WASM & @ziglang fan. AS215820
Also, actually I disliked how Odin handled allocator/context. It's more pragmatic, sure, but just seeing "this function takes no allocator, therefore it doesn't allocate" is so nice.
Also, actually I disliked how Odin handled allocator/context. It's more pragmatic, sure, but just seeing "this function takes no allocator, therefore it doesn't allocate" is so nice.
Not strictly necessary, but it is what makes the E in SSE. Without that it's really just Server Streaming. Which is fine, just not really that technology.
Not strictly necessary, but it is what makes the E in SSE. Without that it's really just Server Streaming. Which is fine, just not really that technology.
SSE can be very fast, but what the LLM web pages usually do isn't real SSE, it's just very slowly streaming the text. Only a technical difference but yeah.
SSE can be very fast, but what the LLM web pages usually do isn't real SSE, it's just very slowly streaming the text. Only a technical difference but yeah.
All the live APIs are WebSockets
All the live APIs are WebSockets
You turn on your device, and your phone just... Shows it? One-click setup and one-click play are just so nice
You turn on your device, and your phone just... Shows it? One-click setup and one-click play are just so nice
I'm also not against just doing build numbers, and it's just build #628 - depends on if the date has any meaningful value.
I'm also not against just doing build numbers, and it's just build #628 - depends on if the date has any meaningful value.
Interesting to see that it seems to look pretty different on lots of devices?
Interesting to see that it seems to look pretty different on lots of devices?
A weird comparison, but like URL parsing. A web dev doesn't have to understand how it works, just use it.
A weird comparison, but like URL parsing. A web dev doesn't have to understand how it works, just use it.
You still have to implement replication, just don't deal with merges. But a stream of operation / states still has to be replicated.
Additionally CRDTs are commonly defined with a separate query layer that exposes the outside data structure.
You still have to implement replication, just don't deal with merges. But a stream of operation / states still has to be replicated.
Additionally CRDTs are commonly defined with a separate query layer that exposes the outside data structure.
I'm thinking this means that they get visibility into the entire TLD? Or just their specific government sites?
I'm thinking this means that they get visibility into the entire TLD? Or just their specific government sites?
All sorts of distributed systems in datacenters, and lots benefit from CRDTs.
Really, a lot of existing tech is extremely similar to CRDTs, just not labeled as such.
All sorts of distributed systems in datacenters, and lots benefit from CRDTs.
Really, a lot of existing tech is extremely similar to CRDTs, just not labeled as such.
Still, it's probably the best RTS. Ofc theres AoE etc, but that's a slightly different style of RTS. I never got into those.
Still, it's probably the best RTS. Ofc theres AoE etc, but that's a slightly different style of RTS. I never got into those.
If I save up to buy new shit, buy it on the credit card and pay it off the next month, that is still considered untrustworthy?
Classic case of "what they doin' over there?"
If I save up to buy new shit, buy it on the credit card and pay it off the next month, that is still considered untrustworthy?
Classic case of "what they doin' over there?"
Repeatedly going back and forth doing the same changes is so time consuming on large repos
Repeatedly going back and forth doing the same changes is so time consuming on large repos