[Kst] [Bug 114364] timezone is interfering with consistent plots

Barth Netterfield netterfield at physics.utoronto.ca
Tue Oct 25 00:36:13 CEST 2005


What format are plank's time variables in, and what timezone do they want to 
display them it?

As to the overall situation:

It seems that we have conflicting desires:
	-the desire to be able to create a plot, and display the time in a fixed 
timezone, regardless of the machine you view it on.
	-the desire to read any plot, and display it in a specified timezone, 
regardless of what timezone the creator of the plot used.

In the former, timezone is a property of the plot.  In the latter, it is a 
setting of the machine.

Both can cause confusion, depending on the situation.  I think that the former 
has more opportunity for confusion, because different plots may be treated in 
different ways.  This is a case where the more flexability you have, the 
easier it is to be confused with no good way to minimize this confusion.

And as Planck points out, the only option which won't cause confusion is to 
mandate that all timezone settings are in UT with no chance to change things.  
This is OK, unless for some reason a user wants to display their plot in 
EST5EDT.

So maybe as a compromize between flexability and 'lack of confusion' we could:
	-Make timezone a kst setting, not a plot setting
	-Make the default UT, unless someone goes and changes it
	-If a plot with a converted timezone is shown, put a timezone indicator in 
the status line.
	-Make a compile time option to disable timezone setting to anything but UT 
for Plank, if they want it.
This makes a window with two plots in different timezones harder to do, but I 
imagine this will be pretty rare.

cbn



More information about the Kst mailing list