Posts by Dinanziame 1344 publicly visible posts • joined 2 Sep 2019
a group of us had a trip to Canada for a weekend
A friend had a similar issue. The problem is not who is paying your salary, the problem is that for a while you were working in the US, even working in the US for so long that you had time to take a side trip in the middle to Canada. Even if you are working for and getting paid by a European company, you are not allowed to work long term in the US without a work permit; you are only allowed there on short business trips for a limited time. So they expect you to get in, and get out. The fact that you are doing a weekend trip across the border to Canada gives the border agent the feeling that you are living and working in the US and going to Canada for a vacation. I suppose Americans rarely leave their country, so the concept of having a leisure trip outside of the US in the middle of a business trip to the US is too complex for the system.
Assuming they’ve been prompted correctly the code *should* be no different to any other.
This sounds like "assuming they have received the same instructions, any employee should write code of the same quality". That's probably true for very simple tasks, but if you dig deeper I think it fails pretty quickly.
It is correct that dots in gmail addresses are ignored — yourusername@gmail.com is the same as your.user.name@gmail.com, or even y.o.u.r.u.se.r.na.me@gmail.com.
However, if you receive emails intended for someone else, it's not because their gmail names are the same as yours with different dots. Those addresses with different dots are all your address, and the people sending the emails are simply sending them to you by mistake.
I remember that similar warnings were given when Wheeler became the head of FCC. Yet, as El Reg ultimately noted, he was not a dingo
However, he was then replaced by Ajit Pai, who was at least a dingo, and quite probably much worse than a dingo.
The war in Ukraine has been very informative, because it's the first honest-to-goodness land war that has happened between industrial countries of relatively equal strength in decades. I think that many militaries are completely redrawing from scratch their assumptions about future combat. Progress, yay.
I was happy with the existing solution. You want to sideload apps, you need to go in settings and explicitly allow yourself to do it. Takes five minutes. But on one hand, people have been complaining that this is way too restrictive and makes it too hard for independent developers who do not want to pay the Google tax (was it really?). And on the other hand, I assume, Google is justifying the new rules by claiming it was not restrictive enough and that some people somewhere were still getting scammed.
I'm not sure how many people's life savings are going to be rescued from scammers by this change, and I wonder if it is worth the millions of people who will now have to wait even one day.
Detecting fraudulent behavior is actually pretty hard, because the scammers are often managing quite well to hide what they do by making it seem plausible. This is a form of social engineering. Of course, it does not help that the person reviewing the account is a vendor hired somewhere in a cheap country with orders to do X reviews per hour or else.
The tricky part is that John Deere sells high-power and low-power machines — but they don't actually build low-power machines. They're high-power machines throttled with software. Kind of like HP would sell you expensive high capacity ink cartridges and cheaper low capacity ink cartridges, but the low capacity cartridges are actually high capacity cartridges that stop working before they're empty. Then they sue you if you manage to find a way to use the remaining ink.
I do think Google technically has the higher ground here; as far as I know they completely respect the robots.txt requests of websites not to be crawled. Of course, that means the websites become invisible and may as well not exist — but at least they have that choice, and Google is respecting it. SerpApi on the other hand is deliberately crawling Google even though Google clearly does not want them to.
Now maybe Google does need to be held to a higher standard due to their overwhelming market share, but that's a different argument.
Surprisingly, Yahoo Finance seems to be the best free website for following stocks in real time. It's funny, because I remember an interview of a Yahoo engineer regretting that Yahoo Finance had fallen so much behind Google Finance (must have been at least ten years ago). Since then, Yahoo finance got a lot better, and Google Finance had to be rewritten from scratch and lost a lot of features, because Google killed the GWT project it was based on. There's a new version of Google Finance rolling out these days, and it's nice to get information about a stock, but the prices are not real-time.
According to the "communist" Guardian, Trump has said: "My goal is to not let any windmill be built. They’re losers."
The "woke" NYTimes is reporting that A Trump ‘Blockade’ Is Stalling Hundreds of Wind and Solar Projects Nationwide.
Does Google even make money in UK? I'm pretty sure that all that lucrative ads business is technically operated from Ireland, and the big offices in London are effectively a consulting and support business which barely breaks even. I believe that UK insisted on this being legal, because they hoped UK-based companies would siphon off money from the entire Europe while only making profits in the UK. They thought they had the most corporate-friendly regulations in Europe, but they never saw Ireland coming. Thanks to that, Ireland's GDP per inhabitant is now twice that of the UK.
To be fair, this is the worst way you can try to use the tool — attempt to recreate existing, polished games. And then complain that there are glitches. And that the experience created in minutes by an AI is not as good as a finished product. Duh. The point is to try to imagine a new game that is completely different from what exists, check how it would possibly look, make changes, explore, find new ideas. When you think you have found something nice, implement the game yourself.
There are people who are apparently working to make that happen, only slightly more sophisticated:
I believe the relevant factor here is not that these images were created by Grok, it is that they were published on Twitter. Note that it is the Twitter service that is investigated, and also subject to DSA; not Grok. People have been using Grok to create such images and much worse for a long time — they simply weren't publishing them. In fact, Grok image creation capabilities were heavily moderated from last October, possibly in preparation for making the tool massively available on Twitter. Whatever can be found today on Twitter pales to what could be generated last September on Grok.
“If you are using the same suppliers, such as web hosting companies, as the majority of illegal websites, then you are helping to build the illegal market,” he said. “And, if you are marketing your products through platforms, including social media, that also promote illegal online casinos, then you are helping to build the illegal market.”
So if I use Cloudflare, which is also used by bad guys, then I'm a bad guy too...? And if I have ads on Facebook, and bad guys also have ads on Facebook, then I'm a bad guy too...? Said like this I think it sounds pretty stupid, but I can't find any other meaning to those sentences.
And why would anybody even bother? It's not like there is a shortage of porn on the internet.
You cannot be that naive. First there is never enough porn, second people create sexualized images of women they personally know, whether for their own gratification, or for harassing said women. Check e.g. this year-old story about South Korea for the problems this causes
If you create sexualized images of children with Photoshop, you are guilty yourself, not Adobe. If xAI allows anonymous people to create sexualized images of children with Grok, xAI is guilty. And if Twitter allows people to post these images, Twitter is guilty.
UK politicians particularly have the habit of going "who will think of the children" as an excuse to ban useful technologies like encryption or VPNs, so it's not surprising they are reacting to this. I will admit that I think xAI bears more responsibility here than a random chat app with encryption — There are a lot of image generators, and the vast majority of them have guardrails in place to prevent this very specific issue. Either xAI have not written any guardrails, or they suck at it, or they deliberately set a very low guardrail. Out of the three, I would advise them to plead incompetence. And Twitter definitely should have barriers in place to prevent the posting of such images as well. It is likely that Twitter deliberately omits safety mechanisms for images generated by Grok.
If I drew a picture of a naked celebrity would that be considered illegal as well?
As far as I understand, pictures of naked people generally become illegal if they look realistic. This is true for famous people like for children, which is why so many Japanese comics representing underage sex are legal (UK excepted), just like South Park episodes showing Trump.
I suspect that from their point of view it is a feature they need to maintain, and that very few people use. Why bother? It's even risky for them to work on something like that. Everybody at Google is measured by how much impact they have. The one who has the least impact in a team gets a "Not Enough Impact" rating, loses their bonus, and is on track for getting fired. And if the feature you work on is used by less than millions of users and has nothing to do with AI, it's likely to be you.
That said, the number of people who both know how to set up and use POP3 correctly and are using Gmail is indeed likely to be tiny.
