Implemented for Hermes 2.0

See also Request: 12543

In our usage, it's a must, as that information is required.   
Generally, though, can seems more appropriate.  The salient point is, 
I think, that once you've committed to that information (regardless of 
whether there is any info or not), you cannot change it later.
Client, job type, etc.
No.
Yes, only the "active" timer is running.  All others are paused.  You 
may switch to another and have it activate while the others pause.
No.
I find the stop watch to have zero utility, so I would consider it a 
replacement.  However, the general utility of the stopwatch may be 
needed, so an addition would likely be best.

Ping?

I would change "must" to "can".
I don't exactly follow the interface flow here. And what do you mean 
with "previously entered information"? Already running timers?

What happens with running timers anyway? Are they paused while another 
timer is running? Is the last active timer resumed automatically once 
a later timer has been stopped?
I'm not sure, do you consider this as a replacement for the stopwatch, 
or next to it? Anyway, it sounds like a nice enhancement.

See also Ticket 2667 and Ticket 6549.



Our particular usage, as developers, is to accumulate time on given 
tickets.  For example, work on Ticket #1, finish it, turn it over to 
test, move on to Ticket #2.  Later, test comes back and complains that 
Ticket #1 isn't finished, so we "switch" back to Ticket #1 and repeat. 
  In a given day, I might have 6-12 tickets under way in this 
waterfall type process.



Another time this arises in when we double duty "on the desk," taking 
support calls.  We'll be working away on Ticket N, when the phone 
rings and we go into support mode.  So, we "switch" our task and want 
to accumulate time on the right one.



Today, one way we handle this is to track our hours in a spreadsheet 
at our desks: at the end of the day, we sprinkle the worked hours 
around roughly on all the tasks we worked.



Timers are more accurate, and it seems like timers would be more 
generally useful if the following modifications were made:

1. When you create a timer, you must fill in the time entry's "static" 
information: client, job type, cost object as well as a "name" for the 
timer.

2. When you click the timer, you see (but cannot edit) the previously 
entered information.  You also see (but cannot edit) the date and 
accumulated hours.  You are prompted to enter a description and 
optionally notes.  You have two options: Save and continue timing, or 
Save and stop timing.



The idea here is to focus the information you enter at each stage of 
the timer -- creation and click -- to just the needed information, and 
to give you the option to continuing timing.



Thoughts?