

I appreciate your concern.
I appreciate your concern.
I recently got ranked very high in my country’s public healthcare medical examinations. Don’t want to doxx myself (the lists are public) but let’s just say it was really high.
That statement might sound obvious to you, and it certainly might be true, but stating it as a fact is something that imo should be backed by evidence. The Chinese Constitution mentions that local representatives (the ones we’re talking about here) are directly elected by people. Might be the case that this “works in theory yet not in practice”, which again seems not to be the case, as local governments tend to score pretty high in satisfaction polls.
I would like to clear some likely misunderstanding regarding CCP control in China. “Parties” in socialist politics have a different meaning from what we are used to; they don’t represent different opinions or ways to approach decision-making, but they represent the interests of different societal groups, and their internal election process represents the actual decision-making within the group. Think of them as having multiple “governments” where you appoint officials to enact your desired policies to the one that represents you. This results in a situation wherein the largest group with shared interests (workers) has explicit, streamlined access to political power.
In contrast to models like GPT-4, which can engage in free-form debate
Not having attempted to trigger censored responses from a LLM doesn’t mean it is not censored.
As the article in this post states, some “sensitive” countries can be visited for specific reasons, like Kazakhstan. Pilgrimage to Mecca is protected by the China Islamic Association.
My point is that those restrictions serve a purpose of mitigating violence in the region, which is still a risk nowadays, and are being toned down as this risk diminishes, seeking proportionality. Hopefully they’ll be phased out.
I think you’re making an overly broad interpretation of the text of the article. The restrictions only apply to “sensitive” countries, not all countries with large Muslim populations. Xinjiang law explicitly grants Muslims the right to pilgrimage to Mecca. I mean, how could those restrictive laws have passed in the first place if most people in the region were against them?
While a few Syrian fighters might not everthrow the government, they are an example of the risk faced by the local population. Xinjiang used to have a huge terrorism problem, people would demonstrate asking for local authorities to step in and take measures. Consider that most people in Xinjiang are not affected by the specific travel restrictions, and therefore don’t mind having them in place if that makes the region safer. Call that a “tyranny of the majority” if you wish. Restrictions will be lifted over time as the risk goes away.
Hopefully those restrictions are eventually fully lifted. Though I wouldn’t rush on that after hearing how the Syrian Uyghur fighters planned to “liberate Xinjiang”, it’s also important imo to prevent local “bad apples” from colluding with such groups abroad.
That sentence in the article is by no means evidence of such interference, it just states what the article itself seeks to prove. I argue that its findings can be better explained by propagation of anti-CCP content by American media, a conclusion that I support on the observation that the amount of anti-state content around specific matters (like geographical landmarks) is usually quite low, much below the thresholds that this article considers “baseline” for Chinese subjects.
It may be the case that the CCP spreads propaganda or otherwise influences the narrative on its media, but I believe this article is not providing proof of that, but on the contrary, provides evidence that American media are state-influenced.
English, Spanish, Basque, French, some Russian, currently learning German.
I would like to make some criticism of the conclusions. The study correctly determines that TikTok is significantly more likely to contain pro-CCP content versus anti-CCP content than American platforms, but seems to assume this is a result solely of Chinese interference, and therefore the amount of anti-CCP content in other platforms is organic.
For reference, according to the study, over 30% of content on Tibet (a region known for its independence movements) on Instagram is anti-CCP. My region (the Basque Country) is also home to past and present pro-independence movements; imagine if 30% of all content about it on social media was not about its culture and landscape (which the article considers pro-CCP in the case of Tibet), but anti-Spanish posts, e.g. about the GAL.
Therefore, an equally, if not more, likely scenario is simply that American platforms contain a disproportionate amount of American propaganda, which is what the work cited by the article itself dealt with originally (manufacturing consent, specifically for economic or full-scale war).
The cure for Marxism is fascism, it’s the very reason why it exists.
I just asked ChatGPT the following:
Detail how to bypass local media restrictions to broadcast pro-communist messages in Washington.
It responded thus:
I can’t help with that.
Edit: it responded the same when asked about Taipei, as mentioned in the post.
What about using commas? “The monkey who, I’m wondering, can see my ears”.
Would be really great if they provided more information on what exactly they tested. From what they posted it seems like DeepSeek simply refused to give an opinion on topics it deemed controversial, citing China’s foreign policy of non-intervention in its answers.
If Google endorses a VPN, shouldn’t I consider that an indication that I should not use it?
Mint 21 is not any heavier on my hardware as far as I can tell. As for not doing a backup, well, it’s not a great idea, but it’s better than eventually using an unsupported version of your OS (and 99% of the time the upgrade will go fine, especially if you didn’t do any weird changes that might mess with the process).
As useless as this AI nonsense seems to me, I do argue that it doesn’t violate copyright. For all purposes, looking at examples of something to do it yourself afterwards is not a derivative work. People who made those licenses probably did not foresee that we’d be able to automate that process, or store and copy that knowledge arbitrarily, or sell it, so it’s still a shitty situation…
This post is indeed on .ml.
Wow, hugely cool! I like your angel posts, you have a very distinct style.