- tz - lists.iana.org

History of time in the UK
by Joseph S. Myers Feb. 12, 2000

Feb. 12, 2000

I have attempted to turn the historical information on time in the UK, given in the europe file, into a web page, which may now be found at http://student.cusu.cam.ac.uk/~jsm28/british-time/ For the most part the information content is similar to that in the europe file. However, I have attempted to include long titles and date for all Acts and Orders; details of those Channel Islands laws I have are included; this history has been prepared with reference to copies of all 80 laws covered; I have attempted to clearly enumerate the probable gaps; I have tried to indicate exactly what was stated and covered by each law, and what powers were used; those orders that cannot be found simply by reference to the annual published volumes are indicated as such; the occasions of a start on Easter Day have been noted as such. Some commentary from the europe file is not included. Additional to those laws listed in the europe file: S.R.&O. 1921 No. 1276 (defining termination of WW1). More careful details of what was happening under what powers in 1922. tz's behaviour is correct, the comments misleading. Expiring Laws Continuance Act (Northern Ireland), 1923. The main Expiring Laws Continuance Act, 1923 did not cover extending the Summer Time Act, 1922's validity in Northern Ireland. Emergency Powers (Defence) Acts, 1939 and 1940 - more details. Interpretation Act (Northern Ireland), 1954 - account of actual effect from reference to the Act. British Standard Time Act, 1968 - new analysis of the time at which it came into effect (subject to proper verification when the relevant files enter the Public Record Office and are opened under the 30 year rule, in 2002 or so). Details of repeals, and what laws revived on its expiry, with what amendment. Summer Time Act 1972 - this consolidated previous laws without substantive changes. The changes dates already applied on expiry of the 1968 Act. Interpretation Act 1978 - _does_ apply to Northern Ireland. Application of modern orders to Isle of Man and Channel Islands - wording quoted for 1994 starts in 1980. European directives and their implementation - some require different end dates for UK and Ireland, others permit them. The 8th directive is available online. -- Joseph S. Myers jsm28(a)cam.ac.uk

1 0

TZ web page: not effortless
by Oscar van Vlijmen Feb. 12, 2000

Feb. 12, 2000

Following the thought given by Paul Eggert (Feb. 10): "But is it worth our time to do, given that we'd be duplicating other people's efforts?" I would like to add: Looking around, for instance at: http://sandbox.xerox.com/stewart/tzconvert.cgi - update: tzdata1999e http://www.timezoneconverter.com/ - update: Oct. 13 1999 http://www.burbs.com/fcgi-bin/tzconvert.fcgi - update: Nov. 15 1998 http://www.bsdi.com/date - update: periodically; seems reasonably up to date but still not tzdata2000a http://www.hilink.com.au/times/ - update: unknown; seems not very up to date (e.g. Tonga) http://www.timeanddate.com/ - update: tzdata2000a ? it can be noticed that it appears very difficult to maintain an up to date timezone converter. At the moment Steffen Thorsen of timeanddate.com spends between 5 and 10 hours weekly maintaining and evolving his site, which is in existence since 1995 (source: http://www.timeanddate.com/steffen/) Very impressive, but for how long from now? Keep in mind that the Olson effort (tzcode & tzdata) exists since 1986. If someone can guarantee that he or she can find the time, energy and resources to maintain for many years an up to date TZ web application, then go ahead. If possible, as Gwillim Law noted Feb. 8, including all historical data. If it cannot be guaranteed that with each tzdata update the web application gets updated too, it could be wise to forget this idea. Don't even bother to produce long lists of time-related (and calendar-related) web links; there are many more and some very good ones at the moment, and it takes a lot of time to keep those lists up to date. My general idea is: without long-term updating guarantees: just stay with a passive web page, promoting the merits of tzcode & tzdata and the "private investigation" services. For the moment, an extra impact should be gained from submitting tz-link.htm to search engines and keepers of lists with time/calendar info. Oscar van Vlijmen 2000-02-11

3 2

Re: Web page idea
by Paul Eggert Feb. 10, 2000

Feb. 10, 2000

From: "Eric Ulevik" <eau(a)ozemail.com.au> Date: Wed, 9 Feb 2000 00:41:32 +1100 The web page could include direct links to the latest version of tzcode and tzdata, and a web-based interface for looking up the current local time around the world. I know of four sites that do this now. Paul Stewart's "Timezone Converter" <http://sandbox.xerox.com/stewart/tzconvert.cgi> publishes the source code, so it wouldn't be hard to clone, if we could find someone who wanted to host it. But is it worth our time to do, given that we'd be duplicating other people's efforts?

2 1

RE: What's in a name?
by Law, Gwil Jr. Feb. 8, 2000

Feb. 8, 2000

Eric Ulevik [eau(a)ozemail.com.au] wrote: [EU]>From: "Oscar van Vlijmen" <o.van.vlijmen(a)tip.nl> [OvV]>> About the naming of the TZ timezone database. This discussion appears to be searching for a name that will give the general public an idea of what we're doing. If, as I believe, the Olson archive is the most complete and accurate collection of historical time zone data in the world (possibly excepting the Shanks atlases), there are a lot of people who should know about it...not just Unix techies. To the general public, "zoneinfo" is not self-explanatory. There are too many types of zone. I think the name selected should contain either "timezone", "time zone", or "tz". [OvV]>> The main issue was: more visible webpresentation. The host of the "official" site of the tz mailing list should know how to get ranked high by search engines. [EU]>For impact, the main web page must be short and to the point. It should offer a very user-friendly interface to the tzdata archive. At <http://www1.tip.nl/~t876506/DSTtable.html>, Oscar van Vlijmen's new page is a good start. The next step would be to open up the full range of historical data. An advantage to having a highly visible Web site is that we would get more updates and error reports from well-informed laymen. The webmaster would have to spend a good deal of time processing those messages. Who volunteers to be webmaster? [EU]>I'd be happy to do some of the editing if someone can host the site. I infer that you're not willing or able to host it. I can modestly offer some of my space on Mindspring. See my page <http://www.mindspring.com/~gwil/statoids.html> to get an idea of what's there. When I say "modestly", I mean that Paul Eggert's existing site at <http://www.twinsun.com/tz/tz-link.htm> would be a much more logical choice if Paul wants to take on this responsibility. I could set aside about four megabytes for time zone info. As a proof of my bona fides, I've contributed quite a bit of IATA time zone data to Paul Eggert since about 1996. By the way, I've resolved to keep my Web site as noncommercial as possible. As I see it, the tz Web site would be edited in the same way as the tz archive. A moderator (presumably the webmaster) would try to achieve a consensus among the contributors, and would then post the results. Yours, Gwillim Law

1 0

What's in a name?
by Oscar van Vlijmen Feb. 8, 2000

Feb. 8, 2000

About the naming of the TZ timezone database. Let's not forget the main issue. The program and data files themselves should not be renamed. Computer programs could have dependencies by name and computer users are accustomed to certain commands, command-line parameters etcetera. The main issue was: more visible webpresentation. ADO asked Jan. 11 if it wouldn't be useful to merge the tzcode & tzdata. I replied Jan. 11 with amongst other suggestions: "An even more useful suggestion would be to place the WWW.htm file in the directory ftp://elsie.nci.nih.gov/pub/ Why don't you advocate your/our work through a web page? Submit this page to several search engines, I would suggest." The fist step has been made; cf.: http://www.twinsun.com/tz/tz-link.htm Let's think now about even more impact on the web. Finding a catchy or unique name of the stuff is only a small part of this undertaking. Oscar van Vlijmen 2000-02-08

1 0

Wachamacallit again
by Oscar van Vlijmen Feb. 8, 2000

Feb. 8, 2000

In reply to the reply from Paul Hill: > I thought it already had a name. It is named after how it is seen on > a Unix system. As http://www.twinsun.com/tz/tz-link.htm says in the > opening paragraph it is "often called zoneinfo", but that doesn't stop > it from being renamed. AltaVista gives 8246 hits when looking for 'zoneinfo'. It is probably wise to rename the database or the web presentation of it, if you want to get a reasonable high ranking and findability. Oscar van Vlijmen 2000-02-07

3 2

Whachamacallit
by Oscar van Vlijmen Feb. 7, 2000

Feb. 7, 2000

Indeed, Eric Ulevik: a catching name for the TZ database seems useful. AltaVista gives however 29 hits with 'tzdb', half of which are Chinese pages and probably due to a Tzdb site (http://202.96.137.66/dzbk/tzdb/tzdb.htm) AltaVista gives 57 hits with 'Olson database', mostly due to RFC 2445 copies. AltaVista gives 174 hits with 'timezone-database'. AltaVista gives no hits with tz-database. Probably tz-database is a useful name. The tz-link.htm webpage should get at least two META tags, describing the contents in order to get registered correctly with most search engines; like this: <meta name="description" content="....."> <meta name="keywords" content="....."> Regards, Oscar van Vlijmen 2000-02-07

2 1

Reference for tz?
by Joseph S. Myers Feb. 5, 2000

Feb. 5, 2000

The recent appearence of a new way to refer to the tz database - `government database' - caused me to wonder: is there any reference that is appropriate to cite when referring to the tz code and data? That is: since there isn't anything much in the way of a web page for the project, has there ever been any form of publication (academic paper, memo, ...), even a long time ago, describing the project/code/data? -- Joseph S. Myers jsm28(a)cam.ac.uk

2 1

Re: New Mail List - timezones.external@software.com
by Paul Eggert Feb. 4, 2000

Feb. 4, 2000

From: Doug Royer [SMTP:doug@home.royer.com] Sent: Thursday, February 03, 2000 6:47 PM The timezones.external(a)software.com list is for public discussion of timezone information. How does that mailing list's charter differ from that of tz(a)elsie.nci.nih.gov, which is a list for public discussion of time zone information that has been in use since 1986? I visited www.software.com, but found no discussion of the new list. How does one join, or view its message archive? IANA as part of the IETF is looking into administrating the names of world wide time zones. How and why will this administration differ from the process already in place for the existing time zone database? 1) Port the government database and 'zic' (Zone Information Compiler) to other OS's and have it also provide the data in VTIMEZONE format. Most UNIX's use the government database format and the zic compiler (man zic). On 1998-07-16 Antoine Leca <Antoine.Leca(a)renault.fr> wrote that he was implementing (1); you might ask him how far he got. 2) Convert the government database into VTIMEZONE records for IANA to administer. Isn't conversion automatic? If so, why would two databases need to be administered separately? One database should be generated automatically from the other. And if someone want to volunteer - make a converter to go from VTIMEZONE format to zic input format to keep the databases in sync. Why would this be necessary, if conversion is automatic? FYI, here is a brief summary of existing sources for time zone data: http://www.twinsun.com/tz/tz-link.htm And here is a copy of the tz mailing list archive: ftp://elsie.nci.nih.gov/pub/tzarchive.gz

5 5

Re: New Mail List - timezones.external@software.com
by Doug Royer Feb. 4, 2000

Feb. 4, 2000

It is very important that when scheduling across timezones that users across those time zones use the same time zone definitions. FYI: The biggest OS and application vendors on Microsoft and UNIX have applications out -now- that understand VTIMEZONE. This includes Outlook, Lotus Notes, Netscape Calendar, Sun, and the list goes on. Oracle is active as are other database vendors. iCalendar and VTIMEZONE have been in the works for years. There are also many other vendors coming up now on the internet that are supporting this format. It would be a little less critical if the data was a month or two out of date. Each side if using VTIMEZONE, would know when it was going to happen. I could write a batch job that woke up once a week to see if the Olson database had changed, if it had, run the converter to VTIMEZONE and send IANA email. I think that IANA could also do this without my help. -Doug > From: CHancuff(a)aol.com > Date: Fri, 4 Feb 2000 14:18:03 EST > > In a message dated 2/4/00 9:48:42 AM Eastern Standard Time, > Markus.Kuhn(a)cl.cam.ac.uk writes: > > > It is much less likely that the respective countries will report > > timezone changes directly to IANA officially. > > I may be the very least qualified on this mailing list to comment > (since the only other time I've posted concerned my intense > interest in discovering the time zone observed by an island > which spends 6 months/year underwater) but ... > > I can see the advantage of having a web presence to which > reports, rumors, and hearsay can also be reported, then brought > to this list for scrutiny. Perhaps a form page, where such things > can be entered. > > I'm trying to imagine the "signal to noise" ratio of a central > reporting venue like this one, but I can't. It might be worth a try. > > Cliff Hancuff > www.ClearDaySoftware.com -Doug ------------------------------------------------------------------- Doug(a)Royer.COM http://royer.com/People/Doug Text Pager: pager(a)Royer.com 801 W. El Camino #131 Mountain View, CA 94040 Ham Radio: N6AAW, Aviation: PP-ASEL

1 0