But isn't he right? Indexing images is pretty intense job. You need to run a model, you need to make sure that the model is accurate enough, you need to constantly train it to recognize new things. If you have text on the image you need to OCR it, again with confidence. You need to implement translations to the search, etc... etc...
To the other hand for $55 you can have infrastructure to run your own media server and a relay giving you all the oversight over your own media.
It is about convenience, trust and control.
Aida
npub1aj…hdt20
2024-10-06 08:21:39
in reply to nevent1q…ue8y
Author Public Key
npub1ajdaw3j4g6aqv86alhn3df8jpulj0mxz3jjgwpm4uh598hc348gqthdt20Published at
2024-10-06 08:21:39Event JSON
{
"id": "e13b1808fae6d8f57847ff2427bb066f3df8563e6347155793b7928f4afcff80",
"pubkey": "ec9bd7465546ba061f5dfde716a4f20f3f27ecc28ca4870775e5e853df11a9d0",
"created_at": 1728202899,
"kind": 1,
"tags": [
[
"e",
"48d1b04d4b2d62dc42b45f1c18e19aa63eb619559f152f6a10ab0e8791e70f7a",
"",
"root"
],
[
"p",
"4eb88310d6b4ed95c6d66a395b3d3cf559b85faec8f7691dafd405a92e055d6d"
],
[
"p",
"8fb140b4e8ddef97ce4b821d247278a1a4353362623f64021484b372f948000c"
]
],
"content": "But isn't he right? Indexing images is pretty intense job. You need to run a model, you need to make sure that the model is accurate enough, you need to constantly train it to recognize new things. If you have text on the image you need to OCR it, again with confidence. You need to implement translations to the search, etc... etc... \n\nTo the other hand for $55 you can have infrastructure to run your own media server and a relay giving you all the oversight over your own media. \n\nIt is about convenience, trust and control. ",
"sig": "412dcda86a681c4471c80ad7e7bda93a7eed5bcdf6d44f7ca90bafeb8e31aec8dc810718faaf9280fe8d69a20226d17b7cd88531d235e61175d24eb1a1c32dfb"
}