On 7 Jun 2005, at 01:54, Mark Davis wrote:
We seeded the project with a set of data that hadn't gone through this process yet, so it hadn't gone through the same level of vetting.
So you did take America/New_York and do s/Eastern/British/g;s/E/B/g. :-) Why pick on just the UK? Why not do that for say Europe/Helsinki, inventing Finnish Daylight Time, or indeed all the en translations? As you suggest below (and I've snipped), surely the Finns would fight their way through your bug reporting system to complain.
We have put out 2 releases since this project was adopted by the Unicode consortium
This 1.3 release is the first with the timezone data though. Was there really a need to rush stuff through unvetted to meet some sort of deadline? With your Unicode Consortium hat on I'm sure you wouldn't have accepted this sort of thing for new additions to Unicode. Folk like Michael Everson need to produce fully researched proposals for that.
We are faced with a huge amount of data, and part of our process has been to try to get tools in place to make it easier to people both to submit data and to vet it. To that end, we've been working on a tool (http://www.unicode.org/cgi-bin/cldr-survey/) that lets us get access to data more easily. For example, en_GB is here: http://www.unicode.org/cgi-bin/cldr-survey?_=en_GB
OK, lets look at that en_GB page. It starts by saying: Vetting is now CLOSED for CLDR 1.3. Thank you! That may be fair enough as 1.3 is now released, but it goes on: To submit vetting information with the survey tool, you will need an account. You should get information on your account from your CLDR contact. If you don't have an account, you can view all the data, but cannot submit. You really aren't doing yourself any favours with stuff like this. If you want feedback and bug reports please think of things from the point of view of people who might submit such stuff. They know about a bit of data, that's why they are trying to correct it, but they probably wouldn't know a CLDR contact from Adam. As I see it you have two problems. The first is a data quality problem with bogus stuff like British Daylight Time and British Time. This is a technical problem that can get fixed. With luck you will change your bug reporting and this survey tool to be less intimidating and you will get more feedback. The second problem is this business of your uniqueness requirements resulting in CLDR trying to tell users that they can't use names that they have used since childhood for what seems to them to be a simple concept like time. I don't know how you fix that, but I'm sure that in any argument about this the users will win. Folk in the UK, me included, aren't going to stop using the names GMT or BST any time soon. Maybe if our government legislates for it we might, but attempts to legislate for UTC have already failed so I doubt that any government will lift the lid off that can of worms for a long while. Peter Ilieve peter@aldie.co.uk