From: EXIS.Diaz Lopez, Rodrigo [mailto:EXIS.rdiaz@iberia.es] Sent: Tuesday, February 01, 2011 12:25 To: tz@elsie.nci.nih.gov Subject: "Etc\GMT" positive and negative
Hi:
I think I found a bug with the "Etc\GMT" zones.
If you use a "positive zone" like Etc\GMT+2 for calculate actual local time in this zone you obtain. local time for Etc\GMT-2 (¡a negative zone!). For instance, its' 5:00 PM (UTC/GMT), so in GMT+2 zone (Greece) the local time must be 7:00 PM. But, if you use TZ Database, you obtain. ¡3:00 PM!
And the opposite for "negative zones".
Is a bit funny. Please, check it.
I found a web page (http://www.gsp.com/support/virtual/admin/unix/tz/gmt/) that use tz database (I think) and shows this error.
Best regards.
P.S.: It is possible that the problem was with the "codes", because in the world there are "GMT+13" and "GMT+14" zones, and these zones do not exist in tz database. However, in tz database exists "GMT-13" and "GMT-14" zones, that not exist in the real world.
Perhaps this is related to the historical "America not wanting to be negative" problem that gives us "EST5" being the same as "-0500" in RFC822 terms? -- -=( Ian Abbott @ MEV Ltd. E-mail: <abbotti@mev.co.uk> )=- -=( Tel: +44 (0)161 477 1898 FAX: +44 (0)161 718 3587 )=-