Hmm, I've a deja-vu... My mind tells me I've read these three comments before on a different topic... If I don't forget, I'll have to check tomorrow on a real PC.
Hmm this doesn't sound like what I experienced a professor doing... But this probably depends on the location and the discipline... Or well at least here in Germany you can more or less pick what you want to do: more being a people / project manager or more own research, or a mix of that... The uncertainty/low chances of getting a tenured position are not different though... And though it might suck, this is something you know, at the latest, after your PhD.
Hmm... You might be able to build one out of a remarkable [1], as that runs linux, has support for external keyboard and might only be lacking in the battery life department, when actively used.
Wouldn’t a simple hack to SSH into something else (maybe over a VPN for consistency) suffice here? There looks to be a keyboard accessory available to get you damn close to what your parent wants.
I wanted to recommend Minisforum, as I have a um560, that is powered through usb-c. But that is not available anymore... So those kind of machines exist...
You could get a frame.work laptop in an external case, slightly larger than hand-sized, though.
In addition to minisforum, there is mele. Mele makes celeron based usb c powered mini pcs. Quite compact. Serves as a good thin client. And lightweight business pc.
Ideas are circulating in my head for a custom solution.
IIRC they have improved the free plan over time, and even mailed users suggesting the relaxed limits might enable moving from paid to free tier [1].
I barely use my tailnet now, might have more of a case for it later, but they are near the top of my "wishing you success but please don't get acquired by a company that will ruin it" list.
But in the end you have to replace at least some of the nitrogen (and other nutrients) you are taking from the field in some way. And without a major shift in consumption patterns (less meat) this will mean fertilizer, as the alternatives usually lead to a much lower yield.
You could do that, the problem is the same as in making nitrogen fertilizer in a chemical plant: energy cost. It just takes so much energy to break the nitrogen tripple bond.
Even if you made a plant that fixes nitrogen extremely efficiently, every joule of sunlight it pumps into the ground is not available as calories you harvest. And fixing nitrogen will take an amount of energy per acre on the order of what you harvested from that acre in a year.
> Well, only being as efficient as existing nitrogen fixing plants (or rather their microbes) would already be quite interesting.
My point is that you can't have corn that is as nitrogen-fixing as a legume and still produce nearly as much corn - the plant (or its microbes) will need the majority of the available photosynthesis products to fix nitrogen. This directly makes the cobs smaller.
> Btw, I don't think plants are close to optimal efficiency in terms of using sunlight. See eg C3 vs C4 plants.
That's true, even photovoltaic panels (which are still far away from their theoretical maximal efficiency) are an order of magnitude more efficient at pulling energy from the sun than plants are. But significantly improving photosynthesis in crop plants is far beyond our current genetic engineering ability.
And I'm not aware of any way to organically fix nitrogen that uses energy outside what is provided by photosynthesis - or gets its energy from digesting dead organic matter, which also doesn't beat the limits of photosynthetic efficiency on a per-acre basis.
> My point is that you can't have corn that is as nitrogen-fixing as a legume and still produce nearly as much corn - the plant (or its microbes) will need the majority of the available photosynthesis products to fix nitrogen. This directly makes the cobs smaller.
I can believe that. However for people who don't want to use nitrogen fertiliser, this might still be useful.
You can see it as an alternative to clover (or manure), that happens to produce eg a bit of grain.
I know. For simplicity, I was talking about the plants in the same generic sense that your gut microbiome is a part of you, and the dead tissues that form your hair and skin are also a part of you.
You still need to hack up the eg cereal plants so they can actually engage in that symbiotic relationship (or perhaps actually directly fix nitrogen all by themselves, without any outside help at all).
in some sense yes, and if you are only interested in good predictions this might work out well. What, maybe due to my limited understanding, is, that this is not theory driven and therefore does not really provide understanding of the underlying process.
> and therefore does not really provide understanding of the underlying process
What is “the underlying process”?
For example, Newton was able to model gravity quite successfully without ever being able to “understand the underlying process”. In fact, physics today still doesn’t have a good grasp on what gravity is. Yet we use the models and equations all the time
In a way, physics is also a collection of black boxes, perhaps just seemingly more elegant boxes
Gravity can be modeled as "things on the earth's surface accelerate downward at about 10 m/s², regardless of their mass". This works very well. Gravity can be modeled as "planets orbit the sun according to Kepler's laws". This also works very well.
Newton realized that these phenomena could be explained as arising from the same underlying process of an inverse square law. This is a much more useful model, and allows predictions that allow us to do things like space flight, even if it is not complete.
The simple ones: advection, latent heat release/absorption from water changing phases, and the Coriolis force. If you need an AI for this, please take a course on differential equations.
The hard ones: droplet/ice crystal formation, cloud feedback on radiative transfer, evaporation at air-sea boundaries. If you can train a model for these processes, please, please tell someone.
"what is the underlying process" is another way of saying, all we have is models. We don't really understand anything. Even gravity. Yet, we can model gravity extremely well for practical purposes.
Exactly. Taking it further, I don’t understand how my hands work. Yet here I am typing away, without even having a good model for it, except just the language I’m using to describe what I’m doing
Physics wants to open the black boxes until it can no longer figure out how to pry the remaining boxes open!
It's not useful to draw a false equivalence between AI-style "the model predicts, that's good enough" and science as a whole which cares very much about the underlying structure.
Why are you assuming that people want to stop understanding the AI models?
If anything, AI researchers are digging deeper into the models too
And people in physics are starting to use AI tools to model physical phenomena
I think that it’s a never ending task to understand all the black boxes. Definitely not possible by a single person. But also at some level you get to circular references. There is no fixed point in the universe, there is no point 0 or origin that we can find. Everything is relative to something else
Wouldn't you be able to somehow couple it with another model that takes the NN data and somehow untangles it's convoluted logic into an isomorphic human readable equation, ie. a model that has one task and that is translating NN logic into human equations.
The training data could be real physics in a simulator held up against evolutionary driven AI logic that competes against it with various goals that are then evaluated and if given a high score then marked as isomorphic and given enough runs you'd get a dataset.
I think at some point it is worth admitting that there are variables you can't account for. Like the precise geography - the models model an area 1 mile square as a single vector, maybe even more coarse. They don't model every tree, rock, and bush. In a neural net you can just have "weight goop" which accounts for the net effect of these unmodeled features, but in a traditional model adding "fudge factors" and extrapolating back from the model to points of interest is tricky.
That's like saying chemistry has nothing to understand because we know Schrodinger's wave equation. Or that we understand biology and psychology just for for the same reason.
Yeah, I think that makes sense. Those systems are similar in that they are made of zillions of tiny parts and there's no way to pull a one equation to rule them all out of it. We could, given infinite compute, but it's just not feasible.
