Skip to content
This repository was archived by the owner on Feb 8, 2023. It is now read-only.
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.

Commit b435c82

Browse files
RichardLittRichardLitt
committedAug 11, 2016
Added notes from Aug 8
See #139. cc @nicola @flyingzumwalt
1 parent 348b176 commit b435c82

File tree

3 files changed

+250
-0
lines changed

3 files changed

+250
-0
lines changed
 

‎notes/2016-08-08-all-hands.md

+120
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,120 @@
1+
# IPFS All Hands Call 08 August 2016
2+
3+
**Moderator:** Matt Zumwalt (@flyingzumwalt)
4+
5+
**Notetaker:** Richard Littauer (@RichardLitt)
6+
7+
**Attendees**
8+
- @nicola
9+
- @RichardLitt
10+
- @jbenet
11+
- @flyingzumwalt
12+
- @haadcode
13+
- @em-ly
14+
- @dignifiedquire
15+
- @diasdavid
16+
- @betaphors
17+
- @Kubuxu
18+
- @haadcode
19+
- @gavinmcdermott
20+
- @ReidWilliams
21+
22+
## Agenda
23+
1. Roll call by alphabetical order (moderator)
24+
2. New call/sprint format and Community Calendar - [issue 134](https://github.com/ipfs/pm/issues/134) and [this comment](https://github.com/ipfs/pm/issues/133#issuecomment-238265884)
25+
3. Call for additional agenda items (moderator)
26+
4. Checking in on the GitHub issue. (@diasdavid)
27+
5. PM Process Discussions [PR 131](https://github.com/ipfs/pm/pull/131) (@flyingzumwalt)
28+
6. Proposal for a breakout call: mfs and `ipfs files`. See [this issue]
29+
8. Proposal for discussion: pubsub. See [this comment](https://github.com/ipfs/pm/issues/130#issuecomment-237293938). (@diasdavid)
30+
(https://github.com/ipfs/specs/issues/98#issuecomment-238032795). (@diasdavid)
31+
9. Proposal for discussion: IPLD. See [this issue](https://github.com/ipfs/pm/issues/124). (@nicola)
32+
10. Schedule for the day
33+
11. Renaming libp2p-swarm to net-machine. See [this issue](https://github.com/libp2p/js-libp2p-swarm/issues/40#issuecomment-238034905). (@diasdavid)
34+
12. RFI about Q2 Update (jbenet)
35+
36+
37+
1. Moderator/notetaker for next time:
38+
* Moderator: @flyingzumwalt
39+
* Notetaker: @dignifiedquire
40+
41+
--
42+
43+
# Notes
44+
45+
- Raise questions by raising hands.
46+
47+
## New format for quick round
48+
- type of discussion:
49+
- Cross-cutting
50+
- Need discussion
51+
- delegate conversations for later
52+
- moderator
53+
- can rotate
54+
- keeps things on track
55+
- keeping things rolling
56+
- keepin' it smooth
57+
58+
## Agenda items
59+
- write them on github
60+
- announce them at the beginning of the call
61+
- writing them in the PAD
62+
63+
## Checking in for Github
64+
- We should continue to check in on the Sprint issue
65+
- But this isn't happening
66+
- We should also have a section about things which we said we could do, but don't
67+
- @jbenet: I don't know if we're going with Kanban or Agile, or so on, and this will make sense later. Super important to track though.
68+
- @dignifiedquire: Writing the tasks is a waste of time if we have Boards. We need to be able to check things at places in time, though, and take a screenshot (or some other way of capturing time-sensitive information), once a week or so.
69+
- @jbenet: We can discuss this, and what models we want, on GitHub
70+
- TODO: @diasdavid Open a GitHub issue about this topic so we can discuss it async
71+
72+
## PM Process Discussions
73+
74+
- Note for new people: We are working on our process right now; in particular, we're trying to get our model down.
75+
- After Lisbon, the goal is to put together a process for managing our projects in the IPFS community.
76+
- We need to find a general-enough process that we can use, but that people can choose to opt into (or not).
77+
- Some IPFS projects may need this now, however.
78+
- So far, a lot of discussion about how we do PM. First, we've been figuring out the model - as @jbenet says, "figuring our the description for the process so that you can implement it with post-it notes if needed."
79+
- There's a document in a PR (https://github.com/ipfs/pm/pull/131) that is good, and we're working on feedback for that
80+
- We need some tool which can be automated, though: roadmaps, milestones, tags, and so on. Ideally, something does this all for you (CI hook, or something similar).
81+
- @jbenet: What should people do this week while we hash this out?
82+
- TODO: @ipfs/devs: Add your sprint todos this Monday for the following work, using the old method, while we figure out how we want this to look. Explicit decision: Use what we used to do that worked (aka huge markdown lists in the sprint issue).
83+
- This pertains to: @RichardLit, @jbenet, @flyingzumwalt, @haadoce, @dignifiedquire, @diasdavid, @Kubuxu, @whyrusleeping, @nicola
84+
- Anyone want to get involved, but havent yet?
85+
- @RichardLitt: Waiting becuase there are a lot of cooks, would like to get involved when you feel it is a good time.
86+
- @dignifiedquire: I'd also like to know, but when there is a draft.
87+
- @haadcode: Review it this week; next week is tooling.
88+
89+
## MFS
90+
91+
@jbenet: All the bike sheds! Can we do this over GitHub?
92+
@diasdavid: If we commit to that, but over a couple of days, not two months, if possible?
93+
@jbenet will do.
94+
95+
## PubSub
96+
97+
@diasdavid: Breakout Discussion!!
98+
- What do we do with our time, research, prototype, and so on.
99+
100+
When: Immediately after this call
101+
102+
## IPLD
103+
104+
Scheduled.
105+
106+
## Renaming
107+
108+
@diasdavid: Should be part of the libp2p discussion
109+
@jbenet: I can't make that call, and this should be on github.
110+
TODO: Can we prioritize that issue? (Yes)
111+
112+
## RFI about Q2 Update
113+
114+
@jbenet: I'm preparing an update on Q2 for the whole IPFS project. Highlighting a lot of important things, too.
115+
I will post a skeleton or a draft and tag relevant people about this (also needs pictures!).
116+
@RichardLitt: Let me haaaalpp
117+
118+
## Done. Retrospective
119+
120+
Done in under 45 minutes. We'll tighten this up.

‎notes/2016-08-08-ipld.md

+53
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,53 @@
1+
# IPLD 08 August 2016
2+
3+
Welcome! Note: please add your name to etherpad on the right --------------->
4+
5+
**Moderator:** David Dias (@diasdavid)
6+
7+
**Notetaker:** (Nicola Greco @nicola) (with em-ly helping!)(\o/)
8+
9+
**Attendees**
10+
@em-ly
11+
@jbenet
12+
@nicola
13+
@dignifiedquire
14+
@diasdavid
15+
@EdwardSilverton
16+
@flyzumwalt
17+
@stebalien
18+
19+
20+
## Agenda
21+
agenda is here: https://github.com/ipfs/pm/issues/124
22+
23+
0.0. CIDv1 https://github.com/ipfs/specs/issues/130
24+
Canonical representation ( https://github.com/ipld/specs/issues/6
25+
Multiple hash pointers in link object ( https://github.com/ipld/specs/issues/5
26+
Use of IPLD pointers in link object ( https://github.com/ipld/specs/issues/3
27+
Allow properties in link object (or reserve that space) ( https://github.com/ipld/specs/issues/2
28+
Relative Paths in IPLD ( https://github.com/ipld/specs/issues/1
29+
IPLD links must be IPLD pointers ( https://github.com/ipld/specs/issues/8
30+
Requiring prefixing of paths ( https://github.com/ipld/specs/issues/7
31+
Mutable pointers ( https://github.com/ipld/specs/issues/9
32+
33+
## Notes
34+
- Calls biweekly, issue will be created with agenda topics that need to be covered
35+
- CID (content ID)
36+
37+
## Action items
38+
- CID
39+
- [ ] more time
40+
- [ ] implementation
41+
- [ ] write its own spec
42+
- [ ] discuss where CIDs belong to (multiformats, IPLD, ?)
43+
44+
45+
46+
47+
----
48+
Ref prefixing of paths
49+
50+
1. For the notion of linking between protobuf to protobuf there is an implicit CID of v0, protobuf, base58. To allow for the
51+
integration of other raw objects, like git objects we need this implicit thing as well. Given that then all links inside networks
52+
are using implicit CIDs, except for IPLD do we want to do the same thing for IPLD inside links?
53+
2. Do we need to use /ipld/<cid> or <cid> for linking to IPLD objects inside IPLD objects?

‎notes/2016-08-08-pubsub.md

+77
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,77 @@
1+
# IPFS PubSub 08 August 2016
2+
3+
Welcome! Note: please add your name to etherpad on the right --------------->
4+
5+
**Moderator:** David Dias (@diasdavid)
6+
7+
**Notetaker:** (Nicola Greco @nicola) (with em-ly helping!)(\o/)
8+
9+
**Attendees**
10+
- @nicola
11+
- @jbenet
12+
- @flyingzumwalt
13+
- @haadcode
14+
- @em-ly
15+
- @dignifiedquire
16+
- @diasdavid
17+
- @betaphors
18+
- @haadcode
19+
- @gavinmcdermott
20+
- @ReidWilliams
21+
22+
## Agenda
23+
24+
--
25+
26+
# Pubsub notes
27+
28+
## Pre-readings
29+
- PubSub discussions so far: https://github.com/ipfs/notes/issues?q=is%3Aissue+is%3Aopen+label%3Apubsub
30+
- PubSub repo -> https://github.com/libp2p/pubsub
31+
32+
## Background
33+
- How do you get fast and (reliable?) communication amongst peers that subscribe to a particular content?
34+
- Everyone that builds on top of IPFS requires pub/sub
35+
- e.g. social networks (orbit, ethereum community!)
36+
37+
## Requirements for PubSub v1
38+
- Subsecond Latency
39+
- Small networks (thousands of notes)
40+
- No? need for packet switching
41+
- There are a few proposed implementations (links?)
42+
- Authenticated streams (data model)
43+
- Way of subscribing and getting notification
44+
- Survive the massive amount of churn in browsers!
45+
- Minimize flood (receiving the same message multiple times)
46+
- Authentication MAY wait
47+
48+
### TODO
49+
- Design
50+
- [ ] list down a list of requirements
51+
- [ ] do we need authenticated streams?
52+
- [ ] Use cases https://github.com/ipfs/2016-Q3-Workshop/issues/17#issuecomment-235006729
53+
- [ ] Identify simplest possible implementation
54+
- e.g. given an annuouncement channel, find a way to form into clusters
55+
that just need a partial view (needed for Orbit)
56+
- [ ] Iron down authenticated streams
57+
- [ ] Trying to identify the different layers
58+
- Implementation
59+
- [ ] create a shared document and start implementing prototypes
60+
- [ ] Testing: Build very basic implementations and a way to spin up a network with thousand of nodes
61+
- [ ] Write down a BIG ISSUE about IPLD
62+
63+
## Lit-review outcome
64+
- Little progress
65+
- Understand concept of treeforming
66+
- We should focus on topic-based pubsub
67+
- Select small spanning trees with not too much effort
68+
69+
## Questions
70+
- q: What happened with ethereum?
71+
- a: not sure what worked
72+
- a: we may need a post-mortem (there are already papers)
73+
- a: not sure if it supports that big of a network
74+
75+
## Layers
76+
- Defining the **interface** for pubsub
77+
- Defining the **data model** for pub/sub (mikola/nicola) https://github.com/ipfs/notes/issues/154

0 commit comments

Comments
 (0)