About rules for Brazil

Paulo Alexandre Pinto Pires p at ppires.org
Mon Oct 21 07:47:30 UTC 2002

Hello, folks.

I have recently downloaded tzdata 2002d, which brings new rules for
Brazil again, because the government again changed when summer time

I suggest that rules for Brazil don't ever be considered permanent,
i.e., make no use of "max" for new rules, but instead use "only".
Regardless of a public poll at National Obversatory's (ON is the
official source of time information for the country) web site, to learn
if Brazilians would like to have fixed dates for start and end of summer
time, the dates were changed again because of plain stupidity --
Electoral authorities forgot about summer time at all, and because of
stupid machine programming, start of summer time had to be delayed.  It
is certain that the new rules will not be used in the next year, when
there will be no elections, and this delay (hopefully) will not be

Reading the source file for timezone data for Brazil is a joke.  Things
like a mayor who tries to change time in his city, and authorities
forgetting about summer time in planning important occasions are
everywhere.  Not having fixed rules is a major cause of such a mess.

This year, the decree stating that summer time should begin not on the
second Sunday of October (yet no such decree ever existed before -- each
ear a _new_ decree said "it starts on day X and ends on day Y"), but
later, was published by the end of its first week, and not reflected on
ON's web site, where most of us used to find such rules.  The result is
that many people, again, had their clocks wrong.

I believe that if clocks eventually go wrong, it would be better if they
just remained in the regular time.  The effect of not advancing the
clock causes less impact (and surprises) than having it unexpectly be
one hour ahead.of what users think it should be, because of assuming
that rules for the previous year will be used again (even though it
sounds very reasonable to our technical ears).

So, we now have

     Rule    Brazil  2001    max     -       Feb     Sun>=15
     0:00   0       -
     Rule    Brazil  2002    only    -       Nov      3
     0:00   1:00    S
     Rule    Brazil  2003    max     -       Oct     Sun>=8
     0:00   1:00    S

but this is wrong, if nothing else, at least because of one very simple
reason (already mentioned): "Oct Sun>=8" or "Feb Sun>=15" may look like
a good aproach, and result of careful observation by some of us, but
they are not in any previous decree, and assuming it may fail in the
future again (and it fails in the past too, if we start looking back. as
far as 1998 and before).  Let's not go beyong the facts we know, that is
we know when summer time starts in 2002 and ends in 2003, and nothing

Besides that, let those of us who are Brazilians try to push the
goverment into using permanent criterion, so that mess doesn't happen

(BTW, I didn't have problems this year, because when I noticed that
ON didn't publish anything until last year's start date was to close,
I dug the Net and found this year's decree).


... Qui habet aurem audiat quid Spiritus dicat ecclesiis.

More information about the tz mailing list