7 ms·
> My opinion is that the Matrix protocol is too powerful. It's going to turn out like TheWeb(tm), where there are only one and a half clients to choose from an
by octoberfranklin 4y ago
> My opinion is that the Matrix protocol is too powerful.
It's going to turn out like TheWeb(tm), where there are only one and a half clients to choose from and instead of a protocol with real standards you just get pseudocode documenting whatever the current version of the dominant client does.
IRC is an IETF standard. Matrix has gone the WHATWG route.
- aaaaaaaaata 4y ago> IRC is an IETF standard v1? v2? v3?
- octoberfranklin 4y agoTLS? v1? v1.2? v1.3? What is the point of your comment?
- Arathorn 4y agoThis is not entirely fair - there are loads of very functional independent Matrix client implementations today already (Element, FluffyChat, Nheko, Hydrogen, Fractal-next, Neochat etc), and a growing number of independent server impls too (Synapse, Dendrite, Conduit, Polyjuice…). It’s true to compare Matrix with WHATWG though - we used the model of WHATWG’s pragmatism as a direct inspiration for Matrix as an implementation-first approach, where we’d then write an open governance spec (having proven it worked) as per https://spec.matrix.org https://spec.matrix.org and eventually merge it into an established standards body like W3C or IETF when stable.
- octoberfranklin 4y ago> there are loads of very functional independent Matrix client implementations No, there's only one. Matrix tries to confuse people by calling the important part of the client a "homeserver".
- a1369209993 4y ago> there are loads of very functional independent Matrix client implementations []today[] There used to be loads of very functional independent web browsers too. There still are, if you count "only works with websites that haven't been up^Wworsegraded in the past decade and a half" as "functional".
- zaik 4y agoIf someone cares about internet standards they probably use XMPP instead of Matrix.