initplus

I assumed outdoor time helps because your eyes spend time focusing at a greater distance. Going outside, but only to use a phone/ereader/book is likely not effective.

It's not about subsidizing the repair. Samsung placed a limit of 7 official repair parts every 3 months per repair shop being supplied by ifixit. How is a phone repair shop meant to operate when they can only get 2.3 parts per month?

'Dr Wright presents himself as an extremely clever person,' the judge stated. 'However, in my judgment, he is not nearly as clever as he thinks he is. In both his written evidence and in days of oral evidence under cross-examination, I am entirely satisfied that Dr Wright lied to the court extensively and repeatedly.'

Pretty damning statement.

Follow up, don't ask "Can I ask you a question?". Just ask the question in the first message.

I'm not the one making assumptions, it's thousands of independent hosts, and all big tech orgs (including specifically Apple in this case) who are making that assumption. I didn't say the assumption was right, just that it's trivial to avoid falling afoul of it by choosing to use a different TLD.

You end up fighting an uphill battle against every third party that blacklists .xyz, It’s not worth the fight just to use a cute tld and save a few dollars on registration cost.

Issue isn’t self hosting email, it’s self hosting it at .xyz.

They had one of the cheapest registration costs. And so ended up with a high concentration of spammers compared to older established tld’s like dot com. Using the tld for legitimate purposes is really challenging due to the high number of systems that flat out blacklist it.

Don’t want to sound like I’m victim blaming the author. But I can tell you exactly the issue with their account: registering with an email on a self hosted .xyz domain. Using sketchy tld’s is just asking for this kind of trouble.

https://news.ycombinator.com/item?id=28554400

In the future? This is almost certainly already the case.

People are searching for comparisons, but these pages have been designed to seem highly relevant to the search engine. Not to be highly useful for readers. That page from Zapier is another example of the same issue. When Google rolls out their new ranking algorithm to zapier.com, I'd expect it will also be de-ranked.

The words "Power Automate" only appear 3x in the page. The title, subtitle, and a link to the same page. Nothing else on the page is relevant to "Power Automate", it's just the same marketing copy from the rest of their website. As a user looking for actual details on how the products capabilities compare to Power Automate this page is useless to me.

Google have been completely transparent about this. They have explained exactly what kinds of pages would be de-ranked under the latest changes.

Yeah I felt kinda hoodwinked. I read their blog post and felt quite outraged. Google stepping on the little guy, “they provided no explanation” etc.

Then I scroll to the bottom of the thread on hacker news, saw this comment, and “oh, that’s why”. They didn’t even take down the SEO spam before complaining about being caught.

I really don’t think I could trust any “privacy focused” email provider that relies on such deceptive tactics to promote their business.

You can downrank your own pages and website by hosting low quality SEO focused content. Spammy content written by others, on their own domain, doesn’t affect your ranking.

Explicitly mentioned on that blogpost:

“…or feel like they were created for search engines instead of people. This could include sites created primarily to match very specific search queries.”

The numerous “Tuta vs …” that all reword the same information clearly meet the “designed for search engines instead of people” standard imo.

These pages, and more on the Tuta website, are a blatant example of SEO spam. Thousands of words long, all rephrasing the same talking points and SEO keywords. No real customer is going to read thousands of words of rephrased crap.

They even admit as much in the blogpost: “We have no idea why Google is no longer showing our website for thousands of keywords that we used to rank for in the past.” Juicing your ranking with SEO spam littered with “keywords” is now penalized in Google ranking. They would rank better by removing 90% of “content” on their website.

They’ve been de-ranked because their website hits all the “SEO optimized” red flags that Google are now downranking for.

Pages like “Outlook vs Tuta Mail” “Gmail vs Tuta mail”, “Yahoo vs Tuta Mail”. All with the same rephrased taking points about their product. Then each of those talking points has its own dedicated page, just saying the same thing over and over again.

Want a better rank? Remove all this SEO crap, leave up the parts customers actually want.

If I get an email from @tuta.com, how do I know if it’s an official staff account, or a member of the public?

He was held in solitary confinement for 108 days for a white collar non-violent first offence. Nobody should be subject to such conditions, for any offence. It is state sponsored torture. The Japanese justice system is dysfunctional, and it has prevented the case from being investigated fairly and appropriately.

No justice system that's actually fair can have a 99.8% conviction rate at trial. It's a process designed to achieve convictions and punishment, not actually discover the truth.

It's not disruption to make me wait 5 weeks to read a digital file through my local library's Libby service. It's an outdated business model using artificial scarcity that isn't effective at getting books onto readers devices. Spotify doesn't make me take out a hold on the artist I want to listen to. Netflix doesn't make me queue to watch the latest release.

This isn’t a China specific thing. The story of Carlos Ghosn is a particularly worrying example out of Japan.

“They didn’t fill out some trivial form” is never the answer. It’s a show of power by someone, either Chinese govt or Meta.

Anecdata yada yada, but both colleagues of mine with a Dell XPS had trouble with them. This is Dell's top of the line productivity laptop segment. Yeah we got top of the line service. But the damn things still broke down in the first place. Critically, I recall one of them blue screened in the middle of some important infra changes and triggered an outage...

The problem is that the existing language concepts that implement these ideas are ambiguous in many languages. Does int *x imply that x is an optional value, or that it’s intended to be shared with/from others, or both of the above?

Speaking of which, why are we stuck with this terrible "trailing space selected" behaviour? It's not the case on all platforms, macos/ios perform fine and only select the actual word but Windows still includes the trailing space. There are posts online complaining about this going back 15 years at this point, it's super low hanging UX fruit.

I think you are coming at this from a particular Python mindset, driven by the limitations imposed on Python threading by the GIL. This is a peculiarity specific to Python rather than a general purpose concept about threads vs processes.

I think you have missed a key detail in the story. None of the video content is in any way criminal.

It’s an attempt to unmask someone the police are investigating for completely unrelated criminal activity (money laundering with cryptocurrencies). The video was apparently something completely innocuous about drone surveying.

Entrapment isn’t just “any time police trick you during an investigation”. It is only entrapment if the police actually induce you to commit the crime you are charged with. Viewing these videos is not a crime.

The first is a somewhat clever attempt to unmask someone ann undercover investigator was already talking to. Police should have narrowed scope of the warrant by only asking for data on viewers within a narrow window after they sent the link.

Even better might have been to directly link to some service that they already control on a honeypot URL, and then gone after the ISP for customer details.

The reason it took so longe is because carriers are still involved with RCS, and carriers suck.

Every successful messaging protocol aside from SMS does so by routing around the carrier and sending messages over data. iMessage, WhatsApp, telegram, signal etc.

It’s not a problem with individual offshore engineers. The problem is the way the business they are employed by operate.

Inference compute costs and training compute costs aren’t the same. Training costs are an order of magnitude higher.