Initial Delay Timer

[Brad]

Wilson,

I tried this feature tonight for the first time and find it works for "start" but doesn't work for "refresh".  On refresh, it just resets itself and starts counting again, maybe after it checks the first group.

I tested by setting for one minute and watched it reset itself several times.  When downloading only one group, seemed to work ok.

Brad

[Alex Nechiporenko]

Brad,

During "refresh all" operation initial delay will be applied before each of the groups. So, for example, if you have set Initial delay to 5min, you will have 5min delay before downloading each of the groups.

[Brad]

Hi Alex.

Wasn't expecting it to work that way :-(

This was a feature I suggested some time ago

http://pgoffline.com/forum/index.php?board=6;action=display;threadid=31

I was hoping it would work the same doing refresh as when trying to download a single group.

I'd been downloading a single group last night and had used up my allotment.  So, set the initial delay to 1 hour and requested a refresh.

The program waited an hour, looked at the first group (which had no new messages) and then went back into another hour delay.

I think it would be more useful if it worked once for an initial pause, and then the refresh would work normally thereafter.

Hmm, I just thought of an solution that would correct another quirk on this feature:

The initial delay isn't a feature you normally use, if you set it in the options tab, it remains as set.  You normally would want to use the initial delay once and then would probably want it to be disabled thereafter - until you need it again.  If the initial delay setting reset it self back to it's normal value (0) after it timed out the first time, both problems would be resolved.

Brad

[Wilson Logan]

Brad,

 We'll certainly fix it so that it only works once at the start of a Refresh.

I'm not 100% convinced about having it auto-reset to 0. I'm envisaging receiving email along the lines of:

" I've set my initial delay to 60 minutes and every time I do a Refresh it resets to zero. YOU PROGRAM IS STUPID.

  Yours respectfully,

 Mr. Angry  
(Baltimore).  "


Cheers,

Wilson.

[Brad]

Well, you know about those guys from Baltimore - 'ya just can't please them.  Almost as bad as those Texans:-)

Let me back up and restate my case.  First of all, this feature works great for what it was designed to do, which was to do an initial delay when downloading.  I might have initially requested it, but Monte accurately described it's use here:

http://pgoffline.com/forum/index.php?board=6;action=display;threadid=83

As Monte stated, the initial delay value normally is left at 0, so having this set as a "one shot" timer would work great, but manually resetting it back to 0 is not a problem.

The only real problem I see as currently implemented is how the initial timer reacts when doing a refresh.  Right now, it triggers before each group, which isn't useful.  (At least I can't see how it would be useful.)

If it's possible, I would prefer the initial timer to trigger once, allowing all groups to be collected, and then the "refresh periodically" timer assume control if it is selected.

This is just "fine tuning" of an excellect feature you've given us.  I really appreciate the way you've responded to user suggestions.

Thanks,

Brad (Did I mention I'm in Texas? :-)

[Wilson Logan]

Brad,

>The only real problem I see as currently implemented is how the initial timer reacts when doing a refresh.  

I agree & we're fixing that right now.

>If it's possible, I would prefer the initial timer to trigger once, allowing all groups to be collected, and then the "refresh periodically" timer assume control if it is selected.

Fine tuning indeed. I'll add it to the to-do pile :)

Cheers,

Wilson.

[ginahoy]

Just to broaden support for this "fine tuning" request, I totally agree with Brad. This feature wasn't really usefully as written. Normally, you don't want any initial delay for refresh since you are just getting recent messages.

In fact the only time I can imagine using initial delay is if I'm getting a big block and unexpectedly trip the limit. That way I can reset the program to continue in x mintues and walk away.

Personally, I think the inital timer should always return to 0. I'd be willing to bet that no one who uses this timer will actually _want_ it to remember the previous setting.  But even if you do get a complaint (there are always those guys from Baltimore), you have a defensible position for why it does what it does. After all, programming constantly involves give and take... the trick is to figure out what serves the most likely use scenario and/or is least intrusive.

If you really want to cover your bases, you could put a check box that says "keep this value", but seriously, I'd only consider that if someone actually makes legitimate argument for needing it!!

My 2 cents...
David Butler