Sprint: Aug 22 → Aug 28

[flyingzumwalt]

Sprint Aug 22

For information about these calls and how they work, read the ipfs/pm README.

Calendar

Endeavour	Lead	Time (PDT - UTC/Z - CEST)	Pad
All Hands Call	@flyingzumwalt	9:00PDT 16:00Z 18:00CEST	all-hands notes
js-ipfs	@diasdavid	10:30PDT 17:30Z 19:30CEST	js-ipfs notes
go-ipfs	@whyrusleeping	11:00PDT 18:00Z 20:00CEST	go-ipfs notes
libp2p	@diasdavid	11:30PDT 18:30Z 20:30CEST	libp2p notes
interfaces repr	@jbenet	12:00PDT 19:00Z 21:00CEST	interfaces-repr notes
mediachain	@jbenet	12:30PDT 19:30Z 21:30CEST	mediachain notes

Please add any agenda items you have to the pad before the project-specific sprint call starts.

[flyingzumwalt]

@parkan has proposed a Mediachain collaboration call (see #145). Should we add that to the day's agenda? Discuss it in the all-hands?

[flyingzumwalt]

Adding coveralls & code coverage to the agenda for all-hands. Just want to check in about where we're at with this.

[jbenet]

I propose this discussion:

• "Representing Interfaces, in libp2p doc repos and more" specifically modules like: https://github.com/libp2p/interface-peer-routing. require(@diasdavid) hope(@whyrusleeping)

doesn't need to happen this week, nor on monday, but best for it to.

[daviddias]

Sounds good @jbenet :) For everyone following, here is the list of interfaces already started/written:

• https://github.com/libp2p/interface-transport
• https://github.com/libp2p/interface-stream-muxer
• https://github.com/libp2p/interface-connection
• https://github.com/libp2p/interface-record-store
• https://github.com/libp2p/interface-libp2p
• https://github.com/libp2p/interface-peer-routing

And for reference, here is the one we use for IPFS itself:

• https://github.com/ipfs/interface-ipfs-core

[flyingzumwalt]

Added "Representing Interfaces, in libp2p doc repos and more" to the all-hands agenda, which might just involve identifying when/where to push that process forward.

[flyingzumwalt]

Added "IPFS Tutorials in Gitbook form" as an agenda item for the all-hands. I'll be proposing this format for assembling the new Tutorials (and gathering/re-formatting the existing ones) See: https://www.gitbook.com/book/flyingzumwalt/decentralized-web-primer/details

[ghost]

@whyrusleeping @Kubuxu @jbenet and I should talk about a vulnerability report we got, maybe the 17:00Z slot?

[parkan]

Hello from team mediachain! Would love to get the following on the agenda:

• large dataset providing: chatty provide on add, direct improvements, workarounds (add w/o provide, alt. routing schemes)
• dataset add w/o copy: (re: avoid duplicating files added to ipfs kubo#875)
• IPLD status/work needed (we want to help out!)

[jbenet]

@parkan great! would a call at 3:30p work for you?

[parkan]

@jbenet sure, works for me

[jbenet]

ok great. we'll add that call at 3:30p. I updated the schedule: #146 (comment)

[parkan]

@jbenet what do I dial into? should I get the url from IRC?

[RichardLitt]

Yeah, I'll throw it in IRC.

[parkan]

Good chat, cleared a lot of things up, have a path forward now (short-term). Thanks again!

One small thing: @vyzo is in EEST and having him participate in the future will be very valuable, is it possible to move the call up? We can likely do another day too, if the earlier hours are too packed.

[kevina]

@parkan @whyrusleeping I was looking over the notes to the mediachain call and I would like to clarify one point concerning ipfs/kubo#875: I have only really broken compatibility once due to the datastore key change and I provided an upgrade path. (In the very early stage I might have changed the format a bit more, but that was well before I created ipfs/kubo#2634.)

[jbenet]

@kevina thoughts whether CIDv1 or your PR should land first? if CIDv1 simplifies a lot, worth waiting?

[kevina]

@jbenet, I would like to see my PR land first, CIDv1 will only simplify a small part that is now well isolated. My PR requires a number of infrastructure changes that I will like to get in ASAP.

[whyrusleeping]

@kevina can you outline somewhere a revised list of tasks we need to do to get your PR in? things have changed a bit since you last updated your wiki landing page

[ghost]

@lgierth todo

• libp2p
  • add packet switching tasks to libp2p roadmap
  • update multigram spec draft
    • move spec to multiformats/multigram repo
  • start go-multigram implementation
• go-ipfs
  • make docker image work with go1.7
  • fix branch naming of hub.docker.com/r/ipfs/go-ipfs
• misc
  • work with an external contributor
  • prototype of ipfs-pages
  • review multihash RFC
  • fix ipfs.io/contact
• org
  • work through email backlog
  • get up-to-date regarding project management, and comms
  • send out new monthly infrastructure report (bling bling)
  • set up waffle board for infrastructure
  • wrap up keythief postmortem
  • take notes about infrastructure role

[kevina]

@whyrusleeping yes I plan to in the next day or so

[daviddias]

js-ipfs goals for sprint

• Finish the pull-stream endeavour (@diasdavid, @dignifiedquire) - refactor: ⎆ Migration to pull-streams (from Node.js Streams)  js-ipfs#403
  • Pass all the tests
  • Solve issue on spdy with pull-streams
  • Add documentation (@RichardLitt, can you help us with this?)
  • Create JS knowledge base in ipfs/community with a FAQ and why we use pull-streams (link to the issue where the discussion happened)
  • Prioritize bitswap tests to pass with pull-streams
• Create a Connection Upgrade for half-closed streams - Similar to the length-prefix stream, just create a thing that encapsulates every packet with one byte, so that we have a way to signal that we are not going to send more data. Useful for libp2p-websockets and other transports that miss this feature.
• Drop the size of js-ipfs-api by shaving off some dependencies (@dignifiedquire)
  • Bundling js-ipfs and js-ipfs-api for the Browser js-ipfs#429
  • [technical exploration] js-ipfs bundle optimized js-ipfs#398
  • Bundling js-ipfs-api ipfs-inactive/js-ipfs-http-client#353
• Activate Code Coverage and CI in all the JS libp2p + ipfs repos (@RichardLitt)
  • https://project-repos.ipfs.io/ (doesn't show the libp2p ones)
• Moar testing for js-ipfs
  • Network tests test: chaotic network tests js-ipfs#299 (@diasdavid)
  • HTTP-API tests Use interface-ipfs-core tests over the http-api with ipfs-api js-ipfs#428 (@diasdavid)
    • Create factory-http
    • test-block
      • upgrade core
      • upgrade http
    • test-object
    • test-files
    • test-config
    • test-swarm
      • upgrade core
      • upgrade http
  • Tests that move a tooon of data over js-ipfs bitswap
• interface-ipfs-core
  • bitswap
  • dht DHT API ipfs-inactive/interface-js-ipfs-core#36
  • mfs Files, the Mutable File System API ipfs-inactive/interface-js-ipfs-core#38
  • name name API ipfs-inactive/interface-js-ipfs-core#37
• Finish up the level-up js-ipfs PR level-up js-ipfs#420

Notes from the Call

[daviddias]

libp2p goals for this spring

• Finalize libp2p roadmap Update project roadmap libp2p/libp2p#2 (@diasdavid)
  • Add more detailed and quantifiable milestones, tag them to the milestones of js-ipfs and go-ipfs, if needed
  • Set expecations (what will be done this quarter)
  • @lgierth to add the multigram goals
  • @jbenet and @whyrusleeping to review
• Make DHT "Providing records" faster
  • Create an issue with the ongoing work to make it faster, let others that are interested know (Mediachain) so they can follow the dev (@whyrusleeping)
  • "disable providing with ipfs add --local" (@whyrusleeping)
• Review and give feedback on PubSub Harness (@whyrusleeping, @jbenet ) - Update 1: Kickoff gavinmcdermott/js-libp2p-pstn#1
• libp2p.io (@RichardLitt)
  • Set up a discussion, define goals (@RichardLitt, @diasdavid)
  • Open issues with the goals so that they are trackable
  • Check the work that @kyledrake started and take it from there - New design for libp2p-website libp2p/website#6

[RichardLitt]

What sort of documentation, @diasdavid?

[daviddias]

@RichardLitt See:

• Node.js Streams vs pull-streams vs ReactiveJS js-ipfs#362 (comment)
• refactor: ⎆ Migration to pull-streams (from Node.js Streams)  js-ipfs#403 (comment)

[parkan]

@diasdavid I think the DHT point is missing "disable providing with ipfs add --local"

[daviddias]

Thank you @parkan, added :)

[dignifiedquire]

@flyingzumwalt @RichardLitt can this be closed?