many such as cases, it's clear why. People hate doing task x, because it's not fun nor valuable to them, but need to do because of duty.

Ironically all nuts friends seem to be talking less, less defensive than other nuts friends who keep talking about other people not buying what they believe nuts.

It's useless as data is not just some graphic semantic, they have relation, business rules on top, not ready to interact with if not all are ready, loaded.

It’s definitely not useless. You’re right that it requires the interpreting layer to be able to handle missing info. The use case at the end of the article is streaming UI. UI, unlike arbitrary data, is actually self-describing — and we have meaningful semantics for incomplete UI (show the closest loading state placeholder). That’s what makes it work, as the article explains in the last section.

Thanks Dan. Yes, I agreed on the ui part, it seems to work in most cases. Some html tags have relation like `<datalist>` or `[popover]` attribute, but if we make all kind of relations trivial then it's benefit for sure.

Yea, and also to clarify by "UI", I don't necessarily mean HTML — it could be your own React components and their props. In idiomatic React, you generally don't have these kinds of "global" relations between things anyway. (They could appear inside components but then presumably they'd be bound by matching IDs.)

Looks neat, the first thing I search for in the docs is:

    Unique indexes  Not currently supported. Requires query rewriting and separate execution engine to validate uniqueness across all shards.

But still looks promising.

Small consolation prize is we can generate unique primary keys: https://docs.pgdog.dev/features/sharding/primary-keys/.

I would like to implement cross-shard unique indexes, but they are expensive to check for every query. Open to ideas!

Aren't UUIDs the a priori collision-free index of choice ?

The problem is which version? Also does the cluster index need to be embedded? Or shard index? Or whatever you are using to track the storage of the record? Should we care?

a nice way to solve this is a hash ring and put it in the ID. you can re-shard the ring as you need to, and can route to where the object lives easily.

Honestly, it doesn't need to be local, API is some 200ms away is ok-ish, make it 50ms it will be practically usable for every majority of interaction.

Indie startup (or just simply call it "selling app online") is not bad if your "capital is lower than $1M" AND "desirable MRR at $10k". You can beat $1M capital, 10% return .. by making $10k MRR SaaS instead.

Yep, 2024 market is considerably too good though. When he about to quit, I'd expect 8% yield (pessimistic) per year, so $6.4k/yr or $533/mo, that's "good" living quality in non major cities in Thailand. But to be fair, it's only good for locals, not for foreigners. And that $80k also needs to be invested 1 year before quitting.

It's doable for extending up to 3x of his runway.

If you get 8% returns per year on average, the expected annualized return over a multi-year period is going to be LESS than 8%.

This is because you won't get exactly 8% each year. For example, suppose the returns over a 3-year period are: 20%, -10%, 14%. In this case, the return over the whole period is 23.21% (= (1 + 0.20) * (1 - 0.10) * (1 + 0.14) - 1). On the other hand, a 8% return each year would have resulted in a 25.97% return over the whole period (= (1 + 0.08)^3 - 1).

When people say x% average annual return this is almost always referring to geometric mean already, so the above math doesn’t apply.

tbh, the 2m app is more clunky than the OG 1m. Virtual scroll is very poor. Jump doesn't seem to work (it works just sometimes).

I feel like this is all missing the points. The point is React is mid. Engineering metric is very strange these days. A guy has made your tool wicked fast, dismissed - fast enough .. wat?

I wonder at what point Vision Pro is highly portable that we don't have to care about small screen ui/ux. Excited for a fancy excel app on this device..