<http://webexhibits.org/daylightsaving/> discussion and corrections

[Paul Eggert]

I compared the current version of the Webexhibits Daylight Saving Time
pages with the tz database at <ftp://elsie.nci.nih.gov/pub/tz/> and
have the following comments.

<http://webexhibits.org/daylightsaving/g.html> says:

 * Fiji observes DST; but our information is that they stopped in 2000.

 * Mongolia observes DST; but our information is that they no longer do
   so (as of 2002).

 * Tonga observes DST from the first Saturday in October to third Saturday
   in April; but our information is that it is now the first Sunday in
   November to the last Sunday in January.

 * Russia observes the same DST rules as Europe; but Europe switches
   at 01:00 UTC, whereas Russia switches at 02:00 local time.

 * Cuba observes DST starting April 1; but our current information is
   that they start on the first Sunday in April, i.e. that they use
   the North American rules except that they change clocks at midnight
   rather than at 02:00.

 * Brazil observes DST from the first Sunday in October to the last
   Sunday in February; but this year it's the first Sunday in November
   to the third Sunday in February (rules vary quite a bit from year
   to year).  Also, equatorial Brazil does not observe DST.

 * We have lots of different reports about the Falklands, but your
   data differs from all of our reports.  Do you have a citation for
   your info?  (The Falklands appears to be somewhat catch-as-catch-can.)

 * Do you have a reference for your information that Rothera,
   Antarctica, is 3 hours behind UTC all year?

 * Paraguay has changed its rules; it now observes DST from the first
   Sunday in September to the first Sunday in April.  See decree no.
   16350, dated 2002-02-26.

A few other comments:

<http://webexhibits.org/daylightsaving/b.html> says "the Eastern Time
Zone portion of the State of Indiana" does not observe DST.  This is
true for most of that part of Indiana, but as you point out in
<http://webexhibits.org/daylightsaving/f.html>, a few counties in
southeastern Indiana observe EST with DST.  So perhaps you could say
"most of the Eastern Time Zone portion of the State of Indiana"
instead.

<http://webexhibits.org/daylightsaving/c.html> says "Eventually, in
1925, it was enacted that summer time should begin on the day
following the third Saturday in April. The date for closing of summer
time was fixed for the first Saturday in October."  As can be seen in
<http://student.cusu.cam.ac.uk/~jsm28/british-time/>, this is
incorrect for both the starting and the ending days.  Perhaps you
could replace that with "Eventually, in 1925, it was enacted that
summer time should begin on the day following the third Saturday in
April (or one week earlier if that day was Easter Day).  The date for
closing of summer time was fixed for the day after the first Saturday
in October."

<http://webexhibits.org/daylightsaving/d.html> says "Nor was the
suggestion of Charles Dowd of Saratoga Springs, N. Y., in 1870, to
establish standard time meridians taken seriously."  This is a bit
harsh, as Dowd revised his proposal in 1872 and the revised proposal
was adopted virtually unchanged by the railways of the USA and Canada
eleven years later.  Certainly Fleming deserves a lot of the credit,
but it goes too far to say that Dowd was not taken seriously.

<http://webexhibits.org/daylightsaving/e.html> says "This [1916]
action was immediately followed by other countries in Europe,
Netherlands, Belgium, Denmark, Sweden, Norway, Italy, France and
Portugal."  You might want to add Luxembourg and Turkey to that list.
Also Tasmania, Nova Scotia, and Manitoba, though these regions are not
in Europe.  It also says "In 1917, Australia, Newfoundland and Nova
Scotia initiated it."  Hmm, perhaps your records for Nova Scotia
differ from ours?  We have Halifax as not observing DST in 1917, even
though it did observe in 1916.  Other areas that introduced DST in
1917 were Russia, Iceland, Spain and Quebec.  (In this area our data
come from Shanks.)

<http://webexhibits.org/daylightsaving/g.html> says "During the
summer, Russia's clocks are two hours ahead of standard time."  This
is not quite true, at least by the ordinary definition of "standard
time".  For example, Moscow standard time (UTC+3) is about a half-hour
ahead of local mean time (UTC+2:30); this is about the same situation
as Detroit, whose standard time (UTC-5) is also about a half-hour
ahead of local mean time (UTC-5:32).  It is true that Russian standard
time tends to be further in advance of local mean time than is usual
for the rest of the world, but it goes too far to say that Russia is
on DST or double-DST all year.

Thanks for your job in maintaining that set of web pages; they're
quite nice.