ipfs

are there any good ipfs explorer tools??(eg. put in a cid and see its content) have just been using explore.ipld.io so far

another ifps thought: is there a good way to use ipfs as a super lightweight public database? if I have a collection of non-sensitive data that’d literally only take up a few hundred rows in a db, is there a lightweight way to either continuously update a set of related files or use some sort of other solution on ipfs?

a few years ago I was pretty into what it would look like to use tools like ipfs to store data meant for applications, but stopped being as interested in/staying as updated on that subject after a while

*ipfs not ifps 🤦‍♂️🤦‍♂️

cofounder @canvas, prev @common, raymond.xyz

a peer-to-peer runtime for distributed applications

we tried using ipfs from the browser initially too, but it didn't really work because peer topologies change too fast for the dht (chrome nerfs background tabs). so you need to sync to a server, which breaks many trust assumptions unless you sync signed data

if you have a desktop app, iroh is excellent, it's low-level but great for things like syncing video / large files

but otherwise that leaves cryptographically authenticated sync engines like @canvas - we have something that's worked well in a few very custom environments (300k row db for common.xyz). now polishing up the out-of-the-box dev experience + doing a ton of iterations based on dev feedback.

there will be a couple of production apps using it by end of june. it's possible to use for apps with small data now! eg https://forum-example.canvas.xyz/ was two hours / 50 LOC to set up

cc @rz curious if you have any thoughts here and also how things are going with @canvas !

It’s certainly possible! The hard part is writes since IPFS is immutable. I took a stab at it with IPCM by recording DB dump CIDs onchain. Doesn’t work well with CRDT but still interesting

DX Engineer | DevRel at OpenZeppelin | Building orbiter.host | stevedylan.dev/links | opinions are my own

It’s certainly possible! The hard part is writes since IPFS is immutable. I took a stab at it with IPCM by recording DB dump CIDs onchain. Doesn’t work well with CRDT but still interesting 

https://pinata.cloud/blog/building-ipdb-a-decentralized-database-using-base-ipfs-and-pglite/

I have been using ipfs in all of my projects, so created an api to call pinata. Not sure if you are looking for something like this?

I've seen at least two database on ipfs implementations. Every time you change the database you have to update your site's hash which isn't a great flow. And those updates are centralized. I think using a small l2/l3 contract as a database is maybe a better option. What exactly are the rows and how often do they update?