It has to be a “historically disadvantaged group” which the legislation defines to be a variety of groups including women. It’s just more anti-white male racism from South Africa.
(I would like to link to the actual law, but the URL ChatGPT gave me doesn't load and I don't know if that's because ChatGPT is wrong or the South African government's website is having problems).
Until a recent Constitutional Court decision, it was an automatic loss of South African citizenship if you naturalised in another country without specifically requesting retention. After that ruling, the process is no longer automatic, but a number of people chose to renounce it regardless so as to avoid any potential impact of extra-territorial application of, for instance, tax law. [0]
Members of Parliament have asked South Africa's government whether Musk is still a South African citizen, but that information is protected by South Africa's privacy laws. So we won't ever know for certain unless he reveals it. [1]
You've had another reply which describes the process to renounce your British citizenship, I'm wondering if you just weren't aware of that, or is there something else backing up your assertion?
As for why someone might do it, some do it to ensure they're not at any risk of extra-territorial application of South African laws, especially tax laws.
To the sibling comment: It's in case you wish to acquire citizenship in another country that does not recognise dual citizenship.
I.e. that [new] country will not allow you to become a citizen of it unless you renounce your original one.
What that might actually mean practically, or if you can un-renounce it later, I don't know. [Edit]: Turns out, according to that page, yes, you can do that with some kind of bureaucratic shuffle. Daft - but sometimes these things are.
No one is being forced to use or contribute back to the project though. In fact, you can even use GPL licensed software, and so long as everyone with the binaries doesn't ever request the source, you do not have to publish it at all.
Does anyone have any evidence that having the most powerful model being the orchestrator, and other sub agents being progressively less powerful, is optimal? Is this just common sense or it has been measured somehow?
For me it could be that this scheme may make the code itself poorly written, verbose or faulty
I think whether it's bottom-up or top-down, the core principle is still that you use cheap models for grunt work and use frontier models for expertise and direction.
In my setup, the orchestrator also drives a lot of the design and planning, and the less capable models get inputs from the more capable models by responding back up the chain. The more capable models also review everything that is produced by the cheaper models.
I've also had systems that invert this and allow cheaper models to call out to more capable models when they need to, but in my observation I found them quite bad at knowing even when to call out and I find wrapping the cheap models a bit more reliable in that problems tend to get caught before they bubble up to the user.
It sure saves tokens and even the 100$ subscriptions provide massive amounts of dev power. Also fair to note - mixing models is perhaps a good idea anyway as this way the entropy in them mixes and this supposedly improves result quality.
https://www.nature.com/articles/s42256-026-01268-y/ This is probably what you're looking for. TLDR; weaker orchestrator + strong workers scored higher than the reverse. Caveat with that, old paper: "High-capability models: GPT-5, Claude Sonnet 4.5, Gemini-2.5 Pro; low-capability: GPT-5 nano, Claude Sonnet 3.7, Gemini-2.0 Flash."
With that said, there is evidence that we recently crossed over a line for 'best practices' when it comes to systems of multiple agents. Comparing the shapes of 'n agents working collaboratively' vs '1 agent spawning n agents to handle distinct work', Opus 5.5[1] and Opus 5[4] performs better with the former, Sonnet 5.5[2] and Opus 4.8[3] performs better with the ladder.
To the limit that any Winit-based app is. Which is to say it builds and runs, and there is basic touch and text input support. But system integration is pretty limited.
They were right at each time for different reasons
In 2024 the tools were severely lacking. Today they work better (if subpar and verbose), but put an undue burden on reviewers (which are already in short supply)
reply