Domain-tailored CRDTs for collaboration without server involvement
Languages and libraries
crdt data structures library
SHOW HN2018
Domain-tailored CRDTs for collaboration without server involvement
Domain-tailored CRDTs for collaboration without server involvement
archagon.net#crdt#collaboration#data sync#offline first#conflict free replicated data types
How HN took it
5 positive, 8 neutral or asking, 0 mixed, 2 critical, out of 15 comments about the project. Thread score +0.20.
Idea: 3 for, 2 againstDocs: 2 for, 0 against
The thread, comment by comment
Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.
positive neutral or mixed critical about another commenter or something else the maker
What HN said
What people liked
Clear, enjoyable walkthrough of the state of CRDTs
From 4 comments
Thank you for posting this -- it was a great walkthrough, and very clear. Cool stuff. :-)
phyzome on HN> I admit that this revelation made some wily political thoughts cross my mind. Could this be the chance to finally break free from the shackles of cloud computing? It always…
anilgulecha on HNNice overview of the state of the art on CDRTs. Like the op, I agreed that we are more likely than not to see large clarity brought to the field thanks to the latest compsci…
xenadu02 on HNWonderful article, appreciate the detail. I've had great success in the past using event sourcing, so anything that is even remotely similar is always a fun read. Thanks!
hesdeadjim on HN
Objections
Conflict resolution cannot be avoided when offline edits conflict; blind merging can silently produce wrong results
From 2 comments
> Most obviously, users should be able to edit their documents immediately, without even touching the network [...] > The user should never be faced with a “pick the correct…
codetrotter on HNI think the author is ok with "one of the edits wins," so long as all users agree on the one that will win (without communicating.) I'm still reading TFA, but this is probably…
repsilat on HN
Requests
Generalize the approach to source code revision control
From 3 comments
Unfortunately I wouldn't expect a lot of substantive technical discussion for a post like this on HN -- this post will expire before anyone has time to read it properly and…
urbit on HNI've had some thoughts along the same lines, but haven't taken the opportunity to flesh them out nearly as far. A git repository is nearly a CRDT that represents a file system…
kd5bjo on HNI suspect you could build such a thing if you could more rigidly define edit operations on source code files. The way we edit code now is roughly "insert/edit/delete string <x>…
blake8086 on HN
Questions
How much metadata overhead do the CRDTs require compared to hash-based approaches
From 1 comment
1-3 is generally true for ORDTs (in archagon's terms). I wonder, how much metadata noms needs to achieve those features? CRDTs brute-force all those distributed/concurrency…
gritzko on HN
How does it compare to Noms and RON
From 2 comments
Hi archagon, Have you checked out Noms (https://github.com/attic-labs/noms)? > But studying Differential Sync further, I realized that a number of details made it a non-starter.…
aboodman on HNHi gritzko! I hadn't seen RON It looks really interesting - I'll definitely take some time to play with it sometime soon. It's so great how much work is happening in this space…
aboodman on HN
Alternatives named
Noms
From 1 comment
Hi archagon, Have you checked out Noms (https://github.com/attic-labs/noms)? > But studying Differential Sync further, I realized that a number of details made it a non-starter.…
aboodman on HN
RON
From 1 comment
1-3 is generally true for ORDTs (in archagon's terms). I wonder, how much metadata noms needs to achieve those features? CRDTs brute-force all those distributed/concurrency…
gritzko on HN
Git
From 1 comment
I've had some thoughts along the same lines, but haven't taken the opportunity to flesh them out nearly as far. A git repository is nearly a CRDT that represents a file system…
kd5bjo on HN