What? Elections in Japan?

My take is that best case scenario you’d arrive roughly at the same time you left.

If you have breakfast in London at 8am, then make it to the airport by 8:30, you’re at the gate at 9:30 after one hour of security and controls, and you’ve made it exactly at the time when boarding starts, which usually is 45 minutes before takeoff on most airlines. You take off at 10:15, arrive at 11:45 (which is 6:45 local time), then still have to go through half an hour of border control and getting out of the airport, and then another half an hour to get to the city centre and have a coffee.

You’d still arrive at about 8:30, but I don’t see the whole ordeal taking any less than 5 hours.

I routinely take a 1.5 h flight to visit my family and while I’m a fair bit away from the airport, I don’t think I’ve ever managed to get door-to-door in less than 8 hours. 6 if we are measuring departures lounge to arrivals.

I hope AMD keep pushing to do things well, because right now the value proposition of anything with an Intel processor is more ridiculous than when Apple charges $300 for an extra 8GB of RAM. Their $600 processors currently offer performance on par with the entry-level Apple Silicon M4. Which is great news for Apple, but not for anyone who wants to use Linux or “the other Mainstream OS”.

Something I find incredibly weird about US company culture is how they talk about overtime like it’s a good thing.

“Our employees worked weekends, days and nights to make this happen! We wouldn’t have succeeded without people who are willing to give up their personal lives!”

I hope they not only succeed but get shares. Doing weekends or nights for a company you don’t (partially) own feels like a con.

If they keep duplicating the ask, soon they’ll be asking for a googol from Google. Hehe.

I’m not sure I’m following, it says developers can opt out!

I don’t think anything with the word “intel” can be taken seriously in value comparisons…

When I got my last laptop I ended up with a MBP because there were no high end options for Linux laptops with AMD. Now the options are better, but back then, the only realistic alternative to a MacBook Pro would have had a third of the real-world battery life if not less, even if I decided to spend £3k. That didn’t seem like an acceptable compromise so there were virtually no laptops in existence that could compete with an M2 MBP.

16 GB of RAM are kinda meh, but I can’t think of many $600 devices that can run three 6K monitors simultaneously at 60 Hz, plus then one at a lower res but still 60 Hz.

How do people have the time to organise vigils and get into “coalitions” and politics in the workplace?

Granted I don’t work at Microsoft, but I feel me and everyone around me is overworked enough that when we have the time to stop working… We head home (or close the laptop if WFH) and rest, not engage in additional activities in the workplace.

Yeah, this sounds like a problem for only the 5% of the world who live in a specific country.

It’s not a waste of resources if you learn something. Think of this as research rather than product development. You can try many things (from VR, to miniaturised computers, to cloud gaming, controllers with wonky form factors…) to see what results in a good experience. You don’t need to get anywhere near a full fledged product to understand those things, so the waste of resources isn’t massive anyway.

I’d bet at the moment people decided “this is useful, I even want this for me, so let’s turn it into a product” the steam deck looked more like a screen, a gamepad and a raspberry pi all taped together or jammed into a 3d printed prototype chassis.

If people have spare capacity to work on these projects, the material cost at such a point can be under <5k which is peanuts for a company like Valve.

To be honest I get your point. We use it at work for summaries of 70-page lists of software commits, and with adequate prompting to “understand” what’s what in our codebase it works remarkably well.

Granted it doesn’t work near as well as a person who spends a month working on such a summary, but it does it in seconds. Then a person can work for a day on reviewing this and tidying up rather than wasting time trying to summarise 100k lines of code by hand.

It seems OpenAI should learn to use it correctly first.

What are detached tabs? Sandboxed? Dragged out into their own window? Genuine question

Oh wow that’s terrible. I did think the poem was AI generated. The author (of the blog post) is right, this does an excellent job… at degrading the art.

That’s not efficient enough, why don’t we make them larger and carry over 400 people instead? And we can do special low friction routes where people want to go, so that there’s even better efficiency!

Or, why don’t we accept maybe that there’s the need for different modes of transport and I’m happy commuting to work 8 miles in a bicycle but my 78-year-old mum sometimes physically can’t walk half a mile to a bus stop to take her to the doctor’s and she needs taxis to exist?

Then that might not be the model the previous poster is talking about, because I have to press perplexity really hard to get it to hallucinate. Search-augmented LLMs are pretty neat.

That’s because it doesn’t learn, it’s a snapshot of its training data frozen in time.

I like Perplexity (a lot) because instead of using its data to answer your question, it uses your data to craft web searches, gather content, and summarise it into a response. It’s like a student that uses their knowledge to look for the answer in the books, instead of trying to answer from memory whether they know the answer or not.

It is not perfect, it does hallucinate from time to time, but it’s rare enough that I use it way more than regular web searches at this point. I can throw quite obscure questions at it and it will dig the answer for me.

As someone with ADHD with a somewhat compulsive need to understand random facts (e.g. “I need to know right now how the motor speed in a coffee grinder affects the taste of the coffee”) this is an absolute godsend.

I’m not affiliated or anything, and if anything better comes my way I’ll be happy to ditch it. But for now I really enjoy it.

You can’t measure this, because it has drivers behind the wheel. Even if it did three “pedestrian-killing” mistakes every 10 miles, chances are the driver will catch every mistake per 10000 miles and not let it crash.

But on the other hand, if we were to measure every time the driver takes over the number would be artificially high - because we can’t predict the future and drivers are likely to be overcautious and take over even in circumstances that would have turned out OK.

The only way to do this IMO is by

• measuring every driver intervention
• only letting it be driverless and marketable as self-driving when it achieves a very low number of interventions ( < 1 per 10000 miles?)
• in the meantime, market it as “driver assist” and have the responsibility fall into the driver, and treat it like the “somewhat advanced” cruise control that it is.

I don’t think so, SpaceX claimed (and NASA apparently verified) that the development costs for the Falcon 9 were $300 million. It’s in the Wikipedia article, also here: https://newspaceeconomy.ca/2022/10/23/how-much-would-falcon-9-have-cost-if-it-was-developed-by-nasa/?amp=1

I was under the impression that the Falcon Heavy was a ground-up development. But in any case the Falcon 9 was cheaper, so go figure…