Libretech | Systems

github.com/libp2p/specs... the local node can automatically ingest the files, generate immutable Content Identifiers (CIDs), and announce them to the immediate local area network via mDNS without ever exposing the data or instruments to the external internet.

Other researchers in the same room can immediately pull the raw files, custom analysis scripts, or shared local ledgers directly from the machine's cache using those

Some people think IPFS doesn't have to scale and get adopted enough to replace the HTTP web; they claim it can just be a cold "archive for rare datasets" or a niche backup protocol forever.

if IPFS doesn't become a living web protocol, then it is nothing but a slow, over-engineered

Some people think IPFS doesn't have to scale and get adopted enough to replace the HTTP web; they claim it can just be a cold "archive for rare datasets" or a niche backup protocol forever.

Always win? Their takes were nonsense to me at first because I had in my mind that everybody understood IPFS would eventually undergird a local-first, sovereign web, so there was no way this was a fragile experiment;

Some people think IPFS doesn't have to scale and get adopted enough to replace the HTTP web; they claim it can just be a cold "archive for rare datasets" or a niche backup protocol forever.

Because random content identifiers (CIDs) and cryptographic hashes stored on DHTs are just worthless bytes if nodes stop hosting them. If we stop making moves towards the only real goal (widespread, active dWeb adoption and active node participation),

Some people think IPFS doesn't have to scale and get adopted enough to replace the HTTP web; they claim it can just be a cold "archive for rare datasets" or a niche backup protocol forever.

The first thing is that these people don't really care about "archiving" anything; they just want a low-maintenance, static dump for data they rarely touch, so the discourse is already skewed from the start. Meanwhile, all the real utility