Hacker News new | past | comments | ask | show | jobs | submit login

I'm one of the people in that thread and I'm unconvinced by Jason's insinuations regarding the hostility of the New Vector team. From everything I've seen, interoperability and federation is extremely important for New Vector. The fact that Matrix is a rather innovative and semi-research project has to be taken into account.

For instance, here[1] is the github thread on verifying event hashes which he refers to in that HN thread. From what I can see, he notes some of his (insightful!) thoughts on the matter, nothing but a civil discussion ensues, with parts of his suggestions getting implemented and others being worked on.

[1]: https://github.com/matrix-org/synapse/issues/1252




And here is three documents/issues that all related to synapse breaking systems in the federation, despite repeated extensive technical feedback.

And to this day, despite people that keep on reporting about the issue, nothing is done to actually solve the problem.

https://github.com/matrix-org/matrix-doc/issues/1194 https://github.com/matrix-org/synapse/issues/4540 https://github.com/matrix-org/matrix-doc/pull/1915

If interoperability and federation is important, why push changes that break those, and do nothing for >6 months?


fwiw, the linked MSC https://github.com/matrix-org/matrix-doc/pull/1915 solves the problem, is merged, and is implemented in https://github.com/matrix-org/synapse/pull/4982.

Yes, it would have been great to have done this sooner, but unfortunately fixing a bug which only impacts deactivating accounts when using custom identity servers took a while to get to the top of the todo list.




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact

Search: