[EMAIL PROTECTED] wrote:
Hi folks,

I would like to inform you about two issues that I am going to address this 
evening. If you disagree, please complain loudly. Otherwise, I will continue 
the CT cleanup...

1. Change time

Every change has a time. This time is used as a guidance for the user while editing his 
document. It is completely irrelevant for LaTeX export. Presently, each new change gets time 
0 (zero). However, when you save your document, the time is substituted by the current time 
(i.e. the time of saving). This "semi-session-like" time management is unfortunate 
for several reasons: First, the change times depend on when and how often you save your 
documents. If you edit a file, (auto-)save it, and continue editing, you will definitely get 
two distinct changes. Moreover, the change time is only displayed after file saving. I am 
going to change LyX's behaviour such that change times always reflect the exact time at 
which the change took place. This is in line with OO and MS Word. To avoid that every 
keystroke results in a separate change, I will also introduce a tolerance interval of, say 5 
minutes. That means if there are two adjacent changes that only differ in their change time 
with delta < 5min, they will be merged (and the later change time is preserved).
Do that mean one eventually have to accept 48 changes for
4 hours of continous typing?

How about always merging changes that are next to each other
and happen in the same editing session?  Use the different timestamps
for changes that are not adjacent, they are much more likely not related.
One change is one change even if it took more than 5min to type it in.
At least from a user perspective. :-)

Helge Hafting

Reply via email to