[JDEV] Message timestamps

[Jon A. Cruz]

David Waite wrote:

> ?? how is this too complex? I thought that wass how my machine handles it
> already. If you have a list of all the timezones in the world, you match
> someone's zone up and go. When you set something up like time zones in
> windows, it actually gives you different choices depending on whether or not
> you are in an area that does not pay attention to daylight savings (I
> believe Indiana is one example)
>

At first it might not seem too bad, but then the devil is in the details.

What does a client do when a country on the other side decides to change it's
timezone? When countries split? When one decides it wants to get the Melinium
first and bumps things up half an hour? If you're a resident of one of those
locales you are most likely going to get your computer tweaked to match things.
But if you aren't, you might not have a clue it even happened.

Also, what about those obscure cases where the people putting that huge
database of timezones makes a mistake? I've seen some Apple Newton users
complaining of this exact issue.

In essence, having the client send GTM + offset makes all comunication a
one-to-one issue, whereas sending a timezone and counting on arbitary clients
to be able to handle it brings the problem to be a many-to-many issue. Just
more unneeded complexity.


BTW, I'm pretty sure that your machine doens't handle it that way you mention.
For dates/times on Internet mail, newsgroups, web pages, etc. all the standards
use GTM + an offset. The only converions at the local end is GTM-to-local for
incoming data and local-to-GMT for outgoing data. The only timezone you deal
with is your local timezone. Later versions of Netscape show this on web page
info by displaying last-modified in local time and GMT, but the GMT is the
actual returned value.


--
"My new computer's got the clocks, it rocks
But it was obsolete before I opened the box" - W.A.Y.