On Thursday 10 August 2017 21:34:19 Christian Loosli wrote: > People pointed out, various times by now, that IRC is the > lowest common denominator and that the rest not only can be bridged, but is > bridged.
The requirements list can also be used to check whether the bridges support
all the functionality that we expect.
Case in point:
During Akademy, the community was split between the Akademy Telegram group,
and the Akademy IRC channel. There was a bridge between them. Messages typed
in one sub-community were bridged to, and displayed in, the other sub-
community.
Watching from the IRC side of things, all the Telegram users were represented
by one IRC user (eh .. TelegramBot?), with messages like
<TelegramBot>: [IsmaelOlea] Bus leaves in 5 minutes!
On the IRC side, how does one respond to Ismael? His nick doesn't autocomplete
(in irssi, konversation or quassel) because he's not actually there as a user;
only TelegramBot is. Clumsy notation such as @IsmaelOlea may work, once the
message crosses the bridge to the other side.
So here the bridge works, but it doesn't really satisfy the requirement (which
isn't on the list, because it's not primarily about bridging requirements, as
far as I can see) that bridged users be not-very-distinguishable from native
users.
Case in point:
Every sarcastic polar bear sticker from Telegram becomes a link on the IRC
side; it's not immediately clear whether the link is something important like
a screenshot of the bug under discussion, or an essential expression of the
speakers emotional state, or just a sarcastic polar bear. Even Quassel's link
preview doesn't help much, if it takes time and effort to get to that preview.
So if the IM-requirements list concludes that sarcastic polar bears are really
important, then that can set goalposts: improving the bridge and the IRC
clients so that some form of spb-stickering is immediately visible.
[ade]
signature.asc
Description: This is a digitally signed message part.
