From: Asmus Freytag (email@example.com)
Date: Tue Apr 12 2011 - 18:59:10 CDT
On 4/12/2011 2:29 PM, Leo Broukhis wrote:
> If not for mathematical usage,
> blackletter wouldn't have been disunified from antiqua, and any
> attempts at "if I send you a message with a blackletter 𝔐, I want you
> to see a blackletter 𝔐" would have been met with "use rich text".
> Unless a context exists in which divination tarot cards coexist with
> playing cards and the otherwise equivalent cards have a semantic
> distinction, there is no reason to disunify.
"I don't buy that reasoning."
Actually, I agree with you on the black letter characters. Outside
mathematics you have essentially two situations.
1) You are representing *text* written in a language using the Latin
script. This provides a clear context that settles the semantics of the
characters. In this context, the coded characters unequivocally
represent the letters of the Latin alphabet, and the unification with
other graphical representation of the same letters is proper. That this
is so is reinforced by several facts. First, there is a wide variety of
type styles for Latin letters where the unification is not in question
(from script, to typewriter, sans-serif to antiqua). Second, there are
innumerable instances of the same text existing as both fraktur (what
you call black letter) and antiqua, and these were created before
Unicode was even considered. Third, readers familiar with both
representations would affirm that they represent the same *text*. Ergo,
there exists a consistent, and agreed upon level of semantics that can
be captured in plain text. (A contract rendered in black letter would be
as binding as one rendered in English script).
2) You are representing the exact shape of an element of text. Perhaps
in a discussion of how the shapes of some of the capital letterforms in
a typical blackletter font are harder to recognize that their
equivalents in antiqua. Here you are clearly concerned with
presentation, and, as the particular example shows, you quickly get to
the point where the actual font matters. (For example, the A can be
rendered more like a squished U in some fonts, which would make a
difference if you are discussing legibility). Here, there is no
consistent, agreed upon semantics that you could capture in plain text,
therefore rich text is appropriate.
(Mathematics has agreed upon semantics for black letter forms, that are
independent of an actual font choice, hence the decision to encode these
Now we come to playing cards.
There are two sets of interrelated issues. One is the question of
regional flavors of the suits, and one is the question of whether parts
of the tarot deck should be unified with the standard deck.
I find the former most worrying, because the regional variations aren't
minor, and unlike text in a language, you can't necessarily guess from
the stream of encoded characters what the regional context might be -
and in some cases there's more than one for the same region. Unifying
all of these makes no sense, because there's no established usage
pattern that anybody can cite that determines what the agreed upon level
of semantics is that needs to be expressed. The claim is made that the
semantics should be determined by historical derivation, but there's no
evidence that authors and readers have come to a consensus about that.
Unlike fraktur vs.antiqua you don't have lots of parallel texts where
the only change is that one style of card image is swapped against
another. There's simply no way to test the assumptions behind the
proposed unification. That's why I called this particular unification
"untested", and will continue to warn people about blindly using it as
Before onecan even consider encoding any additional playing cards, the
validity of the current, implied unification needs to be established
with better evidence.
The same thing applies to the tarot cards. Outside the realm of serious
experts, I doubt that anyone experiences their claimed relation to the
standard playing cards. Unless there's detailed evidence to the
contrary, they seem like separate systems of symbols to me, that just
happen to share some common ancestry - in that sense not dissimilar to
The main point to take away here, is that these unifications are
"untested" - and that should put the brake on adding any new characters
for card symbols until it is much better understood how these symbols
are intended by their users.
This is especially so, because there's absolutely no urgency to act at
this particular juncture. Nothing will be lost by waiting and everything
will be gained by allowing the existing card symbols to be used for a
while to gain further experience.
This archive was generated by hypermail 2.1.5 : Tue Apr 12 2011 - 19:02:21 CDT