datahike

https://datahike.io/, Join the conversation at https://discord.com/invite/kEBzMvb, history for this channel is available at https://clojurians.zulipchat.com/#narrow/stream/180378-slack-archive/topic/datahike
whilo 2020-11-20T09:17:10.097Z

@geodrome @ 1. Yes, one could add a transaction log that works across indices, but it is not obvious to me that this would have any benefits. Actually I am wondering right now whether we do not have a much better design than Datomic because of the hitchhiker-tree. (Again, it is also an optimal transaction log data structure if you want to maintain a form of index over it). @ 2. Our plan for the medium term (after finishing our cljs port) is to add multi-writer support such that you can implement apps on the frontend and synchronize P2P, as we did with replikativ (we collaborate with a partner company on this). We plan to do this conflict-free in CRDT fashion. It is also easy already right now to implement a git workflow on Datahike, I am thinking about doing it as soon as I have a bit spare time. Merging such branches will require potentially user or application driven conflict resolutions, which yes, can create races if you try to do this automatically. Strongly consistent systems that allow multiple writer with strong consistency guarantees (i.e. linearizability or serializability) typically make sure that there are no races but writers write to non-conflicting parts of the state space. We have not done work in this direction yet, but it is a potential future avenue, e.g. through partitioning by attributes of the schema.

whilo 2020-11-20T09:18:11.098100Z

Are you interested in trying out a multi-peer arrangement in the pull-setting described above?

2020-11-20T13:41:12.099100Z

@whilo Thanks for clarifying the medium term direction. At present I have no need for multiple peers. But I’m just thinking ahead.

2020-11-20T13:53:09.099900Z

I’m really not sure I want to deal with (strong) eventual consistency. I find it difficult to reason about. Need to do more reading.

2020-11-20T13:59:22.100800Z

getting “404 page not found” for https://replikativ.io/

dgb23 2020-11-20T15:33:50.101100Z

I’m getting an SSL error

dgb23 2020-11-20T15:34:33.101400Z

<https://replikativ.io/>



The certificate is not trusted because it is self-signed.

whilo 2020-11-20T17:20:40.101900Z

@denis.baudinot Thank you. I don't get this error, is it fixed by now?

whilo 2020-11-20T17:21:25.102400Z

@mestre Have you done fixes to the website lately?

dgb23 2020-11-20T18:28:15.103100Z

@whilo I tried it with firefox again (my std browser) and now also in chrome and safari. I still get the SSL error

dgb23 2020-11-20T18:29:45.104100Z

you might have a caching issue here or some local configuration or anything like that. I suggest you try it out with a different browser or environment

whilo 2020-11-20T21:18:29.105600Z

Maybe there is a caching issue on your side. I cannot reproduce it in multiple browsers including in the tor browser, which evicts all caches on every restart. Can you show the certificate signature that you receive?

dgb23 2020-11-22T21:11:53.107Z

Yes it might have been. The issue is now resolved.

dgb23 2020-11-22T21:12:36.107200Z

sorry for the inconvenience