I have a theory that this focus on ideas vs solutions also divides individual researchers, in what drives them. Agreed that academia celebrates and rewards ideas, not solutions. And maybe that’s ok and how it should be, solutions can be done in industry? But the SNR of ideas feels too high at this point.
> Maintaining context of the overall project and goals while working in the weeds on a subtask of a task on an epic (so to speak) both in terms of what has been accomplished already and what still needs to be accomplished
This is a struggle for every human I’ve ever worked with
This is probably the biggest difference between people who wrote code and people that should never write code. Some people just can't write several connected progtam file without logical conflict. It's almost like their brain context is only capable for hold one file.
Yes. I wonder if the path forward will be to create systems of agents that work as a team, with an "architect" or "technical lead" AI directing the work of more specialized execution AIs. This could alleviate the issue of context pollution as the technical lead doesn't have to hold all of the context when working on a small problem, and vice versa.
This is kind of what the modes in roo code do now. I'm having great success with them and having them as a default just rolled out a couple days ago.
There are a default set of modes (orchestrator, code, architect, debug, and ask) and you can create your own custom ones (or have roo do it for you, which is kind of a fun meta play).
Orchestrator basically consults the others and uses them when appropriate, feeding in a sensible amount of task definition and context into the sub task. You can use different LLMs for different modes as well (I like Gemini 2.5 Pro for most of the thinking style ones and gpt o4-mini for the coding).
I've done some reasonably complicated things and haven't really had an orchestrator task creep past ~400k tokens before I was finished and able to start a new task.
There are some people out there who do really cool stuff with memory banks (basically logging and progress tracking), but I haven't played a ton with that yet.
Do we understand the early dynamics of cancer? Do the hallmarks need to appear more or less at the same time by chance, or can the cancer cells acquire them sequentially, which would then induce a local microevolution process?
Any age. There is no right age to jump out of the window on the 10th floor, no right age to cross a busy interstate by foot, no right age to set a bed on fire. You wouldn’t allow a kid to do it (and similar things) at any age. Would you? :)
There are no goals in evolution. The most fit strain reproduces faster, eventually dominating other less fit strains. There’s no purpose, it just happens.
The "fit" itself is the goal. It is not like water flowing in a direction guided by the terrain. Trees produce thorns to keep away animals. You could say that thorns worked best among multiple options that were tried. But the question is, why is the tree trying out with option at all. Why should it protect itself from animals? Why should it grow and reproduce? A rock boulder, lying next to the tree, doesn't seem to have these goals or intentions.
The water flow has the motivation driven by gravitational force. What is driving the motivation behind the growth the reproduction of an organism? In fact, life, growth and build go against the principle of least action that governs the physical universe.
And I'm not sure why someone has evolved the intention of downvoting me.
The "motivation" behind growth and reproduction is that lifeforms that reproduce more will dominate, while lifeforms that do not die out. This is called natural selection.
Mutations in DNA can have a positive or negative impact on the ability to reproduce. Because of natural selection, positive mutations tend to be preserved while negative mutations tend to disappear. There is no intent behind this process; it is simply a consequence of how life works.
While we can't yet say for certain how life got started, it is likely that the first "life" was made up of self-replicating RNA. Because this RNA self-replicated and produced imperfect copies, it was already undergoing evolution at that point, where the RNA strands which were better at copying themselves (due to mutations) were more plentiful.
> It is not like water flowing in a direction guided by the terrain.
I understand it doesn’t seem intuitive at first, but evolution is actually quite similar to that.
> Trees produce thorns to keep away animals.
It's not that trees consciously produce thorns to keep animals away. Rather, trees with thorns (due to random mutations) are more likely to survive because they are less likely to be disturbed by animals. Over time, those trees are more likely to reproduce and pass on those traits.
Just like water flows down the path of least resistance, evolution naturally selects for traits that improve survival. The "selecting" we are referring to is used in the same sense that water "selects" which route it takes. It doesn't actually select anything, it's just following the laws of physics.
You are still anthropomorphizing. The tree doesn’t want anything. It is a structure of chemical reactions. Sometimes there are mutations that lead to new features. Sometimes these features confer success in the environment at the time. Sometimes that success leads to this new mutant taking over relative to the others. Fast forward billions of years and we go from a sludge of singe celled microbial life to every tree you can see.
Have you heard of Severely Deficient Autobiographical Memory disorder? For me it's memory as it relates to myself that barely functions, I can remember facts just fine.
And so we could follow the sex chromosome and mitochondrial DNA convention—you’d inherit the grandma’s surname from your mother and the grandpa’s surname from your father.
This variant could lead to a split of the set of existing family names into two disjoint subsets, one that is passed through females and one that is passed through males.
I do not think that this split could create problems, unless at the date when the system would be introduced there would happen to be a serious imbalance between the number of family names carried by females and the number of family names carried by males, which might be then preserved by the system.
In capitalism, power correlates with wealth. So wealth inequality yields power inequality. Too much inequality, and a society can hardly feel just. If we only could somehow fix capitalism so that wealth doesn’t correlate with power.
