Another program using the database

I found a program Wclock described at

   http://www.di-mgt.com.au/wclock/download.html

When installed, the program puts several clock faces on the 
Windows desktop.  Additional clocks can be configured and 
existing clocks can be removed.   Unfortunately, its windows is 
always "on top".   It can be reduced to just one clock being 
displayed (and I have just UTC displayed).   

This program uses the TZ database and should probably have a link 

on the tz-link.htm page.

The program stores a Windows-format .INI file from which Posix TZ 

strings can be derived from Olson timezone names.  The program 
apparently will automatically update its .INI file when 
necessary.

Dave Cantor
Groton, CT

In which of four zones should lastThu 24:00 be used

Responses on the mailing list indicate that use of 24:00 in timezone
source files is the way to go where appropriate. The question now is:
where is where?

There are four places currently using a "lastFri 0:00" rule for the
start of DST now and into the future:

	Script started on Tue May 05 09:43:53 2009
	lecserver$ grep -i max.*lastfri.*0:00.*1:00 *
	africa:Rule	Egypt	1995	max	-	Apr	lastFri
0:00s	1:00	S
	asia:Rule	Jordan	2002	max	-	Mar	lastFri
0:00s	1:00	S
	asia:Rule Palestine	2009	max	-	Mar	lastFri
0:00	1:00	S
	asia:Rule	Syria	2009	max	-	Mar	lastFri
0:00	1:00	S
	lecserver$ exit

	script done on Tue May 05 09:44:14 2009

We've seen a report that Jordan should really be "...Mar lastThu
24:00..." What of the other three areas? (Alas, Egypt is the only one of
the three where a look at old newspapers might help.)

				--ado

Re: License for the tzdata information

Re: In which of four zones should lastThu 24:00 be used

"Olson, Arthur David (NIH/NCI) [E]" <olsona <at> dc37a.nci.nih.gov> writes:

> Responses on the mailing list indicate that use of 24:00 in timezone
> source files is the way to go where appropriate. The question now is:
> where is where?

I would guess that all four locations (Egypt, Jordan, Palestine, and
Syria) are inspired by the same mental ruleset, and if we have good
evidence that one should be lastThu 24:00, then we should do the same
with the other four.

Re: License for the tzdata information

> GFDL is something COMPLETELY different from "public domain".  If one
> is right then the other is wrong.

I fully agree. I don't consider copy-left licenses as 'free' or 'open'.

Best

Marc

Re: License for the tzdata information

    Date:        Tue, 05 May 2009 10:31:52 -0400
    From:        "Paul Koning" <Paul_Koning <at> Dell.com>
    Message-ID:  <m1ab5rk4dj.fsf <at> equallogic.com>

  | Can we get an authoritative answer please?

Personally I'd hate for the tzcode (or data) to have any kind of licence.

The reason is that to have a licence, there has to be a licensor,
someone (some legal person, or entity) who is granting the licence.

Currently (unless either ado, or the NIH, wants to be that) there is
nothing around to grant the licence - who do we really think "owns"
this data - who can even say they're responsible for the compilation
of publicly available data?   This is all a co-operative effort,
and if any licence were to be granted, it would have to be done by
everyone who's ever contributed, individually - and that's just a
nightmare to contemplate - and how in any case would we, or anyone,
ever know it was complete - that we hadn't missed a single contributor?
(And for what it is worth, some of the contributors over time - or at
least one I know of anyway, are no longer alive to grant anything, it
would have to be done by their estate(s).)

People here have said that everything (created anyway) is copyright by
someone - and there has to be a licence for anyone else to use it,
and in a sense that might be right - but all of the legal authorities
come from cases where someone is claiming ownership, and someone else
has been trying to deny that (some results may go either way, but there's
always a claim).

I'd barely hesitate to assume that there's not a single case anywhere
where the issue was that that an author was attempting to deny they
held copyright on some material, and someone else was trying to force
it upon them...   That just doesn't happen - who would be bringing
suit, and for what?

That's what I'd like to see with this (and much other) data that we really
want to be "public domain" - where that concept might, or might not
exist - that is, we all simply refuse to claim ownership.  Not as creator
of the data, or as a compiler of the data set.  But not being the owner of
anything means that we also cannot (rationally) give out a licence for
its use (or not a meaningful one anyway).

I do however understand the needs, particularly of large (rich) corporations
that need to do all they can to protect themselves from accidentally
misusing copyright data, and becoming vulnerable to lawsuits because of that.

A licence might seem to help them - but it doesn't (or shouldn't) really,
as the mere existence of the licence doesn't mean its issuer had the
right to do that - the organisation receiving it really still has to
investigate, find who the real owner(s) are, and that the licence is
genuine - barely any different than investigating to find that there's
no determinable owner.

What we might do instead however, is to make it clear to all contributors
of tz code and data (ie: us) that by contributing, we are agreeing not
to enforce any rights that we may have (whether we do or do not) against
any person for any use of the tz code and data in any wy whatever.

If we were to put a notice to that effect - that all contributors have
so agreed, that should perhaps satisfy those people who need some kind
of reassurance before using the data, without anyone ever needing to
claim that they're the owner of all or any part of the tz collection.

kre

RE: License for the tzdata information

Quote:
> Personally I'd hate for the tzcode (or data) to have any kind of licence.

Is there any reason we can't just state in the tzdata as well as the tzcode
that this data is in the public domain? Whether this is equivalent with a
license of copyright or not is only a concern if there is a scenario we want
to avoid (person X claims copyright of the tz data unrightfully). I don't
see that happen. But I am not an IP expert.
Jesper Nørgaard Welen

RE: License for the tzdata information

>>>>> "Jesper" == Jesper Nørgaard <jnorgard <at> prodigy.net.mx> writes:

 Jesper> Quote:
 >> Personally I'd hate for the tzcode (or data) to have any kind of
 >> licence.

 Jesper> Is there any reason we can't just state in the tzdata as well
 Jesper> as the tzcode that this data is in the public domain? Whether
 Jesper> this is equivalent with a license of copyright or not is only
 Jesper> a concern if there is a scenario we want to avoid (person X
 Jesper> claims copyright of the tz data unrightfully). I don't see
 Jesper> that happen. But I am not an IP expert. 

The problem is that the IP status of this material isn't a matter of
opinion, it's a question of fact.

If it's a work of a U.S. Government employee, then it's in the public
domain.

If it's a "sweat of the brow" compilation of public domain data, then
it also is in the public domain.

Otherwise, it is subject to copyright of the authors.  If so, then
a license can be put in place, or the material can be released into
the public domain, but that would have to be done by agreement of the
authors. 

My personal guess is that the first alternative applies.  But that's a
guess, and it's not my call.  That's why I asked for an authoritative
answer -- which I believe has to come from ADO since he's the lead
contributor.   It may well be that getting this answer requires
navigating through red tape.

Arthur, is it possible to get this?  You already made the statement
"there is no license" which sounds a bit like "it's in the public
domain" but it doesn't actually say that.  If you meant public domain,
could you say so?

      paul

Re: License for the tzdata information

On Wed, May 6, 2009 at 12:28 PM, Paul Koning <Paul_Koning <at> dell.com> wrote:

> Arthur, is it possible to get this?  You already made the statement
> "there is no license" which sounds a bit like "it's in the public
> domain" but it doesn't actually say that.  If you meant public domain,
> could you say so?
>I

I second this thought.  If it's possible, a notice in the data files
similar to that already in the code files,

> /*
> ** This file is in the public domain, so clarified as of
> ** 2006-07-17 by Arthur David Olson.
> */

would clarify the situation, and help head off further questions.

  -- Andy

RE: In which of four zones should lastThu 24:00 be used

Paul Eggert writes:
> I would guess that all four locations (Egypt, Jordan, Palestine, and
> Syria) are inspired by the same mental ruleset, and if we have good
> evidence that one should be lastThu 24:00, then we should do the same 
> with the other four.

I disagree that interpreting mental rulesets is a good way forward. Rather I
would recommend to check each occurence where it actually matters and 
create the rules based on that.

For Egypt the only difference of the two rules for the period 1995-2009
occurs in 1998 and 2009, concerning Friday 1.st. of May 1998 and Friday
1.st. of May 2009. We already saw that in 2009 the change took place on the
24.th. of April so did not follow Last Thursday of April at 24:00. I have no
information from 1998, but I doubt that Last Thursday of April would be
correct in 1998.

Jordan is really the inspiration for these rule changes and here it must be
correct.

In Palestine there are no rules further back than 2008 that are relevant for
this ruleset, and for 2008 and 2009 apparently the rule is Last Thursday of
March at 24:00, although none of them have 1.st. of April being a Friday so
we won't know yet.

For Syria we could put the rule for years 2007 and 2009, but these years
were non-critical since either rule would work. And since we already had an
exception in 2008 (First Friday of April) I think it is hard to argue they
go by a rule or a mindset.

- Jesper Nørgaard Welen