Yes, Talk is clearly their best effort to date, but it wasn't great as a product. For instance it didn't have group chat that actually worked.
Also, in retrospect, they did the wrong thing by choosing to offer XMPP as a public interoperability interface.
I liked the idea of XMPP when it came out - or rather, the idea of having an IETF standard for chat. But XMPP had unnecessary problems due to poor design choices, probably informed by people who didn't have experience writing tidy and efficient server code.
If you are going to design a network protocol you _have to_ involve people who know how to write server code in at least a handful of languages so that you know how to design stuff in mechanical sympathy with how you typically implement various aspects. Any standard written by people who are non-programmers or mediocre programmers will suffer.
As a result of poor design choices, XMPP clients didn't turn out well. One example being that group chats were clumsy and ugly affairs in every single client. If they even supported group chat, which not all clients did.
So I can understand Google eventually ditching interoperability.
What they should have done is to just design their own protocol and open it, but at the time the choice was made to use XMPP a lot of people (myself included) really wanted it to succeed.
(I actually blame XMPP for successful chat systems ending up being closed proprietary affairs. It shows how dangerous it is when people get infatuated with certain ideas and do not pay attention to what it takes to implement properly. When you have a poor standard that costs a lot to implement but which people kind of want to succeed it doesn't leave a lot of room for other standards to emerge. I know a lot of people will react negatively to this. I'd encourage people to try to implement an XMPP server well. Not just something that kinda works and pisses away a lot of resources: do it well. That's an unreasonable amount of work)
As for Google: I'm not sure what was worse. Chat never being a goal in itself, but rather something tucked onto other services or the fact that to this day Google manages to make any form of chat or VC service confusing.
Whenever I have to use a chat that isn't part of a calendar entry I do the same head-scratching exercise. Where do I go again? Which account will I end up on now (it seems to have a talent for ALWAYS picking the wrong account).
At this point I'd rather not use Google for anything that has to do with communication. But it is hard to avoid. Unfortunately.
Also, in retrospect, they did the wrong thing by choosing to offer XMPP as a public interoperability interface.
I liked the idea of XMPP when it came out - or rather, the idea of having an IETF standard for chat. But XMPP had unnecessary problems due to poor design choices, probably informed by people who didn't have experience writing tidy and efficient server code.
If you are going to design a network protocol you _have to_ involve people who know how to write server code in at least a handful of languages so that you know how to design stuff in mechanical sympathy with how you typically implement various aspects. Any standard written by people who are non-programmers or mediocre programmers will suffer.
As a result of poor design choices, XMPP clients didn't turn out well. One example being that group chats were clumsy and ugly affairs in every single client. If they even supported group chat, which not all clients did.
So I can understand Google eventually ditching interoperability.
What they should have done is to just design their own protocol and open it, but at the time the choice was made to use XMPP a lot of people (myself included) really wanted it to succeed.
(I actually blame XMPP for successful chat systems ending up being closed proprietary affairs. It shows how dangerous it is when people get infatuated with certain ideas and do not pay attention to what it takes to implement properly. When you have a poor standard that costs a lot to implement but which people kind of want to succeed it doesn't leave a lot of room for other standards to emerge. I know a lot of people will react negatively to this. I'd encourage people to try to implement an XMPP server well. Not just something that kinda works and pisses away a lot of resources: do it well. That's an unreasonable amount of work)
As for Google: I'm not sure what was worse. Chat never being a goal in itself, but rather something tucked onto other services or the fact that to this day Google manages to make any form of chat or VC service confusing.
Whenever I have to use a chat that isn't part of a calendar entry I do the same head-scratching exercise. Where do I go again? Which account will I end up on now (it seems to have a talent for ALWAYS picking the wrong account).
At this point I'd rather not use Google for anything that has to do with communication. But it is hard to avoid. Unfortunately.