IPLD spec refining call - round 1

[nicola]

IPLD stakeholders should get together to agree on some pending refinement in IPLD. These issues are not necessarily new features but restrictions that we may have to specify in the spec to make it less ambiguous.

Partecipants are expected to be informed about the different issues, they should:

• Read the different issues beforehand
• Raise concerns and give opinion offline (so that everyone is ready to agree/further discuss)

Agenda so far:

• Canonical representation (Spec refining: Do not bind the canonical representation to CBOR ipld/specs#6)
• Multiple hash pointers in link object (Spec refining: Multiple links in the link object ipld/specs#5)
• Use of IPLD pointers in link object (Spec refining: Support of IPLD pointers as links ipld/specs#3)
• Allow properties in link object (or reserve that space) (Spec refining: Properties in links ipld/specs#2)
• Relative Paths in IPLD (Spec refining: Relative paths in IPLD ipld/specs#1)
• IPLD links must be IPLD pointers (Spec refining: IPLD pointers must point to IPLD objects only ipld/specs#8)
• Requiring prefixing of paths (Spec refining: requiring namespace in IPLD pointers ipld/specs#7)
• Mutable pointers (Spec refining: Mutable links in IPLD ipld/specs#9)

We can do this in 2 weeks time (who should I ping? @RichardLitt )

---

Ping @Stebalien, @mildred, @jbenet, @dignifiedquire

[dignifiedquire]

who should I ping?

For setting up a call or for what?

[nicola]

@dignifiedquire Yes, setting up the call, adding it to the calendar and finding the right timing

[dignifiedquire]

/me summons @RichardLitt

[RichardLitt]

@nicola Best thing to do is to open an issue in github.com/ipfs/pm about a planned call. Should probably happen on Monday near the sprint. Ping anyone who is interested, and they can follow along there.

LMK if you want me to go through the legwork of this for you! :D

[nicola]

(Is this issue that I opened what I need to open?)

[RichardLitt]

Ah! Wow. I thought this issue was somewhere else. What a doofus I am!

Perfect! This is great. I'll add this to the sprint on Monday the 8th. Ping anyone you think should be there and point them here. :D!

[nicola]

Sorry everyone, I missed @RichardLitt post, this will happen the coming Monday the 8th (@RichardLitt)

[nicola]

@dignifiedquire, @diasdavid, @mildred, @Stebalien can we have this conversation on Monday after the sprint?

[daviddias]

Sounds good to me!

[jbenet]

Most important thing to agree on (IMO) is CID, unless that's already agreed
on.

On Sun, Aug 7, 2016 at 09:22 Nicola Greco notifications@github.com wrote:

@dignifiedquire https://github.com/dignifiedquire, @diasdavid
https://github.com/diasdavid, @mildred https://github.com/mildred,
@Stebalien https://github.com/stebalien can we have this conversation
on Monday after the sprint?

—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
#124 (comment), or mute
the thread
https://github.com/notifications/unsubscribe-auth/AAIcoV4z_dZX8FEPw5BfE5_kZ9UkPkJJks5qddwegaJpZM4JVB5G
.

[nicola]

cc @edsilv

[nicola]

@RichardLitt this is happening after the "all hands" meeting, 12pm EDT

[RichardLitt]

@nicola Sounds good. There are a few other ones happening after the all hands, so let's discuss in that call when to schedule this.

[Stebalien]

IRC? (irc://irc.freenode.net#ipfs)?

[Stebalien]

@nicola I assume this can be closed. Etherpad: https://public.etherpad-mozilla.org/p/r.4e9ac4b53b0d5dab99a8a6995909a8e1

[nicola]

Copying the notes over here

IPLD 08 August 2016

Welcome! Note: please add your name to etherpad on the right --------------->

Moderator: David Dias (@diasdavid)

Notetaker: (Nicola Greco @nicola) (with em-ly helping!)(\o/)

Attendees
@em-ly
@jbenet
@nicola
@dignifiedquire
@diasdavid
@EdwardSilverton
@flyzumwalt
@Stebalien

Agenda

agenda is here: #124

0.0. CIDv1 ipfs/specs#130
Canonical representation ( ipld/specs#6
Multiple hash pointers in link object ( ipld/specs#5
Use of IPLD pointers in link object ( ipld/specs#3
Allow properties in link object (or reserve that space) ( ipld/specs#2
Relative Paths in IPLD ( ipld/specs#1
IPLD links must be IPLD pointers ( ipld/specs#8
Requiring prefixing of paths ( ipld/specs#7
Mutable pointers ( ipld/specs#9

Notes

• Calls biweekly, issue will be created with agenda topics that need to be covered
• CID (content ID)

Action items

• CID
  • more time
  • implementation
  • write its own spec
  • discuss where CIDs belong to (multiformats, IPLD, ?)

---

Ref prefixing of paths

1. For the notion of linking between protobuf to protobuf there is an implicit CID of v0, protobuf, base58. To allow for the
   integration of other raw objects, like git objects we need this implicit thing as well. Given that then all links inside networks
   are using implicit CIDs, except for IPLD do we want to do the same thing for IPLD inside links?
2. Do we need to use /ipld/ or for linking to IPLD objects inside IPLD objects?

[nicola]

Will close this once I create all the necessary issues and document what I have done here

[nicola]

I updated all the issues we have discussed and documented the work done in the captain.log

[nicola]

(should have been closed then!)