> The honest way to say this is that Fable is not useful for bio-related work.
It is way worse than that. Try "How does digestion work?" and you will see "Fable's safeguards flagged this message". It's a stupid rate of false positives.
I wanted to use Fable to discuss a philosophical topic, but halfway through I used the word "cell" and got deflected to Opus.
On the other hand Opus has this awful adversarial-teacher vibe. It pushes back for no useful reason, talks down to you, and acts like it has to prove itself by grading and correcting everything. Instead of working with your claim, it reframes it, declares what the "real" issue is, then tells you what you failed to do.
So Fable refuses me and I can't stand Opus. Nice one, Anthropic, I need to downgrade my subscription.
We can switch in a heartbeat to a competitor, what kept us to Claude was that it had better models for a while. Closing Fable off means they are squarely inferior now.
For me Opus 4.8 was a slow model with a strong habit of talking down to me in an obnoxious way that would not be possible to prompt away. GPT 5.5 is now my main driver for serious work.
The search engine is indeed the last missing component from a sovereign stack. But I think this could be solved locally with little cost. Instead of indexing content on the web we should be indexing sources themselves - where to look for X? - like forums, blogs, docs, feeds, and specialized search engines. We could collectively amass millions of these search stubs that can be used by local models to go and fetch fresh information from the source directly. This means separating the routing layer from the information layer, we don't need to keep information cached from the whole internet locally. The search stubs could fit in a few GB about same size with the local LLM. The cool thing is that sources change much slower than information itself, so the search stub database could be refreshed at a slower pace. We could combine a few million generic stubs with a few hundred personal stubs generated from our own activities. It is trivial to generate these stubs by piggy backing on frontier models.
I think the harness and local context should supply that missing piece between general model and bespoke application. Each application has its own context and action quirks that don't generalize well. Maybe it's just 5% but that is genuinely specific. So its rightful place is in context engineering.
Meditation can also hype up your anxiety depending on individual circumstances and personality. And it is by no means easy, you have to invest much effort to get results. It's like refactoring your code base (mind).
I thought it was going to be even shorter leash - code autocomplete with smaller local models. That raises the level of interactivity and leads to better code knowledge.
I think the obvious solution here is to beef up the test side of the app, much more than when writing code by hand. Tests represent project knowledge in executable format. The LLM does not need to be careful to remember every detail of the tests. You don't need to vet every small interaction, it automates review work as well.
Even better if the project was built from the start to be easier to test and observe. But my golden rule remains - no code without tests, expand test suite all the time.
> When you speak, what comes first, the idea or the word? Do you first feel a thought inside you, and only after that go searching for the right word to wrap around it?
Let's take a feral human, do they achieve the same with language powered humans? No, obviously. It means language does something we can't replace with "consciousness". Maybe something life preserving, putting consciousness downstream of language use.
But more generally - how can 8B humans make a living on this planet? Not without language, that is for sure! We long passed the stage where we could exist without language at our consumption rate. We can't even exist without math, population collapse would be the outcome.
Minds are downstream from language & math. For example, is music downstream of piano or piano downstream of music? I think you can't cut this cleanly. They developed in relation to the other, recursion between tool and art form is already old.
Plain old markdown with links is exactly what I use. But I split memory in 3 parts
1. a log of user comments, all of them collected in a chat_log.md file which is used to validate the agent still follows user intent
2. task level memory, a task.md file starts as intent, becomes plan, workbook with inlined results for each gate, and finally remains as documentation; same task.md is passed to judge agents and back; a project can have 100+ tasks logged in, the task is the natural unit of work
3. a project state mind map which tracks current known value of all decisions and intents, it is used to start the agent new sessions and updated after each task
This is it, a memory of user intent, one of agent work, and one of project state. I used this setup in over 100 projects. My harness uses hooks to constrain the agent
1. can't edit code files without an open task
2. the next current gate ("[ ] plan item") is displayed in the post tool use hook as an instruction pointer, auto-advances as current gate gets checked; this keeps the agent straight even in task.md files with 100s of gates
3. can't close a task until all gates are solved
4. a post user message hook appends to chat_log.md, which is cheap
The harness can be refined by optimizing task.md plan structure for various types of tasks, based on reflection. We can do reflection because we keep the 3 memory types.
Memory works for my harness because I don't split the place where memory sits from the place where work gets done and tracked. The plan items are this place. A single gate starts as intent and ends as a log like of work done. The task.md file is edited as it is solved.
Even if you generate the soundtrack, if you set the lyrics it can be great music. At least for one person, who chose the lyrics. I set some poems on music and I absolutely love the results.