|
| 1 | +--- |
| 2 | +layout: default |
| 3 | +title: 3rd November |
| 4 | +description: ZEPs Meeting Notes for 2022-11-03 |
| 5 | +parent: ZEP meetings |
| 6 | +nav_order: 5 |
| 7 | +--- |
| 8 | + |
| 9 | +# 2022-11-03 |
| 10 | + |
| 11 | +**Attending:** Josh Moore (JM), Jonathan Striebel (JS), Jeremy Maitin-Shepherd (JMS), Sanket Verma (SV) |
| 12 | + |
| 13 | +## TL;DR |
| 14 | + |
| 15 | +Discussions were held on how to move forward with ZEP1 quickly. The summary can be viewed [here](https://github.com/zarr-developers/zarr-specs/pull/149#issuecomment-1302440391). Then the attendees discussed extensions in V3, and JMS is considering trying with non-zero origin. SV joined the meeting after 30 mins. After that, JS mentioned some high-level issues looming around V3 spec. |
| 16 | + |
| 17 | +**Meeting Minutes:** |
| 18 | + |
| 19 | +* JMS: number of PRs that could be merged into the working draft |
| 20 | + - JS: don't want to just close it |
| 21 | + - JM: can we cross link e.g. JMS' PR? Yes. |
| 22 | + - ==> once all cross-linked close PR. |
| 23 | +* JS: when to merge? |
| 24 | + - JM: when it matches the consensus? |
| 25 | + - JS: ok, but don't have merge rights. |
| 26 | + - ==> Let's merge proactively. |
| 27 | +* see: https://github.com/zarr-developers/zarr-specs/pull/149#issuecomment-1302440391 |
| 28 | +* extensions |
| 29 | + - JMS: thinking of trying with non-0-origin |
| 30 | + - JM: think that's a general principle we should try for all issues/PR is "could it be an extension" |
| 31 | + - JMS: thinking of extensions as plugins? Not exactly. |
| 32 | + - JS: how to influence if an implenentation adopts an extension? if there's a concrete implementation / clear interface |
| 33 | + - JMS: agreed and some obvious ones (codecs) but not clear there will be a broader abstraction |
| 34 | + - JS: "index transformer" _perhaps_ |
| 35 | + - or as transformer _if_ multiple of chunking |
| 36 | + - JMS: unfortunate limitation |
| 37 | + - JMS: re: transformers - it doesn't make sense to compose a different storage transform _before_ sharding |
| 38 | + - JS: depends. cache of chunks or shards? also checksum |
| 39 | + - JM: codec is similar |
| 40 | + - JMS: caching enabled in code, but not in zarr metadata |
| 41 | + - JS: that's in spec, yes. "runtime-only" but still before or after |
| 42 | + - JMS: when implementing sharding, would check if it's first |
| 43 | + - want to be able to tell the user "this is the graularity to write" |
| 44 | + - JS: good that it's flexible. like c/f order. |
| 45 | + - JS: mention in implementation "sharding must be first" |
| 46 | + - JMS: composing makes for useful extension point |
| 47 | + - JS: most important point: are we sure enough about the extension points? |
| 48 | +- _Sanket joins_ 🧑🏻💻 |
| 49 | +- Jonathan: high-level issues looming |
| 50 | + - paths/URL discussion (needs an issue) |
| 51 | + - global transformers |
| 52 | + - variable chunk length (possibly origin offset) |
| 53 | + - indexing more abstract |
| 54 | + - upgrade path! (`{“extension”: [“@v2-layout”]}`) |
0 commit comments