[tz] 2018e Morocco

Brian Basarich BBasarich at taxography.com
Tue Oct 23 18:12:17 UTC 2018


Paul,

Honestly, I don't see how it's more complicated to maintain.  We're talking about adding one new rule and capping an existing one.  Which as I point out, is exactly what was done with the 2014-2021 DST start rule for Morocco.

Further, I'd asked if this supersede-overlap occurs elsewhere in the database, to which you said you thought so.  So I went back and examined the entire set of rules, all 1954 of them, and you're right.  It does occur, but only one other time.  It's HK:

Rule	HK	1965	1976	-	Apr	Sun>=16	3:30	1:00	S
Rule	HK	1973	only	-	Dec	30	3:30	1:00	S

And that supersede-overlap could have been avoided as well with an adjustment to the existing 1965-1976 rule and the addition of one more rule.

All the other rules in the database have this nice property - they all apply operationally.  Wouldn't it be nice if they all did?


-----Original Message-----
From: Paul Eggert <eggert at cs.ucla.edu> 
Sent: Tuesday, October 23, 2018 10:47 AM
To: Brian Basarich <BBasarich at taxography.com>
Cc: Kiley Walbom <kiley.walbom at taxography.com>; Time Zone Mailing List <tz at iana.org>
Subject: Re: [tz] 2018e Morocco

On 10/23/18 7:53 AM, Brian Basarich wrote:
> I understand the need for a rule that supports 2038 and beyond, but instead of having overlapping rules, why not create the rules like this:

Because that would be more complicated to maintain.

> I realize in 2038 the DST end date will likely be in September, and the 2038-max rule would have to be adjusted as future DST end date rules are added.  But I don't see how the above is any different than leaving the 2013-max rule.

Exactly. What we have for Morocco has the same effect as the rules you suggested, except it's easier to maintain.


> And aren't you eventually going to cap the 2013 rule end rule at 2035 down the road?

Nobody knows. It's possible Morocco will cancel DST entirely next year 
and that this will all be moot.

> Does this happen elsewhere in the database?  That is, where multiple rules apply to the same year, but one rule operationally supersedes the other?
I think so, yes.



More information about the tz mailing list