On 2012-07-24 19:21, Mark Rejhon wrote:
[About changing XEP-0301 introductory example to transmit fragmented
words versus whole words]
On Tue, Jul 24, 2012 at 1:15 PM, Mark Rejhon <marky...@gmail.com
<mailto:marky...@gmail.com>> wrote:
Right now, votes are roughly evenly split (half a dozen Yea's, and
half a dozen Nay's over the last 2 years)
Note -- this includes offline discussion with companies on three
different continents, as well as members within the Real Time Text
Taskforce, etc.
There should be no need for voting here. It is just about clear logic
and trying to make the specification consistent and understandable.
Both Kevin and me detected the little logic gap in the text of section
4.1 and the example.
The text says:
" The <rtt/> element is transmitted at regular intervals by the sender
client while a message is being composed"
The example shows <rtt/> elements perfectly divided in the words "Hello"
, "my", "Juliet".
The reader just starting to try to grasp the protocol will of course
wonder "Hmmmm, how come that the typer has completed the words on these
regular intervals? Is there some other meaning buried here with even
words that I have not yet understood?" And from that on, the reader will
be less receptive to grasp the details of the protocol but will merely
search for the answers on these questions.
So, I again suggest that you let the first example just show what the
text says by a more random division of the transmitted text.
The word-by-word thoughts can be saved for 6.1.4.
/Gunnar
