You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The Zulip compose box in all clients has always been a tricky balancing act for the UX, for any situation where the message list the user is looking at doesn't correspond to a single conversation (i.e. any "interleaved" narrow). We want it to be easy to reply in the conversation you want to reply in; but hard to accidentally reply in an unintended conversation out of those you're looking at; and also to be easy to start a new topic.
The current prototype takes an approach that's optimized more for implementation simplicity than for this UX balance. That's fine for the alpha, but we should revisit it for the beta, or certainly before launch. We should have either the behavior that's in zulip-mobile, or something we think will be better.
For topic and DM-conversation views, give a compose box that doesn't offer changing the destination conversation (topic or recipients, respectively).
For other views (e.g. all-messages; stream views; arbitrary narrows), don't show a compose box until the user asks for one, with "reply" or "quote and reply" on a specific message. The compose box will send to that message's conversation, without offering a change of destination, since that could lead to confusion and accidents.
The Zulip compose box in all clients has always been a tricky balancing act for the UX, for any situation where the message list the user is looking at doesn't correspond to a single conversation (i.e. any "interleaved" narrow). We want it to be easy to reply in the conversation you want to reply in; but hard to accidentally reply in an unintended conversation out of those you're looking at; and also to be easy to start a new topic.
The current prototype takes an approach that's optimized more for implementation simplicity than for this UX balance. That's fine for the alpha, but we should revisit it for the beta, or certainly before launch. We should have either the behavior that's in zulip-mobile, or something we think will be better.
@chrisbobbe writes at #147 (comment) :
The text was updated successfully, but these errors were encountered: