Tex Texin wrote as follows.
>William, Peter,
>
>Unlike Peter, I wouldn't object to a new term with a precise meaning, as
>it might be useful, even if it's absence hasn't been a problem to date.
>
>However, I have to agree with Peter the name seems very inappropriate as
>the roots do not easily map onto existing concepts, and there are
>existing roots that would work better. graphimap or glyphimap or
>glyphicode would be more natural.
The purpose of the introduction of the word holomap is that it does not
apply to all glyphs in the font, only where, in an advanced format font, the
providing of a direct code point access to the glyph is optional. For
example where a particular glyph is normally accessed through a mechanism
such as a ZWJ sequence yet the map of the whole result is provided by a
direct code point access route as well. For example if a glyph for U+0065,
a lowercase letter e, is produced in a font and mapped to U+0065, then that
is not a holomapping. However, if a glyph for an st ligature is normally
accessed by s ZWJ t then making the glyph also accessible by U+FB06 is a
holomapping as it was not essential for the glyph showing the st ligature.
A holomapping is thus an optional addition to a glyph.
Thus I suggest that graphimap, glyphimap and glyphicode are not suitable as
the words seem to imply that the process can occur with every glyph which is
stored within a font, which is not the intended meaning. I chose the hol-
root so a to convey the idea that to holomap a glyph is to complete a whole
collection of access possibilities to that glyph.
>I am not proposing these nor do I want to enter a discussion of the best
>term for the concept.
Well, you made a comment so it is only fair that I have the chance to
respond to your comment.
>I just want to highlight that if you are going to
>create a definition of a relationship between existing objects, you
>might choose terms with a basis in existing usage. Were you to define a
>new concept or new object I could understand going outside the domain to
>create a term.
Well, when one generates a glyph one can map it to a code point. So, to
complete a whole collection of access options using a map, surely holomap is
an entirely appropriate coining.
>
>A holomap sounds like what I would draw were I to program the Holodeck
>on the USS Enterprise. It should at least refer to encoding of
>3-Dimensional characters which can be viewed from any direction. (To my
>mind anyway.)
Well, the hol- prefix can be used in words such as holography and hologram,
yet there are also words such as holistic and holograph.
The holodeck feature on the USS Enterprise is very interesting and has, in
fact, led me to some interesting ideas about fonts. My favourite item in
the whole of Star Trek is the episode "Elementary, dear Data", which takes
place to a large extent in the holodeck, where one of the characters says
"Arch". This has led to many interesting thoughts about simulations and how
simulations can be controlled and managed while they are running, for
simulations usually have their parameters set from outside the simulation,
yet here the parameters of a simulation are controlled from within the
simulation. The concept of an arch-like feature is something which I am
thinking about for courtyard codes. In courtyard codes the feature is that
codes in the in-coming character stream from a Unicode file can open a
separate window and place upon it controls such as buttons, scrollbars and
textboxes so that the end user may send information to software routines
which are stored in the font along with the glyph design. This is as a way
to implement the "rotating furniture" glyphs which were suggested and which
I then tried to design a method to implement. Yet the technique would seem
to have many possible uses in education.
>
>But I share Peter's sense that there is no problem to be solved and
>therefore a discussion of the term to improve it, is not warranted or
>useful. If the term were distinguishing one concept from another, or
>having a great deal of precision, or rooted in industry terminology and
>therefore more naturally occuring like Peter's verbified cmap, it might
>be worth pursuing. The term is also not saving much wording relative to
>"creating a font table".
Well, using the word holomap one could say the following.
The font designer produced a font where the st ligature is holomapped to the
U+FB06 code point.
Given the definition of the verb holomap, that sentence conveys the meaning
that the accessing of the glyph is not intended to be primarily by using the
U+FB06 code point. How can the sentence be reworded using the wording
"creating a font table" so as to convey the same meaning?
>
>Simply throwing out terms and definitions, without establishing a need
>and ignoring existing industry terms, seems to be a self-inflating and
>glory-seeking action, rather than a desire to make a helpful
>contribution. I mention this not as an attack, just to make you aware of
>the potential perception.
Well, I found that every time that I wished to express in writing the notion
that which the word holomap describes it took a lot of text to explain what
I was meaning. I was seeking to introduce a word with a precise meaning so
that discussions could take place easily. So there is a need.
I am unaware of what existing industry terms you are suggesting that I have
ignored. I am learning more as I proceed and still have lots to learn, so I
would be interested to know.
Why you consider that this could be seen to be a self-inflating and
glory-seeking action simply amazes me. Coining a new word for an existing
situation simply solves a small problem in expressing concepts clearly. How
can that be self-inflating and what glory is there to be sought for doing
that?
Yet if you feel that some people might interpret my coining the word and
posting it to the list in that manner, well perhaps I shall have to act
bearing in mind that some people might react that way, so thank you for the
tip. Perhaps I can mention the Myers Briggs personality type indicator, for
which there are various references on the web. I only found out about the
Myers Briggs Type Indicator in late 2000, I wish I had known about it years
ago, it is very fascinating and has given me much insight into the way
people think. I shall try to work out in my own mind the way in which the
perception of my actions which you suggest some people might form relates to
personality as expressed using the Myers Briggs Type Indicator.
>
>I am willing to admit, that those of us (Americans?) that verbify, have
>less of a need for new terms not based in existing noun roots. Perhaps
>others that do not verbify, might find a term useful.
Well, glyphs are mapped and hol- is a well established prefix.
>
>tex
>
>Peter_Constable@sil.org wrote:
>>
>> On 08/03/2002 03:30:17 AM "William Overington" wrote:
>>
>> "A new word for the English language"
>>
>> Correction: a new word for William Overington's ideolect.
I had not met the word ideolect before. Upon looking it up I found that I
indeed have one! Wow! I am reminded of the gentleman in a play by the
French playwright Molière who found out that he had been speaking prose all
of his life!
>>
>> We don't need a new word coined for this purpose. Those of us involved
>> in font development and the digital font industry have managed fine
>> without it thus far; I don't think any of us have developed any sudden
>> urge for a new way to express ourselves in technical discussions. And
>> if we did, I doubt "holomap" is what any of us would choose. We'd
>> probably do some verbing (so common among English speakers, it seems)
>> and say "cmap" as in "It is almost never necessary to cmap a ligature
>> glyph."
>>
What about Tamil? In Chapter 9 of the Unicode Specification is the
following.
quote
It is important to emphasize that in a font that is capable of rendering
Tamil, the set of glyphs is greater than the number of Tamil characters.
end quote
If someone wishes to produce a display of Tamil text in a program such as
Microsoft WordPad upon a Windows 98 platform using an advanced format font
which contains all of the necessary ligature characters, would it be
necessary, in order to get the job done, to have a font where the ligature
glyphs are holomapped?
I find it strange that the Unicode Standard does not provide code points for
the ligature characters so that holomappings for those ligature characters
could be provided in the fonts. Perhaps I am missing something here, in
which case I would be pleased to learn, yet it does seem to me at present to
be the case that for some applications holomappings of those ligature glyphs
would be useful.
Certainly, in relation to a ct ligature for English, I recognize that for
most purposes a ct ligature is optional, yet, as I understand it, and do
please correct me if I am wrong in this for I am not congruently certain on
the matter, for Tamil, the ligatures are obligatory for correct rendering of
the text.
Nevertheless, although a ct ligature is optional I still feel that it would
be helpful if the Unicode Consortium could allow some more ligatures into
the U+FB.. block, as a collection of holomappings. Certainly using U+E707
for ct and so on from the golden ligatures collection solves the problem to
some extent, yet I feel that code points for these ligatures in regular
Unicode would be helpful and help to preserve our typographic heritage. I
am aware that my golden ligatures collection is not viewed as a good idea by
everyone, yet at least the golden ligatures collection is Unicode
compatible, as are the other code point allocations for items such as chess
fonts. I have seen chess fonts that use the code points of the basic
English alphabet and Fraktur fonts where some of the code points in the
range from hexadecimal 80 through to FF are used for ligatures. It does
seem to me that the Unicode Consortium could encourage a move to Unicode
compatibility by encoding such items as the various pieces for a chess font
and the various Fraktur ligatures into regular Unicode as holomappings.
Certainly, the theory, for Fraktur ligatures, is that everyone will one day
use advanced font technology fonts in order to display the ligatures. Will
it happen? Who knows? I suggest however that encoding the Fraktur
ligatures into the U+FB.. block will be an important and valuable transition
to getting Fraktur fonts compatible with Unicode. At present, a Fraktur
font which is constructed as an ordinary TrueType font needs to have code
points for the ligatures. So, most of the code points need to be made up.
At least the golden ligatures set is Unicode compatible. I really think
that this whole issue of encoding ligatures needs to be looked at again,
without someone claiming which way any vote will go before members of the
Unicode Consortium even have a chance to decide if they will actually
discuss the matter in Committee and take a vote on the matter. There is a
meeting of the Unicode Technical Committee due to take place later this
month and I wonder if some of the people who are going will please consider
raising this issue so that the matter can be discussed in Committee without
any prejudicial stating in this discussion list by anyone of what the
decision is purportedly going to be prejudicing this matter from being
properly discussed.
William Overington
5 August 2002
This archive was generated by hypermail 2.1.2 : Mon Aug 05 2002 - 00:48:17 EDT