I used to do this purely so that I could attach a name, then I found out that you can add a message when stashing: `git stash push -m "trying a thing"`
Stash is just a stack of commits. If you use stash this way too much probably you will mess something, since it needs to keep that stack data structure.
The solution: create a branch or tag with the things that you're trying. If you want to apply it, use `git merge --squash`. This way your unfinished work lives outside the stash stack!
git stash is given the UX of a stack, but it's more a list or a set of commits. The commits in your git stash don't need to be related (their commit parent pointers can point at very different branches). At some point if you are heavily using lots of stashes you tend to switch to referring to them by commit message and/or stash number instead of thinking it a stack you build bottom to top and always and only pop from the top.
- If the server is responding with JSON, the client has to parse, decode, and validate the JSON. Then it has to transform it into HTML DOM nodes. Finally, it can inject them into the DOM.
- If the server is responding with a bespoke non-JSON data structure, it's the same situation as above except even more work because a third-party, non-standard library has to decode this structure into something (most likely JSON) and then transform the JSON into HTML nodes, etc. etc.
- If the server is responding with straight-up HTML, the client uses the browser's HTML parser to directly transform it into HTML nodes and inject it into the page. And by the way, the browser engine is essentially a function that is highly optimized for exactly this use case–parsing bytes into DOM nodes.
'We' might use the same API but it's widely accepted that generally the same API doesn't work for all contexts. See https://samnewman.io/patterns/architectural/bff/#general which details the many well-known problems with having a single API for different contexts.
Yes but isn’t this the beauty of decentralized protocols? The cost is spread amongst all, it doesn’t require one entity to shoulder the cost for everyone.
It’s not really decentralized in that way. If you’re loading a bunch of pages in your mobile browser, your device is paying the power consumption cost of deserializing, parsing, and running all that JS. It doesn’t save on that because some other user also did it.
Yeah but when you're reading the actual source code you're not thinking in terms of bytes, you're looking and files and line counts. Htmx is deliberately maintained as a single, no-dependency file of about 5000loc. The maintainers are of the opinion that it reduces complexity, and I tend to agree.
Monero ideally combined with something like this but one or the other would also be an improvement:
>Brave uses a blind token protocol based on Privacy Pass, which decouples payment identity from service usage. These privacy-preserving subscription credentials allow the browser to verify you have a valid purchase of a premium product (like Origin) without learning anything about you.
It depends. If you have a Claude Code subscription then it defaults to an hour, if you use Claude via an API or third party then it defaults to 5 minutes. You can opt-in to the 1 hour TTL but it obviously costs more.
> 5-minute cache write tokens are 1.25 times the base input tokens price
> 1-hour cache write tokens are 2 times the base input tokens price
reply