[missing-sync-palmos-talk] 90-minute alarms become 1 hour on Palm

[Jim C.]

Brian,

Thanks for the explanation. 

I do think that bumping up how you break down the seconds would
definitely do the trick.  Since the Palm only allows for whole hour
increments (when using the hour setting) the only people affected would
be those that have their alarms set for 1 hour in advance... which
would convert to 60 minutes.  The semantics would change, but at least
their alarms would still sound at the correct time, and those of us
that need 90 minute alarms would remain unaffected.

Naturally, this could be extended to include bumping up the hour
cut-off to 99 hours before converting to days, though I'm not sure how
much demand there is for 36-hour advances on alarms (or any day plus
hour combo, up to 4 days and 3 hours [99 hours]).  But, as with the
minutes, only the semantics would change (2 day alarms becomes 48
hours), and the alarms would still sound at the correct time - which I
have to believe is the more important factor.

Thanks again for your response, and here's hoping that you decide to
make this change.

- Jim

--- "Brian M. Criscuolo" <bcriscuolo at markspace.com> wrote:

> Hi Jim,
> 
> No way to fix it, unless we change how we're doing things. Because  
> the alarm trigger comes out of the sync engine only as a number of  
> seconds (unless you enter a specific date) we have to break it up  
> into something that will be reasonable on the Palm.
> 
> Currenly, if it is less than 3600 seconds (1 hour) we set the Palm  
> unit to minutes. If > 1 hour and < 1 day we set it to be hours and > 
> 
> 1 day we set to days.
> 
> We round down when calculating, so that the alarm doesn't appear  
> after the event.
> 
> I don't know what number of minutes would satisfy everyone in this  
> case. Maybe 99 minutes? That's the maximum I can enter in the Palm  
> field.
> 
> I'll look at it a little more closely.
> 
> b