There's a subtle but firm distinction between augmenting human productivity and augmenting human intelligence. The first is an economic, capitalistic endeavor first (though productivity contributes indirectly to collective intelligence); the second is a more pure pursuit, I think.
Noteworthy features of Starlark
Starklark is Bazel's configuration language, and not designed to be general-purpose. Nontheless, there are some features that seem useful even for a dynamic G/P language.
- Single final assignment at the top level. Module-global functions, variables cannot be re-bound. This helps reading code, and simplifies tooling.
- Deterministic iteration order for dictionaries, and in general determinism (a program run twice always produces the same outcome, modulo things like time). Determinism seems like a generally desirable property, for things like testing/reproducible builds.
- No mutation of iterator during iteration. Mutating the iterator (like a list) during iteration panics the program, to avoid iterator invalidation errors.
- No [checked] exceptions. Panicking the program for any non-anticipated error might seem problematic, but "makes the language simpler and reduces the number of concepts." Exceptions also become API surface for the language, so not having it helps language evolution.
- Strings are not iterable. This avoids bugs from passing in a string instead of a length-1 list of strings to APIs expecting a list.
Starlark, a configuration language for the Bazel build tool, has a tree-walking interpreter implemented in pure Go at google/starlark-go.
It seems like the canonical implementation of Starlark is the Java implementation in the Bazel source tree, but the Go version is used "in production" in web playgrounds, debuggers, and so on.
starlark-go is notable because it's one of a vanishingly small number of production language implementations that are tree-walking interpreters rather than bytecode VMs. The implementation guide says:
The evaluator uses a simple recursive tree walk, returning a value or an error for each expression. We have experimented with just-in-time compilation of syntax trees to bytecode, but two limitations in the current Go compiler prevent this strategy from outperforming the tree-walking evaluator.
The details of why exactly that's the case is interesting, and documented further in the link, but seem inherent to Go's current compiler design and philosophy. It also supports Oak's current (tree-walking) evaluator design, which is nice.
Typing on a typewriter for a while and then going back to a shallow laptop keyboard is a trippy experience.
How happy is the blameless vestal’s lot!
The world forgetting, by the world forgot.
Eternal sunshine of the spotless mind!
Each pray’r accepted, and each wish resign’d.
— Alexander Pope
Now that I have much more flexibility over my work/life cadence working solo (and with a healthier sleep schedule), it seems like I'm settling into a pretty consistent schedule of:
- Early morning / before meetings on deep work, building, writing, thinking
- Meetings in the early afternoon
- After lunch / evening on reading, research, gathering raw materials and letting them simmer
and it feels good. It lets me "end the day" whenever I feel tired or ready to end the day, without having to rush to complete something, and it seems like my mind is clearest and most ready to work after I wake up when my mind is clear. I'm trying to stick with it.
Future (desktop) operating systems — a collection of inspirations
- Artifacts, which is interesting for its document-focused design, the way it implements transclusion ("Links with context"), and a horizontal arrangement of documents as a way to organize workflows.
- Mercury OS, which has a striking visual design and showcases what a workflow and information-focused design for productivity could look like.
- Desktop Neo, which has a nice implementation of horizontally scrolling panes that connect apps together into workflows.
- Alexander Obenauer's Itemized OS, which reimagines personal computers to be more focused on information and open metadata to allow apps to extend each other around the information that matters, rather than organize into walled gardens.
What would this look like in 500 years?
Whenever I'm thinking about the long future of some technology or problem, like the future of AI and the nature of computation, there's a thought-experiment format I like to use to think more openly. It goes like this:
What would X looks like 500/1000 years into the future of humanity?
In the case of that blog, the question was, "What would computing hardware for AI look like in year 2500?"
I think this kind of thinking helps avoid the status-quo bias that's so easy to fall into when we try to imagine the future projecting out from current technology.
- If we project out from the current computing landscape, it's easy to think that binary logic gates on silicon wafers is the right computing substrate for nearly everything we'd want to do. But would humanity still be using silicon wafers to run intelligent, conversational, omnipresent computers five centuries from now?
- It's easy to get sucked into the allure of gene editing in the next few decades and century, but what capabilities over biological processes will humanity have in a thousand years? Surely, we won't just be adding and splicing DNA. Surely, the long future of dominion over the biological source code is something much more expansive, perhaps the equivalent of "software engineering" for biological processes, perhaps a world where there are more artificial life forms in the universe than organically evolved ones.
3 phases of creative production — a hypothesis about the structure of a creative process
Gathering raw materials
- Through reading and consuming media, listening to others, thinking aloud in conversations, or just pure contemplation, we gather a bundle of ideas and related experiences that start to point to some insight or further idea.
- A lot of creative value comes from this very unstructured stage of creation where ideas are gathered without any goal directed work. I call this "simmering".
Active thinking
- In the active thinking phase, the we look at the gathered raw materials and try to form structure out of them to produce something that's coherent and clear and perhaps even actionable.
- This implies and requires a "sense-making" process, of going from unstructured information to emerging a structure out of it.
- In the process of active thinking, we might find the need to collect even more raw materials. But this time, we're looking for information in a more goal-directed way. Information may also be discarded or deemed no longer relevant.
Production/drafting
- For us to create some creative work/output from ideas or insights gathered during active thinking, we need to crystallize it into a form that can facilitate communication. I call this "drafting" or "production".
- This stage is about serializing thought into media. Production may often also serve the role of communicating with ourselves, i.e. when taking notes for my future self, to remember my past thoughts.
See also: Creative shadow-casting
Resilient creative processes
Sometimes when I have a dozen tabs open (which is far too many compared to my normal amount of 2-3) and a bunch of stuff lined up to read, I wonder: if I miss these ideas that are open before me this time around — like, if I forget about what I read, forget to read them, or misunderstand them — would I encounter these ideas again? If I don't cross paths with these ideas and the ideas they will lead to now, will those ideas cross paths with me again in the future?
Obviously, this varies on the type of information and topic, but I think it's an interesting idea to entertain, that I'm calling the resilience of a creative process.
In a resilient creative process, similar ideas emerge in dense, connected forests over and over again, so missing any single idea once doesn't influence the general path I take through my idea maze.
In a brittle creative process, connections between ideas are sparse, but ideas could cover a wider ground together. The cost of this breadth is that missing an idea somewhere might mean you don't hit upon that one insight that you need to complete the next step in your work.
Recognizing this spectrum, it might be interesting to intentionally try to make a creative process more resilient or more sparse, depending on the subject matter.