From: William Overington (WOverington@ngo.globalnet.co.uk)
Date: Tue Jun 24 2003 - 06:32:56 EDT
Michael Everson wrote as follows.
> I do the best I can. At the end of the day my document won its case and
the five characters were accepted.
This raises an interesting matter.
In that the document proposes U+2693 for FLEUR-DE-LIS it would seem not
unreasonable for fontmakers now to be able to produce fonts having a
FLEUR-DE-LIS glyph at U+2693.
However, what is the correct approach? Is it that the characters must
remain either unimplemented or else implemented as Private Use Area
characters until Unicode 4.1 or whatever is published, notwithstanding that
the hardcopy Unicode 4.0 book is not yet available? That will probably take
quite some time. It appears to me that there should be some system devised
so that when a few extra symbols are accepted into an already established
area that those characters can be implemented in a proper manner much more
quickly than at present.
However, such speeding up of the process might not always be a benefit. For
example, the proposed U+267F which has, in the document the name HANDICAPPED
SIGN could, if there were a fast track process, be all the more quickly
incorporated into databases as a way for officials to make automated
decisions about people much more conveniently without considering the
individual circumstances of each person so tagged.
I am rather concerned that the name HANDICAPPED SIGN is being used without
any jusitication or discussion of the name of the character. The character
has now been accepted it appears.
I am rather concerned about the Orwellian nightmare possibilities of this
and believe that vigilance is a necessary activity to protect freedom. Just
think, data about someone can be expressed with one character which can be
sent around the world to be stored in a database which is not necessarily in
a jurisdiction which has laws about data protection. Automated decision
making is a matter covered by United Kingdom data protection law, yet does
the law have any effect in practice? For example, some credit card
application documents now have in the small print items about the applicant
agreeing to accept automated decisions. And also, does every user of
computer equipment obey the law?
I gather that in the United States there is a concept of a Social Security
number and that it has now become the widespread practice that people who
are nothing to do with the administration of social security now routinely
ask (and maybe even require) someone to state his or her social security
number before they can do anything. I wonder what is the effect of saying
that the number is for social security purposes and one is not willing to
state what it is. Perhaps even questioning why that information is needed
will go against one.
The issue of the name for what Michael has named as HANDICAPPED SIGN needs,
in my opinion, some discussion. If that discussion widens into what
purposes for which Unicode could or should be used and whether the political
and social implications of encoding symbols is something of which people
should be aware, then fine.
For example, would DISABILITY LOGO be a better name? I have seen the logo
used in signs in shops with the message "Happy to help" referring to help
for people with any disability where help is wanted, not just for people in
wheelchairs. So having the logo in fonts so that such signs could be
printed might well be helpful. Yet I feel that some discussion about the
implications of encoding this logo need to take place, particularly as the
N2586R document suggests as seemingly obvious the potential for use in
databases. For example, could the sign be made as not to be interchanged?
Is it best not to encode it in Unicode at all as being too dangerous in some
of its potential applications? If this symbol is implemented without some
protection for rights, could there be a basis for compensation by someone
disadvantaged by the use of such a symbol in a database?
An Orwellian nightmare scenario of just encoding the symbols and "leaving it
to" people who use Unicode as to how they use the symbols is not attractive.
William Overington
24 June 2003
This archive was generated by hypermail 2.1.5 : Tue Jun 24 2003 - 07:14:32 EDT