Archive Team projects

[ion1]

@ArchiveTeam seems to be awaiting a proposal from IPFS developers:

http://archiveteam.org/index.php?title=INTERNETARCHIVE.BAK/ipfs_implementation

[davidar]

Cc @jbenet

[jbenet]

@davidar @ion1 and anyobody else: would you like to help me with this? i dont have the bandwidth to tackle this at the moment. as i see it, we need the following (all of which are useful for our own archive needs)

• a written up proposal explaining how IPFS works, in particular describing versioning, swarming, api, programmability, benefits.
• living example with one of our archives (ideally a large one)
• easy way to make a webpage that uses a local ipfs (or even the gateway) to find number of providers for a given object.
• create an "Archive" ipfs object that points to a all the data, but which also has metadata (maybe this is our package.json thing).
• ipfs-cluster or something close to it ready, with RAID like behavior, for one massive pinset. not needed for something like their "multiple torrents" approach (multiple pinsets).

stretch goals:

• webpage that renders an "Archive" ipfs object with live stats, and links to explore the content, and easy instructions to get started in replicating stuff. this viewer is basically a simple distributedd webapp.
• make it easy to embed a "Pin it" button, almost like github stars.

If a few people are interested in helping out, we could have a quick discussion sometime soon to kick things off.

[davidar]

a written up proposal explaining how IPFS works, in particular describing versioning, swarming, api, programmability, benefits.

I could probably draft something over the weekend.

living example with one of our archives (ideally a large one)

Would arXiv work?

The remaining issues are things I'd really like to have anyway, and I think the Alexandria folks are interested in too (CC: @DevonJames)

Also relevant #25

[davidar]

@eminence has also been working on some of this stuff already

[jbenet]

I could probably draft something over the weekend.

thanks would be huge

Would arXiv work?

yep.

[rht]

(https://archive.org/details/arxiv exists, but its content isn't systematically added)