At 12:57 PM 2/19/03 +0100, Dominikus Scherkl wrote:
If you have issues with the Unicode BETA charts that you would like to see addressed, please follow the instructions on http://www.unicode.org/versions/beta.html about providing beta feedback. Nobody monitors this list for the purpose of extracting feedback buried in the general discussions.Hello.> But why is the Hot Beverage character listed under the > heading "Weather Symbol" in the Miscellaneous Symbols > code chart ? This is by far not the only place where the category in the character description is simply wrong - or gone wrong by the introduction of new characters which doesn't fit.
In fact, they will be. The names list file that was used for the beta code chart was machine-merged from the Unicode 3.2 nameslist plus the list of proposed new characters. The tool does a good job merging blocks and characters (since they have a code position or range that gives them a fixec location in the list), but category headers and general comments (the ones in italics in the list) don't always get merged correctly, or the fact that the new characters interrupt an existing category is not apparent when we work with the list of proposed characters.Especially in the charts which already were pretty full new characters often have no place under the category they would fit - the charts become more and mor mixed up. (e.g. the new arabic presentation form is no currency symbol) I knew, there is no way to avoid this (nothing worse than an re-ordering can be done to an ongoing standard), but the category-names can (and I think should) be reviced.
The final nameslist will be machine-merged from a different source of data. It will take the character codes, names and decompositions from the Unidata.txt file in the Unicode Character Datatabase, and all the other information from an 'annotation' file, from which the correct annotations will be inserted into the nameslist at the correct place.
In the process of preparing the annotation file we review the information and add subheaders and comments and make other changes. To see the best available state of the nameslist, look at http://www.unicode.org/Public/4.0-Update/NamesList-4.0.0xxxx.txt where 'xxxx' are some letters that change with each beta draft level. [I just looked, the new file is not there yet, but will be in a few days.]
That file is the plain text file that drives the charts generator. If you see headers still missing in that file when it comes out, you might want to send an official beta comment and we'll fix it. [We will not republish the PDF beta charts before they are final, since that's a very time consuming process.]
Asmus Freytag
Technical Vice President
The Unicode Consortium
