[Standards-JIG] JEP-0085: suggested triggers

Peter Saint-Andre stpeter at jabber.org
Mon Sep 12 14:30:53 CDT 2005


Ian Paterson wrote:
> To summarise, the question is: should 'paused' be sent if the user
> "starts interacting with another chat interface or application", or
> should the client wait for 5 seconds of inactivity.
> 
> 
>>Ian Paterson maintains that it is important for the user to know
>>if another chat participant has switched focus, even for less 
>>than the suggested paused timeout. This does not seem all 
>>that useful to me (again, it's prospective notification)
> 
> 
> If I think someone is composing a reply to me, then (if I'm not too
> time-stressed), I wait instead of switching immediately to another
> window to continue with whatever I was doing. If while composing the
> other person is interrupted (by another chat window or a VoIP call or
> whatever), I don't want to hang around waiting for the 5-second timeout
> before I'm informed the person has paused (for an indefinite period). I
> want to know now immediately, so *I* can decide (depending on how busy I
> am) either to return to what I am doing until they finish composing the
> message or to wait a little longer (they are not yet 'inactive' after
> all, and if they start composing again I will know immediately, and if
> they don't start composing again I know that immediately too).
> 
> Power (information) to the users.

But power tends to corrupt. :-)

Here is the scenario that seems odd to me:

1. You and I are chatting

2. You ask me a question about a web page that we're both talking about 
(I have both the chat window and the browser window open).

3. I start typing a reply but while I'm typing I change application 
focus from our chat window to the browser window for a split-second, 
then finish my reply and hit Return. (For example, I may do this because 
the part of the web page you've asked me about is obscured by the chat 
window.)

Compare this with...

3a. I start typing a reply but while I'm typing I change eye focus from 
our chat window to the browser window for a split-second, then finish my 
reply and hit Return. (I don't need to change application focus because 
the part of the web page you've asked me about is not obscured by the 
chat window.)

3b. I start typing a reply but while I'm typing I space out for a 
split-second while I mentally recall the browser window from eidetic 
memory, then finish my reply and hit Return.

I fail to see any material difference between 3, 3a, and 3b from your 
perspective.

> The paused state is good, it gives the other person *immediate*
> information without any negative emotional connotations, and it is
> immediately corrected if the user starts composing again.

Define "immediate". :-)

> Finally, the plain truth is that: *as soon as* I "start interacting with
> another chat interface or application" I am paused. I'm not typing in
> the input field anymore. True the pause might only last 2 seconds, but
> the other person will be informed about the end of the pause immediately
> too.
> 
> FYI, in practice most pauses only last about 2 seconds, however they
> were triggered. But the instant feedback associated with toggling
> between 'paused' and 'composing' helps to make IM chat a more dynamic
> communication experience.

If the suggested paused trigger is set to 2 seconds (not 5 seconds) then 
  are you happy?

What I'm saying is that I don't think the proximate cause for the pause 
matters all that much (application focus, eye focus, spacing out, etc.). 
All that matters is the fact that the chat participant has indeed paused 
for 2 or 5 seconds or whatever.

Peter

-- 
Peter Saint-Andre
Jabber Software Foundation
http://www.jabber.org/people/stpeter.shtml
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3511 bytes
Desc: S/MIME Cryptographic Signature
Url : http://mail.jabber.org/pipermail/standards/attachments/20050912/58d86850/smime.bin


More information about the Standards-JIG mailing list