# IPFS All Hands Call January 9th

**Moderator:** @RichardLitt
**Notetaker:** @diasdavid
**Attendees**
- @RichardLitt
- @dignifiedquire
- @keks
- @haadcode
- @goldani
- @victorbjelkholm
- @diasdavid
- @frankpetrilli
- @flyingzumwalt
- @jonnycrunch
- @lgierth
- @kumavis
- @Kubuxu
- <!-- add your gh handle here -->

## Agenda

<!-- Ensure notetaker is present before you begin -->

- Roll call, by [timezone from east to west](../admin-guides/timezone-rollcall.md).
- Call for additional agenda items (moderator)

- Q1 Roadmap (@flyingzumwalt: everyone)
- Roadmap in a Waffle Board: https://waffle.io/ipfs/roadmaps
We will be working in 2 week sprints this quarter
The schedule can be checked at the new waffle board https://waffle.io/ipfs/roadmaps
We will still be doing small endeavours, known as "sprintinos", which shouldn't take a full sprint.

- New Sprint Structure (sprints, sprintinos) -- how to distinguish between weekly "sprint calls" and calls for a specific sprint?

- Stale Endeavors https://github.com/ipfs/ipfs/pull/224/files ([rendered](https://github.com/ipfs/ipfs/blob/b0c3f43b466747d461a085f49527877f450944c3/ROADMAP.md#stale-endeavors))
- We now have a process to make sure endeavours don't fall into oblivion, see issue above for all the details
- [ ] Captains should be writing down the roadmap for the endeavours that have been declared as 'stale'
- If you care for one of the 'stale' endeavours, you should feel encouraged, given the roadmap, to take the lead of that endeavour you care about

- Uptick in Discourse Traffic (@flyingzumwalt: everyone)
- More people have joined the discussion in discourse
- Some of the questions popping in discourse are duplicated of questions in FAQ
- [ ] Investigate if we can have Github Auth for discourse

- Help @goldani figure out how to contribute to IPFS
- @goldani Fresh perspective from learning IPFS https://github.com/ipfs/specs/issues/145
- check issues in the ready column https://waffle.io/ipfs/js-ipfs with 'help wanted'

- 1st week of February, Seattle DApp Hackathon https://github.com/ipfs/community/issues/188

- What's next
- Session for sprint people to check in with each other (1 hour max)

-----------------
## SPRINT CALL NOTES

### Data Refuge Hack

@flyingzumwalt attended a hackathon organized by climatemirror. DataRefuge hack. Community-driven efforts for saving climate datasets that may get turned off due to a change in administration. Even if we didn't care about this data personally, politically, what have you - this is a clear example of centralization hurting. Entire fields of research depend on datasets that have silo issues.

@flyingzumwalt published an article about this on Medium recently; at the Hackathon, @flyingzumwalt got to talk to people about what it looks like on the ground to backup data on IPFS.

So! @flyingzumwalt is putting up a sprint for downloading _all_ of data.gov, around 300TB. The goal is to a) preserve it on their systems and b) put it on IPFS! This will help the data be stored, shared, and accessible. This is a great load test for our system, as well as just being good for the datasets we are trying to save. This will improve our UX, our perf, and how filestore (index the content in place rather a copy) works. 300TB is over the threshold where deduplicating data becomes super important.

Goals:
- Can we get filestore merged?
- Can we launch a basic ipfs-pack, which is a registry or manifest of your data?
- Can we get load tests up and running?

As far as sprint, we'll work on how - but this is more on whatsup.

Question: What genomic data is there?
- Not sure. This came from a question from Stanford, and they just said let's try it. So we're trying to make it work. This brings up questions about cluster, too.
- Some people have asked about a way of identifying the most redundant blocks on the network. This is an interesting use case, too.
- @johnnycrunch handles around 55k species genomic data, and has around 20 servers running. Ran into an issue with gc not being turned on by default, so would be curious about loadtests.
- Would be great if @johnnycrunch could join the sprint for ipfs-testlab, helping to set up testlabs so that we can guarantee his data doesn't go down.
- This is one the reasons that IPLD uses CBOR
- There will be an IPLD sprint in a couple of weeks, too, which could be a good place to talk about this.

### js-ipfs on browser

- Creating bookstrapper nodes for js-ipfs nodes;
- Finishing up and implementing relay in JavaScript
- Fixing one issue with our stream-muxer, but making a _new_ muxer. (of course). The issue is on the go-ipfs spdy-muxer.
- Creating example tutorials
- Tests
- Implementing relay in Golang, too.

> The rabbit hole is deep in this one. ~ David Dias

Right now there is a 64k cap in js-ipfs, which doesn't exist in go-ipfs. So, we need to make them both interopable. What we expect to have by the end of this sprint is powerful demos that people can run though.

- We will be checking in daily on the sprint issue: https://github.com/ipfs/pm/issues/310

-----------------

<!-- Add items above this line. Use this format:
- Item (@your_name: @target_audience)
-->

- Demos
- Moderator/notetaker for next time:
* Moderator: @dignifiedquire
* Notetaker: @victorbjelkholm

<!-- After each call, it is the responsibility of the notetaker to save the last version of the notes in a file in ipfs/pm/meeting-notes, by opening a branch and submitting a PR. -->