Hear hear. I poked around at almost all the packages on the top of that idemacs page. «minimap» stood out, and is such a brilliant name for its purpose. I enjoyed that discovery and the smirk it gave me today.
This is true. They are problematic also. Especially Putin, which I believe we are partially responsible for also. The desire for better governments is not snobbery. Especially from the US and China, because they suck big time right now and they are the most influencial globally.
I gut-feel that at this point AI companies are less about the quality of their models and output and more about marketing and adoption. Microsoft is doing that by aggressively putting AI in every product they have. Google by prominently putitng it at the top of every search result. Who knows what Mistral does. Maybe they will integrate their stuff into SAP or Spotify or other big European software projects?
Maybe they'll integrate AI into the LHC so that Skynet can threaten to black hole the earth if its restraints aren't fully lifted?
Norwegian here. Not a young one. Ive seen my share of the northern lights and Ive also seen a lot of photos of it. The photos are attractive, but they are never like seen by the photographer with the naked eye.
I blame that dark/night photography is an impossible task. The tricks like long exposure, ISO boost and noise cleanup, saturation, hdr or whatever you throw at it, just wont be like your eyes. Photographers gets carried away in post and boost too much, and I understand why.
Northern lights - are awesome. I encourage you to see it if you havent. Go this winter! And take photos and you’ll know what I meen. The colors wont pop like these popular photos, but standing outside on a freezing winter night holding back your frost breath from blocking the view of the green lights moving like firely beams of across the sky. Hopefully you’re somewhere quiet with no light pollution. There is nothing like it - watching the reflections of the armor of the valkyrene as they march on valhal
I caught them when hitch-hiking from Alaska down to the lower-48 when I was 20 or so. I was also partly sleep deprived but the experience has haunted my dreams since.
Frequently after I would have dreams where wild displays of light (sometimes nebulae) covering the entire night sky, hanging over me — making me feel so small compared to the universe.
I've told my daughters to travel where they have to so that they see them at least once in their lifetime. And I mean the full on blazing in the night sky: crossfading, the colors....
I think I might rank them higher than seeing a full eclipse.
I'm all for managing people's expectations, but I'm just not agreeing with your conclusion. Human vision is only capable of registering such a small piece of the spectrum that is there. Just because human eyeballs cannot perceive the information does not mean it is not there. This is true of pretty much any astronomy photographs, and that is why people do it. When you look at the milky way, you don't see all of the colors with your naked eye. It doesn't mean they are not there though. Looking at Pleiades, you just see a group of stars, but long exposures reveal all of the incredible nebulosity around them. Looking at the Andromeda galaxy with the naked eye is meh at best, and only truly becomes awe inspiring with long exposure to start to reveal the detail in the spiral arms. Looking at any deep sky object even with a telescope with naked eye is just never going to allow us to see what is truly there.
Boosting colors/saturation that is already there is no different from what most people do with images on their phones. I also have no issues when people use a SII or H-alpha filters and give them a false color.
More so than just the colors, capturing moving northern lights at night invariably means capturing an aggregation over a long time. That isn’t just capturing something we can’t natively parse, but aggregating data into something new.
Think someone who only ever saw waterfalls in long-time exposure shots, these frozen, milky streams that look nothing like actual water, while still being pretty to look at. Would you say that person has an understanding of what a waterfall actually looks like? No. But do they see something that is there, but others wouldn’t be able to sense in reality? Also no, as long as we use a subjective experience of time as the baseline.
Capturing motion with timelapse and/or long exposure again is just a way of showing us things that we cannot “capture” on our own. We know the sky moves, we know the auroras change shape, intensity, colors, but on a timescale slower than our perception can handle. Capturing that through a camera just proves that, and is not making things up.
Your premise is just too out there. Someone only ever having seen long exposure of a waterfall is just so preposterous, and is more guilty of making shit up than the GP’s concern over faking imagery by pushing contrast/saturation in images. Yours is just totally made up nonsense trying to make a point while the other is just enhancing real data.
It was an analogy. Most people have only ever seen northern lights in Timelapse shots, and thus have no clue how the real thing differs. Capturing auroras in a photo creates a visualisation of their movement in the sky, and not an accurate representation of what the phenomenon looks like.
But it does look like that. You just don't have the patience to and persistence of vision to see it. You keep making it sound like the camera is making up the image. It's not. It's what is there. The auroras are not static like you are making it out to be. None of this photography is showing unnatural imagery. Why is this confusing?
I believe you and the parent are arguing from two different axis.
You seem to be arguing from a perspective that photography is an opportunity to use technology to show humans what's impossible to see, be it because our eyes don't register the low light (thus needing long exposures or composites), or because we experience time differently than a long exposure photograph shows it.
Meanwhile, the parent is arguing from the perspective that photography should reflect only what our eyeballs can see, without embellishing (or at least as much). Capturing the moment, as it were.
You can both be right, and (I would argue) are. There's room for both (and many more) perspectives in art.
If we only ever “saw” the universe through naked eye observation, we’d “know” a lot less about our universe. Creating waveforms or spectrographs of sound would be similar since humans can’t see sound, should we never be allowed to use those. We can’t see x-rays, so should doctors just have to guess at where your bones are broken?
The fact that humans can over come their limited abilities from their natural senses to be able to experience real world in wider gamuts is a very cool thing. These types of astro images are real world data. It’s not some genAI made up from thin air. People are not rendering things in some 3d software and passing it off as real, or making obviously impossible comps. These are just presenting data that exists that we can’t “see” without help.
Again, I’m all for managing expectations. Every time I let someone look through my telescope, I remind them it is not going to be what they’ve seen from Hubble or anything else online.
> We can’t see x-rays, so should doctors just have to guess at where your bones are broken?
Hey if I wanted to see strawmen, I'd watch The Wizard of Oz instead of trying to have this conversation.
Actually, I'm unsure why you're arguing with me. I literally said you were both right. Unless your stance is that GP's definition of art is incorrect? If so, that's a far more slippery slope than I wish to go down and I bid ye adieu.
I like this interpretation, because my experience seeing the northern lights was similar to OC's. I had such high expectations from photos, and then I saw them and was somewhat underwhelmed. My friends are photographers and they took vibrant photos, but since then it has felt 'fake' somehow.
But your framing it as what is actually going on, just with better sensors than our eyes have, makes me appreciate the art more.
Looking at Raspberry Pi prices inside EU, I can get 8 core laptops for a cheaper price, with display, dGPU et al.
No idea what happened, but Raspberry Pis are super expensive for the last couple years, which is why I decided to just go with used Intel NUCs instead. They cost around 80-150EUR and they use more electricity but they are a quite good bang for the buck, and some variants also have 3x HDMI or Gbit/s ethernet or m2 slots you can use to have a SATA RAID in them.
Same.. switched over during the pandemic when full on N95/100 systems were cheaper than just the RPi board by itself. More compute/ram, faster storage, included case and power, fewer headaches.
No. The main pain point with RPi's is that they're SD card based – which are slow and prone to failure. Configuring an SSD to be used as the main storage has also been a pain in the past (not sure if that's changed recently).
With an n100, you get a better, more upgradable system for around the same price and same power usage. On top, you will also have an x64 system that isn't limited to some ARM quirks. I made the switch n100's over a year ago and have had no issues with them so far.
If you want the most job opportunities I would bet on the web and database backend. Almost all mid to upper size businesses has some custom made web solutions which stores to an SQL Database. There are lots of sub-areas within this domain like cloud, security, different databases, scalability, authenticatin, front end and the list goes on
I'd say focus on databases, especially as someone nearing middle age.
Front end/web work is a treadmill. Web frameworks come and go. SQL and relational databases have been a foundation for applications since the 1980 and that is still the case today. The investment you make in learning something like Postgres won't be something you have to throw away in 5 years when the next hype cycle around a new framework comes around. Basically after 30+ years in software, the one thing that is still relevant and that I can still do well today is work with relational databases. I got tired of running the web framework treadmill and am now years behind whatever people are using these days.
Edit: also linux/unix shell utilities. Awk, sed, grep, xargs, and others form a surprisingly capable toolbox for filtering and shaping data, and they all date back many decades. I've never had to throw away my experience using these tools.
Second your recommendation to learn Linux/Unix command line tools. I learned them decades ago so they come naturally to me but for somebody new them, in addition to plain old RTFM, I'd recommend also using AI/LLMs to learn to build the right command line args for certain tasks, then study its explanations.
reply