<div dir="ltr"><div dir="ltr">Fabrício,</div><div dir="ltr"><br></div><div dir="ltr"><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sun, 31 Jan 2021 at 17:40, Fabrício Cabral <<a href="mailto:fabriciofx@gmail.com">fabriciofx@gmail.com</a>> wrote: </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div>America/Fortaleza is further than America/Recife. So, by logic, Joao Pessoa people choose America/Recife instead of America/Fortaleza.</div></div></blockquote></div><div><br></div><div>For recent and current timestamps, there is no difference between America/Fortaleza and America/Recife. Both of those entries have used UTC-03 year-round since October 2002.</div><div><br></div><div>The only modern timestamps for which these two entries differ is over one week in the October 2000, when Pernambuco appears to have ended Daylight Saving Time after one week while Paraíba and other neighboring states ended it after two weeks. This difference is shown in lines 983–1000 of the 'southamerica' file in version 2021a.<br></div><div><a href="https://github.com/eggert/tz/blob/2021a/southamerica#L983-L1000">https://github.com/eggert/tz/blob/2021a/southamerica#L983-L1000</a><br></div><div><br></div><div><font face="monospace"># Maranhão (MA), Piauí (PI), Ceará (CE), Rio Grande do Norte (RN),<br># Paraíba (PB)<br>Zone America/Fortaleza -2:34:00 - LMT 1914<br> -3:00 Brazil -03/-02 1990 Sep 17<br> -3:00 - -03 1999 Sep 30<br> -3:00 Brazil -03/-02 2000 Oct 22<br> -3:00 - -03 2001 Sep 13<br> -3:00 Brazil -03/-02 2002 Oct 1<br> -3:00 - -03<br>#<br># Pernambuco (PE) (except Atlantic islands)<br>Zone America/Recife -2:19:36 - LMT 1914<br> -3:00 Brazil -03/-02 1990 Sep 17<br> -3:00 - -03 1999 Sep 30<br> -3:00 Brazil -03/-02 2000 Oct 15<br> -3:00 - -03 2001 Sep 13<br> -3:00 Brazil -03/-02 2002 Oct 1<br> -3:00 - -03</font></div><div><br></div><div>This difference of one hour for one week in 2000 is the reason that America/Fortaleza and America/Recife are separate entries in our database, so technically America/Fortaleza is more correct to use for João Pessoa, but there is otherwise no difference since 1914.</div><div><br></div><div>Please see our project's guidelines for creating new timezone identifier entries:</div><div><a href="https://ftp.iana.org/tz/theory.html#naming">https://ftp.iana.org/tz/theory.html#naming</a></div><div>In particular, among our goals are to "[u]niquely identify every timezone where clocks have agreed since 1970", which means that "[i]f all the clocks in a timezone have agreed since 1970, [we] do not bother to include more than one timezone". Notably, a city being large or being a capital of a region or country is not sufficient for us to add a new entry. The United States, for example, is mostly covered by a small handful of entries, and many US states do not have separate entries of their own as they have always matched neighboring states since 1970.</div><div><br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">The time zone changes are the same of America/Fortaleza</blockquote><div><br></div><div>The listing above shows that our records agree that clocks in João Pessoa and the state of Paraíba have matched clocks in Fortaleza, the state of Ceará, and several other neighboring areas since 1914, which is well before our 1970 cutoff. If that information is incorrect, we'd be happy to hear of reliable references to support a proposed change. But otherwise, we're unlikely to add a separate America/Joao_Pessoa entry.</div><div><br clear="all"><div><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature">--<br>Tim Parenti</div></div></div></div><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"></div>
</blockquote></div></div>