johan winge <johan.wing-@...> wrote:
original article:http://www.egroups.com/group/elfscript/?start=8
> To summarise:
> a : Three dots pointing upwards or  ^
> e : Acute accent  /
> i : One dot  .
> o : Curl up open right  /
> u : Curl up open left  / or  \
> y : Curl down open left  _/
> å : Not sure. Maybe curl down open right  /_  or \_  ?
> ä (æ) : Two over dots  ¨  (or maybe three dots pointing downwards or
v)
> ö (ø) : Not sure. Maybe sort of a tilde '-,  ex Angasule, or some
curl, or
> three dots or...

Interesting... This would work for Danish too...!

> One issue not being adressed is whether to place the tehta above the
> preceding or following tengwa. I have always placed them above the
> preceding one -- that's what I find most logical. Are you accustomed
to
> write in the other way?

I'm a die-hard Quenya-speaker, so I actually have *always* (and
sometimes to my detriment) written the other way round... ;)

Mark A Miles
Dept. of Scandinvian Studies
University of Edinburgh

Michael Everson wrote:
> Tengwar proposals which have been seen by the standards committee:
[...]
> http://www.egt.ie/standards/iso10646/pdf/tengwar.pdf
> The third proposal (follows some corrections by Arden Smith).

Missing the modified <malta> for _mh_ in the King's Letter (_SD_).
There's also an Old English _hw_, which looks like a <thúle> above
and like a <rómen> below, in the same volume of _HoME_.

> http://www.indigo.ie/egt/standards/iso10646/pdf/tengwar-vowels.pdf
> discussion of vowel representation and modes. Inconclusive and under
> revision by me.

A few comments on the margins:

(1) Referring to a the tendency of a language to have vowel-final
or consonant-final words as its morphological structure is at best
controversial.

(2) It is not always true that `<númen> = _nn_ in Sindarin';
in the King's Letter (in _SD_), the most important published
document in Sindarin-punctuated mode, the use of <númen> and
<óre> is very similar (though not identical) to what is seen
in Quenya.  I therefore suggest you replace <óre> with <númen>
throughout and cut out the parenthetic remark.

(3) I don't understand all the examples at the end (`Comparison
of encodings').  _Tolkien_ #3, #4 and #6 seem redundant.
In _Tolkien_ #7 <hyarmen> should be <yanta>; and shouldn't
the long carrier be a short one?

NB extra _s_ in _discusssed_.

> See the Cirth proposal at
> http://www.dkuug.dk/JTC1/SC2/WG2/docs/n1641/n1641.htm

Er, no, that's about the tengwar too.  Wrong document number?

--
<fa-al-_haylu wa-al-laylu wa-al-baydA'u ta`rifunI
  wa-as-sayfu wa-ar-rum.hu wa-al-qir.tAsu wa-al-qalamu>
                        (Abu t-Tayyib Ahmad Ibn Hussayn al-Mutanabbi)
Ivan A Derzhanski                     <http://www.math.bas.bg/~iad/>
H: cplx Iztok bl 91, 1113 Sofia, Bulgaria          <iad@...>
W: Dept for Math Lx, Inst for Maths & CompSci, Bulg Acad of Sciences

Correction of address:

See the Cirth proposal at http://www.dkuug.dk/JTC1/SC2/WG2/docs/n1642/n1642.htm

Also see the revision at http://www.egt.ie/standards/iso10646/pdf/cirth.pdf

Ivan said:

>Missing the modified <malta> for _mh_ in the King's Letter (_SD_).

Looks like an m-s ligature to me.

>There's also an Old English _hw_, which looks like a <thúle> above
>and like a <rómen> below, in the same volume of _HoME_.

Or is this an h-w ligature?

I think the final -s curls have to be dealt with as ligatures using
Zero-Width Joiner. But if you could wait for my new document as I asked it
would help.

>(1) Referring to a the tendency of a language to have vowel-final
>or consonant-final words as its morphological structure is at best
>controversial.

It's pretty much the explanation for choosing one mode over the other, as I
understand.

>(2) It is not always true that `<númen> = _nn_ in Sindarin';
>in the King's Letter (in _SD_), the most important published
>document in Sindarin-punctuated mode, the use of <númen> and
><óre> is very similar (though not identical) to what is seen
>in Quenya.  I therefore suggest you replace <óre> with <númen>
>throughout and cut out the parenthetic remark.

Isn't it better to stick with the canonical description in AppE.

>(3) I don't understand all the examples at the end (`Comparison
>of encodings').  _Tolkien_ #3, #4 and #6 seem redundant.
>In _Tolkien_ #7 <hyarmen> should be <yanta>; and shouldn't
>the long carrier be a short one?

It is a draft and would you wait. Thanks. ;-)

Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

Greetings Michael and all!


> Guidelines on how to proceed with discussion:
>
> 1. Do not post "competing" proposals. I am managing the proposal, and, as a
> member of ISO/IEC JTC1/SC2/WG2 and am deeply involved in the encoding of
> many scripts in the Unicode/10646 it is well to leave management of the
> proposal in my hands. I assure you I will take all comments and criticisms
> and suggestions into account, but a few years ago Måns Björkman (I think it
> was Måns) published a separate table and I _still_ haven't completely
> figured out the delta between his and mine because he moved things around
> and renamed them as well. Hence Rules 2 and 3.

Actually, the "competing" proposal referred to was compiled by Martin T.
Kutschker, though I provided him with a great deal of input. As I have
not studied the principles of Unicode in great detail, I don't feel
qualified to come up with a proposal of my own, and I shall be more than
happy to lend my aid to that of Mr. Everson.


> Personally, although I know that Tengwar is everyone's favourite, I would
> like to hold off on it for a little while (I am preparing a big new summary
> paper on it) and instead begin by looking at what I think is in fact a more
> complete proposal, namely, that for the Cirth. See the Cirth proposal at
> http://www.dkuug.dk/JTC1/SC2/WG2/docs/n1641/n1641.htm

I agree the Cirth proposal seems more complete, and since I doubt I can
contribute much to it, I will let this matter rest until Michael feels
the time is ripe for criticism of the Tengwar proposal.


Yours,
	 Måns


--
Måns Björkman                      "Twas brillig, and the slithy toves
Törnby                                Did gyre and gimble in the wabe:
SE-179 75 Skå                            All mimsy were the borogoves,
Sweden                                    And the mome raths outgrabe."

I propose unifying away the Cirth combining characters E5, E6, and E7
in favor of U+0302, U+0331 or U+0332, and U+0323 respectively.

The Unicode Standard says (version 3.0, p. 179):

# The combining diacritical marks in this block [viz. U+0300 to U+036F]
# are intended for general use with any script. [...] [T]he characters
# in this block may have multiple semantic values.

The choice between U+0331 and U+0332 depends on whether length marks
applying to consecutive runes are merged into a single long line
or not.  If so, U+0332 is appropriate; if not, or if the evidence is
unclear, U+0331 is appropriate.  What if anything is known about this?

--

Schlingt dreifach einen Kreis um dies! || John Cowan <jcowan@...>
Schliesst euer Aug vor heiliger Schau,  || http://www.reutershealth.com
Denn er genoss vom Honig-Tau,           || http://www.ccil.org/~cowan
Und trank die Milch vom Paradies.            -- Coleridge (tr. Politzer)

Also, DORIAN in character names should be DORIATHRIN.

--

Schlingt dreifach einen Kreis um dies! || John Cowan <jcowan@...>
Schliesst euer Aug vor heiliger Schau,  || http://www.reutershealth.com
Denn er genoss vom Honig-Tau,           || http://www.ccil.org/~cowan
Und trank die Milch vom Paradies.            -- Coleridge (tr. Politzer)

Ar 13:47 -0500 2000-03-30, scríobh John Cowan:
>Also, DORIAN in character names should be DORIATHRIN.

I knew that.... :-)

Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

Ar 13:44 -0500 2000-03-30, scríobh John Cowan:
>I propose unifying away the Cirth combining characters E5, E6, and E7
>in favor of U+0302, U+0331 or U+0332, and U+0323 respectively.

http://www.egt.ie/standards/iso10646/pdf/cirth.pdf

Ah. You mean xx6A, xx6B, and xx6C, don't you?

Circumflex, underscore, and dot below?

The circumflex mark is used in Cirth to indicate nasalization.
The underscore/macron below is used in Cirth to indicate length.
The dot below is used in Cirth to indicate numeric value of a rune.

What about the punctuation? Ken Whistler will kill us if we add yet another
MIDDLE DOT. Runic has a single and a multiple dot separator, but Cirth
definitely has one, two, three, and four dots, and I don't really favour
linking Runic and Cirth anyway.

>The choice between U+0331 and U+0332 depends on whether length marks
>applying to consecutive runes are merged into a single long line
>or not.  If so, U+0332 is appropriate; if not, or if the evidence is
>unclear, U+0331 is appropriate.  What if anything is known about this?

I can't remember seeing examples of the length mark. Anyone?

Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

Hei Måns,

>Actually, the "competing" proposal referred to was compiled by Martin T.
>Kutschker, though I provided him with a great deal of input. As I have
>not studied the principles of Unicode in great detail, I don't feel
>qualified to come up with a proposal of my own, and I shall be more than
>happy to lend my aid to that of Mr. Everson.

For the present you can just look at the Tengwar repertoire and note if you
think you find anything missing, but hold off sending it in till you see my
amazing contribution.

Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

Michael Everson wrote:

> Ah. You mean xx6A, xx6B, and xx6C, don't you?

Yes.  I was looking at N1642 instead of your most recent version.

--

Schlingt dreifach einen Kreis um dies! || John Cowan <jcowan@...>
Schliesst euer Aug vor heiliger Schau,  || http://www.reutershealth.com
Denn er genoss vom Honig-Tau,           || http://www.ccil.org/~cowan
Und trank die Milch vom Paradies.            -- Coleridge (tr. Politzer)

OK, folks, the draft is available at
http://www.egt.ie/standards/iso10646/pdf/tengwar.pdf

It is a draft. This will be obvious when you look at it. Comment is invited.

Guidelines on how to proceed with discussion:

1. Do not post "competing" proposals. I am managing the proposal, and, as a
member of ISO/IEC JTC1/SC2/WG2 and am deeply involved in the encoding of
many scripts in the Unicode/10646 it is well to leave management of the
proposal in my hands. I assure you I will take all comments and criticisms
and suggestions into account, but a few years ago Måns Björkman (I think it
was Måns) published a separate table and I _still_ haven't completely
figured out the delta between his and mine because he moved things around
and renamed them as well. Hence Rules 2 and 3.

2. Do not worry about the order of the characters in the table right now.
Ordering is irrelevant to the UCS character set encoding, and we can settle
on the aesthetics of the table layout LATER.

3. Do not worry too much about the character names right now. We do need to
discuss them and settle on good ones, but they're less important than Rule
4.

4. We need to determine whether the character set is complete or not. Are
there characters missing? Are there characters which are duplicates of each
other and should be unified? Are there characters which are unified and
should be differentiated? Do we have citations for EACH character we encode?

Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

On Thu, 13 Apr 2000, Michael Everson wrote:

> OK, folks, the draft is available at
> http://www.egt.ie/standards/iso10646/pdf/tengwar.pdf
>
> It is a draft. This will be obvious when you look at it. Comment is invited.

I'm not convinced about the idea of modes needing particular
font/rendering logic. Whilst I appreciate that this is done in order that
the underlying spelling is the same, this isn't really a useful property
to have : there are only a few Quenya and Sindarin sounds that share the
same character in their normal modes, and they are different languages, so
words and names will (in the general case) be spelt differently anyway...

This would complicate writing a renderer, and make it impossible to
represent different modes in plaintext.

apart from that...

   xx2D (short carrier) is mentioned in Appendix E :

   "When there was no consonant present in the required position, the
    /tehta/ was placed above the 'short carrier', of which a common form
    was like an undotted i."


   xx70 - xx7D come from a manuscript that I think was published in /Vinyar
   Tengwar/. Perhaps someone else knows which issue.

I see no obvious (or subtle) errors.

--
Robert
"Any person who knowingly causes a nuclear weapon test explosion or any other
nuclear explosion is guilty of an offence and liable on conviction on
indictment to imprisonment for life." -- Nuclear Explosions Act 1998.

Ar 20:01 +0100 2000-04-13, scríobh Robert Brady:

>I'm not convinced about the idea of modes needing particular
>font/rendering logic. Whilst I appreciate that this is done in order that
>the underlying spelling is the same, this isn't really a useful property
>to have : there are only a few Quenya and Sindarin sounds that share the
>same character in their normal modes, and they are different languages, so
>words and names will (in the general case) be spelt differently anyway...

You would get Quenya regularly represented: N-E-Ld-E 'nelde',
N-E-L-T-I-Ld-I 'neltildi' and S-I-L-M-A-R-I-L~-I-X-O-N 'SILMARILLION. But
Sindarin would be incorrectly represented: N-L-E-D-E 'neled', NLEThLI
'nelthil', and S-L-I-V-R-A-L~-I-X-I-R-O 'SILMARILLION'.

>This would complicate writing a renderer, and make it impossible to
>represent different modes in plaintext.

Actually not. You just have to have smart fonts to get the ligatures (as
you do for Arabic and Devanagari) and even if you didn't you would still
get a legible text (with the tehtar simply showing as nonspacing:

Quenya   N´LD´ 'nelde', N´LT·Ld· 'neltildi'
Sindarin N´L´D 'neled', N´LTh·L 'nelThil'

Yes, we'll have to have good fonts, but we would need those ANYWAY in order
to get the nonspacing tehtar to combine correctly over their consonants.

And the alternative would be expensive inputting software to meet Sindarin
or English expectations. This is what you'd have to do to meet the user's
"underlying spelling" expectations, which they probably want to use on
inputting because it is lots easier.

>apart from that...
>
>  xx2D (short carrier) is mentioned in Appendix E :
>
>  "When there was no consonant present in the required position, the
>   /tehta/ was placed above the 'short carrier', of which a common form
>   was like an undotted i."

Oops, that's an error in the names list. xx2D isn't the short charrier, it
is halla. The short carrier is xx2E. So I don't have the attestation for
halla yet.

>  xx70 - xx7D come from a manuscript that I think was published in /Vinyar
>  Tengwar/. Perhaps someone else knows which issue.
>
>I see no obvious (or subtle) errors.

Apparently it was Quettar 13 and 14, but I do not have these. (I need them.)

Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

Spelling errors in some of the examples have been corrected at
http://www.egt.ie/standards/iso10646/pdf/tengwar.pdf


Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

On Fri, 14 Apr 2000, Michael Everson wrote:

> Actually not. You just have to have smart fonts to get the ligatures (as
> you do for Arabic and Devanagari) and even if you didn't you would still
> get a legible text (with the tehtar simply showing as nonspacing:
>
> Quenya   N´LD´ 'nelde', N´LT·Ld· 'neltildi'
> Sindarin N´L´D 'neled', N´LTh·L 'nelThil'
>
> Yes, we'll have to have good fonts, but we would need those ANYWAY in order
> to get the nonspacing tehtar to combine correctly over their consonants.

Ok. That convinces me. I'm just concerned about the complexity, as it is
not likely software companies will devote any effort to Tengwar support, so
it should be that adding support is a 'no-brainer'. (Less conventional
operating systems such as Linux will certainly have support for Tengwar).

Have the committes of ISO and Unicode responsible for such matters
approved, in principle, to the encoding of Tengwar (and Cirth) in the UCS?

--
Robert

Ar 18:56 +0100 2000-04-14, scríobh Robert Brady:

>Ok. That convinces me. I'm just concerned about the complexity, as it is
>not likely software companies will devote any effort to Tengwar support, so
>it should be that adding support is a 'no-brainer'. (Less conventional
>operating systems such as Linux will certainly have support for Tengwar).

It is a complex writing system, and I suspect it won't really get encoded
until we have in implementation font. I'm concerned about whether I got the
Old English vowel encoding right.

>Have the committes of ISO and Unicode responsible for such matters
>approved, in principle, to the encoding of Tengwar (and Cirth) in the UCS?

While they hiss and spit at Klingon, the feeling does seem to be that
Tengwar and Cirth are legitimate scripts for encoding. (Not that this was a
reason I proposed Klingon along with Tengwar and Cirth, oh my no....)

Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

Robert Brady wrote:

> Have the committes of ISO and Unicode responsible for such matters
> approved, in principle, to the encoding of Tengwar (and Cirth) in the UCS?

They neither approve nor disapprove until a specific proposal is in hand.
What we are doing is refining the draft of such a proposal.

--

Schlingt dreifach einen Kreis um dies! || John Cowan <jcowan@...>
Schliesst euer Aug vor heiliger Schau,  || http://www.reutershealth.com
Denn er genoss vom Honig-Tau,           || http://www.ccil.org/~cowan
Und trank die Milch vom Paradies.            -- Coleridge (tr. Politzer)

Fellow Elfscribes,

Here are my first comments on Michael Everson's proposal to encode the
Tengwar in Unicode.

1. "compare Quenya [...] nelde ‘three’, [...] neltildi ‘triangle’ with
Sindarin [...] neled ‘three’ and [...] nelthil ‘triangle’ (the use of
[númen] for n in Quenya and [óre] for n in Sindarin (since [númen]
generally = nn in Sindarin) is irrelevant here)."

Though it indeed is irrelevant, I would like to point out that in the
attested Sindarin mode where ómatehtar are used (see the third copy of
"The King's Letter") _númen_ actually represents /n/ and _óre_
represents word-final /r/. The unattested Sindarin mode used by Mr.
Everson is, if nothing else, a slight distraction and should therefore
be corrected.


2. "In Old English, the vowel preceding a consonant is written above it,
and following a consonant is written after it: [...] þære."

This is the case in the first version of "Edwin Lowdham's manuscript".
In the second version, the consonant is pronounced *before* all vowels
attached to it; if there is both a superscripted and a subscripted
ómatehta, they together represent a diphtong (as in _weorulde_).


3. I very much doubt that xx32 TENGWAR LETTER ANNA SINDARINWA,
supposedly attested in the Moria Gate inscription, is a letter distinct
from xx16 TENGWAR LETTER ANNA. This decorative version of _anna_ simply
follows the general style of the characters in the inscription, as a
comparison with the samples of _númen_ and _vala_ should demonstrate.


4. The title page Tengwar inscriptions in _The Silmarillion_ and _The
Lost Road_ were not written by Tolkien, and should thus not be treated
as reliable sources. From this follows that xx34 TENGWAR LETTER REVERSED
PARMA and xx35 TENGWAR LETTER REVERSED FORMEN should be removed (xx4B
TENGWAR SIGN RIGHT CURL BELOW is attested in "Edwin Lowdham's
manuscript").


5. Attestations:

* xx2D TENGWAR LETTER HALLA is described in Tolkien 1965a, in the last
footnote to the section describing the Tengwar.

* xx2F TENGWAR LETTER EXTENDED LONG CARRIER is, as far as I know, not
attested.

* xx33 TENGWAR LETTER OPEN ANNA is attested in Tolkien 1992, third
script sample.

* xx37 TENGWAR LETTER ASCENDING STEMLESS VALA, as well as xx6E TENGWAR
LETTER REVERSED ROOMEN, are attested in Tolkien 1992, second script
sample.

* xx54 TENGWAR SIGN GRAVE was, to my knowledge, never used by Tolkien.

* xx56 TENGWAR SIGN THREE INVERTED DOTS ABOVE is attested in Hammond and
Scull, _J.R.R. Tolkien: Artist and Illustrator_, ill. ## 184, 186.

* xx63 TENGWAR EXCLAMATION MARK and xx64 TENGWAR QUESTION MARK occur in
the "Namárie inscription" in Donald Swann, _The Road Goes Ever On: A
Song Cycle_.

* The distinction between xx66 TENGWAR SECTION MARK and xx68 TENGWAR
LONG SECTION MARK is probably by necessity typographic rather than
graphemic.

* The numeric tengwar xx70--xx7D were published in Doughan and
Bradfield, _Quettar_ Special Publication #1, 1997.


6. Missing characters:

* The numeric _12_ had a variant form resembling a lower-case _9_, i.e.
with a diagonal line descending from the bow of the circle (ibidem).

* The quotation marks and the _&_ sign used in "Edwin Lowdham's
manuscript" may perhaps be excluded because they occur in other writing
systems as well; but then the double dots, the acute accent and the
breve should be excluded on the same grounds.

* Tolkien 1965a mentions a variant form of the a-tehta "like a
circumflex".

* In the first copy of "The King's Letter", a special variant of xx30
TENGWAR LETTER STEMLESS ANNA is used in word-final position.

* I am of the opinion that the glyph used for _mh_ in "The King's
Letter" is a distinct grapheme, and *not* an m-s ligature. I base this
on the following:

1) We have no other samples of the s-curl being used for anything other
than /s/.
2) The curl in this letter is shorter and more "compact" than any s-curl
by Tolkien I've seen. Also, this would be the only sample of an s-curl
occuring somewhere else but word-finally.
3) Adding a curl to a letter seems in Sindarin to have been the
preferred way of "de-voicing" a letter: compare _rómen_=/r/ vs.
_arda_=/rh/; _lambe_=/l/ vs. _alda_=/lh/.


Suilaid,
	 Måns


P.S. I've pointed this out before, and it really is a trivial point, but
it was *not* me who published a "competing" Unicode proposal some years
ago, though I provided the compilator with much input. I would beg Mr.
Everson to consider rephrazing his Guidelines before posting them again.


--
Måns Björkman                      "'Min drotzete signe then helge and!
Törnby                             Nw hauer iak Swerighe i mynne hand.'
SE-179 75 Skå                   Tha swarade en riddere, het herra Knut:
Sweden                           'Jak tror, tik brister alt annat wt'."

Mans Bjorkman wrote:

> 6. Missing characters:
>
> * The numeric _12_ had a variant form resembling a lower-case _9_, i.e.
> with a diagonal line descending from the bow of the circle (ibidem).

Sounds like a glyph-level distinction.  Are there any cases where the
difference actually *makes* a difference to plaintext interpretation?
That is the general criterion for creating a new Unicode character.

> * The quotation marks and the _&_ sign used in "Edwin Lowdham's
> manuscript" may perhaps be excluded because they occur in other writing
> systems as well; but then the double dots, the acute accent and the
> breve should be excluded on the same grounds.

Indeed, I believe that all of these should be unified away in Unicode,
and represented as font differences, just as the Latin round period
is unified with the Armenian square period as U+002E.

> * Tolkien 1965a mentions a variant form of the a-tehta "like a
> circumflex".

Again, a font difference only (representing a difference in handwriting).
Unicode is not supposed to support all possible palaeographic variants.

> * In the first copy of "The King's Letter", a special variant of xx30
> TENGWAR LETTER STEMLESS ANNA is used in word-final position.

More details?

--

Schlingt dreifach einen Kreis um dies! || John Cowan <jcowan@...>
Schliesst euer Aug vor heiliger Schau,  || http://www.reutershealth.com
Denn er genoss vom Honig-Tau,           || http://www.ccil.org/~cowan
Und trank die Milch vom Paradies.            -- Coleridge (tr. Politzer)

Can anyone supply me with a photocopy or PDF of the Quettar publication(s)
which described the digits? Måns says there was a special publication, but
I heard somewhere that there were two other articles. Maybe the special
publication combined the two?

Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

On the Elfscript site I posted links to the discussion documents, for your
collective convenience.

Thanks for Måns and John for their recent comments.

Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

John Cowan wrote:
>
> Mans Bjorkman wrote:
>
> > 6. Missing characters:
> >
> > * The numeric _12_ had a variant form resembling a lower-case _9_, i.e.
> > with a diagonal line descending from the bow of the circle (ibidem).
>
> Sounds like a glyph-level distinction.  Are there any cases where the
> difference actually *makes* a difference to plaintext interpretation?
> That is the general criterion for creating a new Unicode character.

That's the problem... we really don't know. It may have been used in
special contexts, it may have been a distinct grapheme, but we have
exactly one sample of it, and we're not told how it was used.


> > * The quotation marks and the _&_ sign used in "Edwin Lowdham's
> > manuscript" may perhaps be excluded because they occur in other writing
> > systems as well; but then the double dots, the acute accent and the
> > breve should be excluded on the same grounds.
>
> Indeed, I believe that all of these should be unified away in Unicode,
> and represented as font differences, just as the Latin round period
> is unified with the Armenian square period as U+002E.

Which had me thinking: is the Anglo-Saxon _&_ sign codified already?


> > * Tolkien 1965a mentions a variant form of the a-tehta "like a
> > circumflex".
>
> Again, a font difference only (representing a difference in handwriting).
> Unicode is not supposed to support all possible palaeographic variants.

Fair enough.


> > * In the first copy of "The King's Letter", a special variant of xx30
> > TENGWAR LETTER STEMLESS ANNA is used in word-final position.
>
> More details?

Well, the only difference is that a short "hook", similar to a comma, is
attached to the lower end of the bow.


Yrs,
	 Måns


--
Måns Björkman                      "'Min drotzete signe then helge and!
Törnby                             Nw hauer iak Swerighe i mynne hand.'
SE-179 75 Skå                   Tha swarade en riddere, het herra Knut:
Sweden                           'Jak tror, tik brister alt annat wt'."

Michael Everson wrote:
>
> Can anyone supply me with a photocopy or PDF of the Quettar publication(s)
> which described the digits?

There is a PostScript file that summarizes the article at the TolkLang
web page <http://www.dcs.ed.ac.uk/misc/local/TolkLang>.


> Måns says there was a special publication, but
> I heard somewhere that there were two other articles. Maybe the special
> publication combined the two?

Yes, that is correct. I believe they were nos 13 and 14, but I might be
mistaken.


--
Måns Björkman                      "'Min drotzete signe then helge and!
Törnby                             Nw hauer iak Swerighe i mynne hand.'
SE-179 75 Skå                   Tha swarade en riddere, het herra Knut:
Sweden                           'Jak tror, tik brister alt annat wt'."

Ar 00:15 +0200 2000-04-21, scríobh Mans Bjorkman:
>Michael Everson wrote:
>>
>> Can anyone supply me with a photocopy or PDF of the Quettar publication(s)
>> which described the digits?
>
>There is a PostScript file that summarizes the article at the TolkLang
>web page <http://www.dcs.ed.ac.uk/misc/local/TolkLang>.

I've seen this. I do not want to use the summary. I want to use the primary
article.

Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

I'd be grateful for any comment.
http://www.egt.ie/standards/iso10646/pdf/cirth.pdf

Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

>Michael Everson wrote:
>> Can anyone supply me with a photocopy or PDF of the Quettar publication(s)
>> which described the digits?
>Måns Björkman wrote:
>There is a PostScript file that summarizes the article at the TolkLang
>web page <http://www.dcs.ed.ac.uk/misc/local/TolkLang>.

"http://www.geocities.com/fontwizard/tengnum.zip" (170K)
This ZIP file contains TIF images (easily inserted into most
word processing programs) showing:
1) Quettar Special publication No. 1 (Sept. 1987), pages 1 and 4.
2) Beyond Bree, (Dec. 1984), page 1,  containing a reprint from
    Quettar, No. 13 (Feb 1982), containing a note from CJRT dated
    16 May 1981.
I hope that this helps.  If desired, I can convert the TIFs to PDFs
early next week.

My questions are (may have already been asked):
a) Will Tengwar from passages edited and published by CJRT be
    included in Mr Everson proposal?  Such as title page
    inscriptions.
b) Will the Tengwar proposal include the additional characters
    used in Tolkien's Old-English mode? (I didn't see them in
    the character chart.)

I will continue to examine the proposal and hope to supply this
group with additional comments next week.

A warm thank you to everyone contributing to this endeavor.
(Especially to Michael and Måns!)


--
  Daniel S. Smith - Fantasy Fonts for Windows
  email: fontmaster@...
  http://www.geocities.com/TimesSquare/4948/
  <><

I've made what are I think about the last changes I can make to the Cirth
proposal. I've reviewed the sources and added typeset examples of original
texts. Are there any significant texts we've missed?

Please see http://www.egt.ie/standards/iso10646/pdf/cirth.pdf

Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

>Though it indeed is irrelevant, I would like to point out that in the
>attested Sindarin mode where ómatehtar are used (see the third copy of
>"The King's Letter") _númen_ actually represents /n/ and _óre_
>represents word-final /r/. The unattested Sindarin mode used by Mr.
>Everson is, if nothing else, a slight distraction and should therefore
>be corrected.

(Call me Michael.) I am only too delighted to be corrected.

>2. "In Old English, the vowel preceding a consonant is written above it,
>and following a consonant is written after it: [...] œÊre."
>
>This is the case in the first version of "Edwin Lowdham's manuscript".
>In the second version, the consonant is pronounced *before* all vowels
>attached to it; if there is both a superscripted and a subscripted
>ómatehta, they together represent a diphtong (as in _weorulde_).

Hm. "In some styles of writing Old English"? I'll have to look at Lowdham
again.

>3. I very much doubt that xx32 TENGWAR LETTER ANNA SINDARINWA,
>supposedly attested in the Moria Gate inscription, is a letter distinct
>from xx16 TENGWAR LETTER ANNA. This decorative version of _anna_ simply
>follows the general style of the characters in the inscription, as a
>comparison with the samples of _númen_ and _vala_ should demonstrate.

I agree, and thank you for pointing it out. In the Luthien MS there are
also lots of these decorative serifs.

>4. The title page Tengwar inscriptions in _The Silmarillion_ and _The
>Lost Road_ were not written by Tolkien, and should thus not be treated
>as reliable sources. From this follows that xx34 TENGWAR LETTER REVERSED
>PARMA and xx35 TENGWAR LETTER REVERSED FORMEN should be removed (xx4B
>TENGWAR SIGN RIGHT CURL BELOW is attested in "Edwin Lowdham's
>manuscript").

Arden Smith pointed out the Christopher Tolkien MSS. I don't know, what is
consensus? Obviously we don't want to encode just any tengwa any body
invented, but what about these?

Michael Everson  **  Everson Gunn Teoranta  **   http://www.egt.ie
15 Port Chaeimhghein Íochtarach; Baile Átha Cliath 2; Éire/Ireland
Vox +353 1 478 2597 ** Fax +353 1 478 2597 ** Mob +353 86 807 9169
27 Páirc an Fhéithlinn;  Baile an Bhóthair;  Co. Átha Cliath; Éire

Michael Everson wrote:
>
> (Call me Michael.) I am only too delighted to be corrected.

Okay, Michael. :)


> >4. The title page Tengwar inscriptions in _The Silmarillion_ and _The
> >Lost Road_ were not written by Tolkien, and should thus not be treated
> >as reliable sources. From this follows that xx34 TENGWAR LETTER REVERSED
> >PARMA and xx35 TENGWAR LETTER REVERSED FORMEN should be removed (xx4B
> >TENGWAR SIGN RIGHT CURL BELOW is attested in "Edwin Lowdham's
> >manuscript").
>
> Arden Smith pointed out the Christopher Tolkien MSS. I don't know, what is
> consensus? Obviously we don't want to encode just any tengwa any body
> invented, but what about these?

CJRT's Tengwar inscriptions contain many oddities that his father never
used ("stemless anna" for /s/, vowels being placed freely below the
letters, etc.). Either Christopher hasn't studied his father's English
tengwar inscriptions in great detail, or he has chosen to ignore them
and made up his own conventions. If these mirrored tengwar of his are to
be considered at all (and I beleive they shouldn't), it ought to be as
allographs of the letters they most resemble, since they carry the same
sound value.

Of course, I don't know if there is any "consensus" in this area, but I
hope so. :)

	 Måns


--
Måns Björkman                      "'Min drotzete signe then helge and!
Törnby                             Nw hauer iak Swerighe i mynne hand.'
SE-179 75 Skå                   Tha swarade en riddere, het herra Knut:
Sweden                           'Jak tror, tik brister alt annat wt'."

Aiya!

Today I have finished my Russian translation of Helge Fauskanger's article about
Black Speech and have put it on my site. I invite everybodu to look it. It lies
on http://drauger.chat.ru/bs.htm. I will be glad to receive your opinions about
mu translation ahd whole site.

Best regards,

Drauger
(mailto:drauger@...; http://drauger.chat.ru)