Lord Howe Island DST and the Olympics, Macquarie Island DST, etc.

[Paul Eggert]

Oscar van Vlijmen noted some discrepancies between your excellent web
page <http://www.dstc.qut.edu.au/DST/marg/daylight.html> on Australian
daylight saving time and the public domain time zone ("tz") database
maintained by Olson et al (please see
<http://www.twinsun.com/tz/tz-link.htm> for details).  Can you please
let me know any thoughts you might have about these discrepancies?

* Your web page says `Next year, daylight saving will start much
  earlier in NSW and Victoria'.  This is also true of Tasmania and
  ACT; e.g. see <http://www.sacentral.sa.gov.au/information/daylite.htm>.

* Your web page implies (but does not explicitly say) that Lord Howe
  Island will start DST early this year for the Olympics, at the same
  time as the rest of NSW.  Is this true?  If so, the tz database
  needs to be updated.

* Your web page says `Macquarie Island ... changes over to daylight
  saving time when Tasmania does'.  But The Australian Bureau of
  Meteorology FAQ <http://www.bom.gov.au/faq/faqgen.htm> writes
  `Macquarie Island follows Tasmanian practice irrespective of any
  local use of DST.'  (I'm not sure what they mean, but they use
  different wording for Giles and Lord Howe Island.)

* While we're on the subject, that same web page says that Giles
  Meteorological Station uses South Australian time even though it's
  located in Western Australia.

* Your web page gives time zones for Heard and McDonald islands.  I
  assume that this is for when they have visitors?  I'm under the
  impression that they have no permanent human inhabitants.  I bring
  this up because the tz database omits locations that are not
  permanently inhabited, to keep things manageable; but if Heard and
  McDonald are permanently inhabited we need to create an entry for
  that location.