Proposed change for new U. S. law; also, what about Indiana?

[John Hawkinson]

Paul Eggert <eggert at CS.UCLA.EDU> wrote on Tue, 16 Aug 2005
at 17:09:17 -0700 in <873bp9e85u.fsf at penguin.cs.ucla.edu>:

> > [ I had originally sent a version of this on Tuesday morning, but it
> > failed to go through because apparently lecserver.nci.nih.gov is still
> > firewalled, and elsie works only because of an MX record. ]

lecserver now has an MX record, so I think this is all solved.
> > Can we wait until then and include that number?
> 
> I wouldn't wait.  We can update the comment later.

Given how long its been, I suppose you're right. I was expecting it
within a day of Arthur's patch, but it's been more than a week now.

> > p.s.: I keep wondering if its a good time to propose changing the
> > default name of US timezones to US/Eastern -style rather than
> > America/New_York, because our timezones are set based on national
> > policy
> 
> Unfortunately that's not quite true.  The daylight-saving rules are
> national policy, but whether to adopt daylight-saving is decided more
> locally.

Sure. But for the most part, when someone knows what time zone they're
in, they know they're in US/Eastern, not "the same time zone as New York."

Where a more specific entry is available, like Monticello, KY, (or
US/East-Indiana) of course users should use that. But it doesn't seem
to speak to the general case of whether US/Eastern should be prefered
to America/New_York for people outside of New York, NY. Like
Monticello, AR; Monticello, FL; Monticello, GA; Monticello, IA;
Monticello, IL; Monticello, IN; Monticello, LA; Monticello, MN;
Monticello, MO; Monticello, MS; Monticello, NY; Monticello, UT; and
Monticello, WI.

It certainly seems to make more sense from a user interface
perspective.

> We will continue to support names like US/Eastern indefinitely, but
> the location-based names are more accurate.

But only if you're in that location. Otherwise they're less accurate.
My proposal isn't based on accuracy though -- I'm more concerned
with usability...

--jhawk