3 ms·
I swear, HN somehow tracks what I am doing. The last few days I also looked into CRDTs, Automerge, etc, and here we go. Happens so often, it is uncanny. Here i
by auggierose 3y ago
I swear, HN somehow tracks what I am doing. The last few days I also looked into CRDTs, Automerge, etc, and here we go. Happens so often, it is uncanny.
Here is a good overview article, which has pointers to other articles:
https://cacm.acm.org/magazines/2022/11/265835-research-for-practice-convergence/abstract https://cacm.acm.org/magazines/2022/11/265835-research-for-p...
To me it seems that while state-based CRDTs are easy to understand, operation-based CRDTs are actually what is used in practice. Furthermore, it seems to me the difference between operation-based Automerge, and operational transform (OT) is actually not that big.
- kortex 3y agoHave you heard about the Baader-Meinhoff phenomenon before? You will now. https://en.m.wikipedia.org/wiki/Frequency_illusion https://en.m.wikipedia.org/wiki/Frequency_illusion
- codeulike 3y agoI just read about that yesterday
- dangoodmanUT 3y agoI was literally just reading about this >.>
- strbean 3y agoI'd bet there's a dose of this, as well as a bit of the same factors that lead to multiple discover[0]. Seems like conditions may be right for a boom in the application of CRDTs. 0: https://en.wikipedia.org/wiki/Multiple_discovery https://en.wikipedia.org/wiki/Multiple_discovery
- squidsoup 3y agoNah, it's definitely the Noosphere.
- alephnan 3y agoIt’s also possible that the cycles of tech employment means programmers are exploring similar topics over time.
- yawnxyz 3y agoI hadn't thought or heard about CRDTs for many months until stumbling across Automerge yesterday. And here we are!
- rstuart4133 3y agoOpen access version: https://martin.kleppmann.com/papers/convergence-cacm.pdf https://martin.kleppmann.com/papers/convergence-cacm.pdf