The idea is that the more pressure you use, the more unwanted materials like fines and oils will get though the paper. So if you press slowly and stop when you hear a hiss, you should have a better brew.

It does make sense, if imagine pressing through in 5 seconds vs 30 seconds, that the paper filtration would work better in the slower press. But I'm not sure if anyone has scientifically measured this.

Actually wait, it's coffee. Someone has definitely scientifically measured it and probably published a two hour YouTube video with their results.

Hey it was my turn to post this quote today!

Do you add whole milk to coffee? The casein and fat should help to reduce acidity and make it easier on the stomach.

If you can't do that, I've heard of people adding a sprinkle of baking soda as a buffer to black coffee. I'm not sure how much you'd need, probably just a tiny amount that you'd barely be able to taste.

Haven’t tried the latter, will give it a shot. Thank you!

Life being non-stop stressful for a majority of the population is not a personal illness, it's a societal illness. Although societal illness can definitely lead back to mental illness which is a very personal affliction.

I do agree that it shouldn't be the job of GPs to prescribe away mental illness though, any more than they should be telling you what eyeglasses to wear. Those jobs should go to psychiatrists and optometrists, respectively. The GP should merely refer you to the specialist.

It does beg the question though, since society is so clearly sick and appears to be getting worse in many countries, whose job is that to fix? The obvious answers are either "politicians" , or "all of us". But politicians seem just as afflicted as the rest, or even to be adding to the sickness in many cases. And saying we all need to come together to fix it might be a truism but is basically useless.

I don't have that experience with Nextcloud Memories.

Everything works well and it's comparably fast with Google Photos for me, and scrolling to specific dates works fine.

How long ago did you try it? I've only been using it for a few months so maybe it's improved over time.

I would disagree there. I've tried lots of photo managers, and for organizing thousands of photos, I think Google Photos has it pretty much nailed. When choosing a photo/video manager, "works pretty much exactly like Google Photos but without all the AI bullshit and privacy issues" is a major selling point for me. Ideally it would even have the same shortcuts so that my muscle memory still works.

It's just a dog whistle. People hate tax, so calling things that are not tax "tax" triggers anger in people without deep critical thinking skills. Or people with skills but not enough energy or time to use them on this particular issue.

We all know what a measuring cup is.

The issue is that flour is quite compressible. So if you've opened a fresh bag and scooped out a cup from the top, you'll get quite a different amount than if you dig in the cup and compress the flour against the side of the bag.

It's not like scooping out rice or sugar, where a cup is a cup, assuming you're careful to use a measuring cup and fill it exactly level.

I'm not sure of the exact error margins (there is research on these things though if you care to look) but at a guess I'd say you should expect 15% error margins on scooping white flour even if you're extremely careful to fill the measuring scoop to the same level each time. Probably a bit less on heavier flours.

Note, I run a bakery, although I'm not a baker myself. But I am the one focused on making sure we have repeatable processes, as much as that's possible for sourdough!

Volume measurements are extremely innacurate and should be avoided for pretty much everything in baking, except maybe water. Not a big deal if you're baking at home - bread with +-15% flour will be totally edible - but if you want repeatable bread quality, use a scales with accuracy to 0.1g. You can buy one for $20 that's good enough for home use.

If you're going to weigh even one thing, weigh the salt. Small variations can have a big effect on yeast activity and final bread flavor. Volume measurements of salt are not at all accurate because crystal sizes vary a lot. People assume (even some bakers that I've met!) that salt is just added to bread for flavor, but it's more like a chemical reaction rate control dial, and you need to be very accurate in how much you turn that dial.

I used Django for a project a couple of years ago. Of course, there's a lot to learn. But generally I found that everything makes sense and I was able to build a moderately complex site with user accounts within a reasonable time. Work took me in different directions, but if a project came up I'd use it again, no problem.

Last year I taught myself Next.js for a project. Everyone would say that's a modern framework, with a modern website. I already know React, I'm quite familiar with prisma.js. Learning and using it was (and is, because now I have to maintain the project) painful, confusing, and full of footguns. I wanted to host on Cloudflare but half the stuff doesn't work so I'm forced into Vercel. Takes ages to understand how images work, how server side works, and so on, and those things are still confusing to me. Constantly I ran into tricky problem about getting data to a client side component, or server side, because some UI library wasn't server side or something like that. Even getting the two fonts I'd chosen into a client side component took me several hours! And I still felt like the solution I came up with was hacky and fighting against the framework.

I regret learning the "modern" framework. I don't regret learning Django. Don't let fancy marketing fool you into using a bad tool, or drive you away from a good one.

Don't judge a web framework by it's website.

> much better than Gemini 3 Pro for writing a lot of code

I know that people here are myopically focussed on code, but that's not what the majority of people use AI for.

If Opus 4.5 is better than Gemini 3 for code, but the same or worse for most other uses (which seems to be the case according to benchmarks), that's great for us but terrible for Anthropic.

Claude still can't even draw basic pictures, for example.