VortexLain

This guy vibecodes his whole stack, from X server, WM, terminal emulator, to shell and text editor. It would've been weird not to have an AI-generated image in an article =)

This person's idea is that if LLMs can create personalized and efficient software on the fly, it's good enough to use such software instead of more bloated, existing software which has to accommodate for usecases outside of his.

Sortof like suckless.org, but vibecoded on the fly.

I've found his Rust android development framework quite interesting, and will probably try using that to vibecode some software I've always wanted to have for my own personal use, but I've never had the time to code properly, especially considering it won't be useful for anyone else. Of course, I'll probably release it under a FOSS license, in case somebody finds it useful anyways.

They can navigate that, if specifically targeted towards that, with proper adversarial reviews (including human reviews).

The question is, if one takes their time and has enough competence to do that.

If a harness wants to upload the whole codebase in one batch, it should explicitly inform the user about it. Instead of exfiltrating the data without consent.

After this, I reconsider whether I want to use grok 4.5 just over an API at all.

Trying to prevent this will cause the same outcome as trying to prevent terrorists from using search engines: it'll only disproportionately harm regular users. Imagine if current AI misuse paranoia would've been applied to search engines, internet libraries (knowledge is useful for planning acts of terrorism), and computing in general.

Probably yes, use an external drive for that sort of thing

In my personal, subjective opinion GLM-5.2 is on par with GPT-5.3

Depends on a lab, but they do have plenty of compute and engineering. So this would only slow down the progress.

What is the average monthly token price for daily reasonix use?

Nope

A lot of lost media used to be considered garbage before it has gotten completely lost. Culture is always worth preserving, at least for historic purposes.

This is a problem of trust towards a software which runs on the user's machine and secretly conducts malware-like stenographic data exfiltration.

Anthropic does their best with banning accounts. As the result, shady API reselling market emerges. OpenAI on the other hand doesn't really discriminate based on a country like that (but a VPN is required nevertheless).

Codex CLI is FOSS, unlike Claude Code, so Codex is less likely to do things like that, and it's one more reason to avoid Claude Code and Claude in general. Hopefully, many eyes will be looking into Codex for malicious things like that.

And which they disable in any country with significant internet censorship, where having a VPN actually matters.

Sometimes it seems like Black Mirror screenwriters work at Meta as a side hustle.

Welcome to the permanent underclass everyone.

It is better to at least do what can be done instead of accepting a defeat without even trying.

Russian Aurora OS was an official Sailfish OS offspring, focused on MDM devices, but Sailfish cut ties with Aurora in 2022, after the Russia-Ukraine war has emerged. It's now developed independently of Sailfish, although they share the same code since the codebase was unified before the split.

Could you please elaborate, which software is usable on mobile Linux except for Firefox? I've seen multiple people using mobile Linux, and they were using Firefox and webapps for everything, no exceptions.

It seems like in 5 years using stolen ID scans as a basic privacy and anonymity tool will be as common as using VPNs today. What a wonderful world.

Allowing contributions only from big tech companies sounds ideologically questionable from free/libre software movement perspective, and it emboldens decisions which go against the user's interests, such as removing manifestv2 in Chromium.

So every adult citizen is forced to open a credit line?

Oh yes, let's induce regulatory capture and ban open-weight models in the name of alignment.

During Iran shutdowns I've been researching what ways Iranians manage to get to the internet by mimicking as whitelisted resources (such as hcapcha). ChatGPT had refused to lookup information written in Farsi since "circumventing state regulation is a crime".

Now this is very bad, as bad as it can get. As soon as all local services will stop working in sanctioned countries, those countries' governments will force all users to either install a root certificate or lose access to all local services and websites. And then it will be possible to use that root certificate for MITM attacks. In the worst case scenario, after the majority of users will install the root certificate, state DPIs will MITM all traffic and will block all un-MITMable traffic.

I really hope this will have proper GNU/Linux support, otherwise it will end up the same way Qualcomm ARM PCs did.

Starting with the days of Siri, i've been evaluating all chatbots of that nature by writing them a meaningless string of text and seeing how they answer. GPT-3 was the first system which instead of refusing to answer or answering meaninglessly has identified that the string of text has no sense.

You could have told the guard you're going to a competitive programming contest.