> There exists a higher level problem of holistic system behavior verification
This is the key observation. Strict, code-level verification of every line, while valuable in the small, doesn't contribute very meaningfully to understanding or building confidence in higher level system properties.
I see a future of modules with well defined interfaces and associated contracts. Module level contracts can be formally verified or confidently believed via property based testing. Higher level system behavior can then be proven or tested by assuming the module contracts are upheld.
Module boundaries allow for varying the desired level of confidence module by module (testing, property based testing with variable run time allotments, formal verification).
LLMs can handle each step of that chain today, allowing humans to invest specifically where we want more confidence.
Modeling workflows as state machines is incredibly powerful.
In addition to everything the author mentioned, the constraints of state machines allow a workflow platform to provide a ton of additional guarantees and capabilities around consistency, state propagation, reliable timers, inter-instance messaging, etc.
We built our workflow execution platform [1] around state machines and we've seen great results. We find our workflow code is incredibly simple and easy to understand.
Rust ADTs and pattern matching are so much better than other mainstream languages I find that once my code compiles it actually is almost always correct.
The next step is to encode your transition logic in the From impls between the enum structs and you've got yourself a first-rate state machine.
It's very similar to a DAG you'd make for a workflow but not necessarily acyclic and events can be sent from external sources as well as internal processes.
States can be grouped into parent states, which gives you some additional nice things like parallel states and also a totally reasonable way to zoom out and hide info in visualizations.
When you have a pure function or a single effect, that should definitely just be a function rather than its own machine. Sate machines are most useful for orchestrating those effectful functions and for situations where you want to accept external events. So you're still writing regular code but you're writing it in small functions that get invoked by the state machine as needed.
That is, state machines don't replace code. They're just a really nice way to organize it.
Great point! State machines are a really nice way to orchestrate multi-step LLM invocations.
They're also a great way for LLMs to produce code. It's human-readable so you can vet that it's doing what you want it to and it's high-level enough that the remaining bits to fill in tend to be small functions that AI can easily generate.
So happy that others are getting behind this idea. Protecting critical backend state by establishing a state machine that owns it makes so many things so much easier to reason about.
Recently opened up access to a backend as a service [0] for running arbitrary state machines as reliable workflows or real-time backends. After building a bunch of systems on top of this, I'm definitely convinced that this is a better way to structure a lot of systems.
We built https://www.teampando.com out of a similar love for state-based thinking. We turn natural language requirements into state machines, have a Figma plugin to connect designs to states (didn't realize sketch.systems had embeds that can support that - very cool!), and can export XState state charts.
State-based thinking is already helpful just to get your thoughts organized as you're building something but it's super helpful to give everyone (eng and non-eng) on the team something concrete to talk about and play with.
Unfortunately, much of the advice in vouge today argues against thinking about ontology. That advice comes from an epistemological stance that you can't know the right ontology beforehand so you shouldn't even bother trying. Instead, I would argue, in those situations, you should design for the ability to easily evolve your ontology and system.
Sorry, more investigation prior to mitigation than mitigation. I think moving the lambda aspects of sst to a lambda extension would allow for a workflow like this:
1. Find production issues
2. Replicate production issues in user facing tools
3. Run a local `sst start <fn> --filter '.userId == 'my-user-id'` to have remote <fn> connect to my local for any requests that match `evt.userId == 'my-user-id'`
4. Replicate issue and debug with the request sent to this particular lambda, potentially many calls away from the user action
5. Patch and deploy.
Being able to debug in situ would be fairly incredible in a complex application.
None of that is meant to take away at all from what you've built. What you have here is really awesome and I'm excited to use it.
This is amazing. Congratulations! I've been thinking along similar lines and so happy this exists.
Have you considered extracting the lambda portion into a lambda extension so that you can deploy this as part of any cdk/cloudformation stack or standalone lambda and toggle on/off local debugging based on some registration in dynamo (e.g. store a predicate applied to the event and proxy to local if the predicate evaluates true)? This would allow something like this to be hugely useful in mitigating outages.
I didn't catch it in the repo yet but I'd you're proxying local AWS calls through the lambda to allow quickly catching IAM and security group/routing issues, that would be incredible.
Happy to help with this stuff and will definitely be a user!
Our galaxy has ~10^70 atoms. At an atom per compute element, we will use 1/10^10 of our entire galaxy for our own compute? 10^60 is a bit more than the number of atoms in our sun. Given our sub-exponential space advances over the last 50 years, it seems... ambitious... to turn our sun into a computer in the next 300.
If they're saying we'll be an interstellar spacefaring civilization in much less than 300 years, they really buried their lead :)
10^60 computing elements (presumably doing some computing that we humans want done) with only ~10^80 atoms in the universe? I'm as optimistic a technologist as they come but that seems a bit... extreme
As RyanZAG says, all production has diseconomies of scale in unit complexity. The production process has economies of scale, meaning that churning out more units of equivalent complexity reduces the marginal cost of churning out the next unit of equivalent complexity.
Software exhibits this same economy of scale in production. Take Google's machine learning platform. They allow multiple functional teams to churn out roughly-equivalently-complex machine learning-powered widgets in less and less time. Contrast that with a startup building a single machine learning-powered widget and the marginal cost to Google is significantly lower.
I definitely agree that we'll always need people who think like programmers. We can develop tools in the vein of those you mention to significantly enhance the productivity of those people though. I haven't seen many tools like that that help in distributed systems or that allow one to easily visualize and understand an entire system.
> There exists a higher level problem of holistic system behavior verification
This is the key observation. Strict, code-level verification of every line, while valuable in the small, doesn't contribute very meaningfully to understanding or building confidence in higher level system properties.
I see a future of modules with well defined interfaces and associated contracts. Module level contracts can be formally verified or confidently believed via property based testing. Higher level system behavior can then be proven or tested by assuming the module contracts are upheld.
Module boundaries allow for varying the desired level of confidence module by module (testing, property based testing with variable run time allotments, formal verification).
LLMs can handle each step of that chain today, allowing humans to invest specifically where we want more confidence.