Daniel Wigton on Nostr: My idea is to build a wasm based VM with a distributed filesystem. Then over time add ...
My idea is to build a wasm based VM with a distributed filesystem. Then over time add distributed io and compute. Imagine having your code editor available and running on all your devices without having to sync your code. Just move from laptop to desktop and pick up where you left off. The VM even moves the running code to the most effective core across all your machines.
That way compiling from your phone would automatically use your desktop etc.
It gets even more powerful if you have trusted contacts with mutual resource sharing agreements. They can act as an off-site back up as well as giving everyone with enough friends access to supercomputer like performance for highly parallel (and not too sensitive) workloads. Blender rendering for the masses!
Published at
2024-08-07 20:34:43Event JSON
{
"id": "f0897707519a3f11ddb3c61c352868da8fed74bbc1d88dbe5d306281e1fccf84",
"pubkey": "75656740209960c74fe373e6943f8a21ab896889d8691276a60f86aadbc8f92a",
"created_at": 1723062883,
"kind": 1,
"tags": [
[
"e",
"aab50e0c1a1e653bd938730fb721ba3e662f08e26844a87bea4f01d4d64b0c28",
"",
"root"
],
[
"e",
"496b4c496c7885b140878cb8a93160bf4e0171541ca80e1e7903e579b6a65ecb",
"",
"reply"
],
[
"p",
"9a4acdeb978565e27490dca65c83e9f65745eaec1d9a0405a52d198c1489913b"
]
],
"content": "My idea is to build a wasm based VM with a distributed filesystem. Then over time add distributed io and compute. Imagine having your code editor available and running on all your devices without having to sync your code. Just move from laptop to desktop and pick up where you left off. The VM even moves the running code to the most effective core across all your machines. \n\nThat way compiling from your phone would automatically use your desktop etc.\n\nIt gets even more powerful if you have trusted contacts with mutual resource sharing agreements. They can act as an off-site back up as well as giving everyone with enough friends access to supercomputer like performance for highly parallel (and not too sensitive) workloads. Blender rendering for the masses!",
"sig": "f79c6f22dea61f81cdf32bd2f5d77a9370e0fb8bbcbb3c50dfbc846e96b1632eba94f0967fdc6a51660518b1c4f32ba48093eddeec252f3e24fe7bd6b90cd4f7"
}