Why is it an unfair amount, though? Who gets to decide what compensation is fair and what isn't?

Interesting. So that explains where the inspiration for the cult-like elements in Severance came from. The parallels between Lumon and IBM were fairly clear to me, save for this part.

> without needing to pay a rent or ask for permission

Firstly, the ability to “build” the best and most capable software is still locked behind frontier models, so rent is still and will always be due.

Secondly, OSS is about giving users the option to be in control of and have visibility over the software they run on their machines.

But that doesn’t mean that humans do not want or deserve recognition for the work they do to provide these libraries and tools for free, which is IMO partially why copyright and attribution are critical to OSS as a movement.