[9fans] yet another font package converted to plan9

[9fans] yet another font package converted to plan9

[Joel Salomon]

regarding http://pages.cpsc.ucalgary.ca/~mirtchov/p9/fonts/ esp. the
sample given for 9x18ja
(http://pages.cpsc.ucalgary.ca/~mirtchov/p9/fonts/9x18ja.png) - where did
the text come from? The question mark at the beginning of the line in
Hebrew seems to imply that right-to-left text has been implemented in
plan9 - which I know it hasn't. Besides the text *is* backwards - those
words that are intelligible, that is. The sentence itself is gobbledygook
(at best :-)

Are the phrases in the other languages equally meaningless?

--Joel

Re: [9fans] yet another font package converted to plan9

[andrey mirtchovski]

On Mon, 4 Aug 2003, Joel Salomon wrote:

> regarding http://pages.cpsc.ucalgary.ca/~mirtchov/p9/fonts/ esp. the
> sample given for 9x18ja
> (http://pages.cpsc.ucalgary.ca/~mirtchov/p9/fonts/9x18ja.png) - where did
> the text come from? The question mark at the beginning of the line in

the examples are linked to from Markus Kuhn's web page:

	http://www.cl.cam.ac.uk/~mgk25/ucs/examples/

> Hebrew seems to imply that right-to-left text has been implemented in
> plan9 - which I know it hasn't. Besides the text *is* backwards - those
> words that are intelligible, that is. The sentence itself is gobbledygook
> (at best :-)

there's no right-to-left input in plan9, yes. using this was simpler than
trying to fit the output of 'unocode 0x0-0xffff' in a single screen :)

> Are the phrases in the other languages equally meaningless?
>

the slavic ones are :)

andrey

[9fans] tfel-ot-thgir support

[Skip Tavakkolian]

> The question mark at the beginning of the line in
> Hebrew seems to imply that right-to-left text has been implemented in
> plan9 - which I know it hasn't

I plan to work on this, but I have to get a lot smarter on the
subject.  Also, it is low on my priority list, unfortunately.

Has anyone done any critical thinking on this?

Re: [9fans] tfel-ot-thgir support

[boyd, rounin]

> Has anyone done any critical thinking on this?

looks nasty.  you could try asking:

    http://www.itamarst.org

he's a clever guy.  met him in a previous incarnation.

Re: [9fans] yet another font package converted to plan9

[Jack Johnson]

andrey mirtchovski wrote:
> On Mon, 4 Aug 2003, Joel Salomon wrote:
>>Hebrew seems to imply that right-to-left text has been implemented in
>>plan9 - which I know it hasn't. Besides the text *is* backwards - those

If someone were to tackle some facility for right-to-left text, would it
make sense to try to do top-to-bottom at the same time (i.e. arbitrary
text orientation?), or do you think it would be better to have a system
that was entirely one orientation or another?

I haven't been in an environment with languages in multiple
orientations.  Would it make sense to read something on the order of 'rm
c.hsabta' or 'rm hsabta.c' where hsabta is atbash in Hebrew glyphs?  Or
would it be relatively easier to just read 'c.hsabta mr' from right to left?

In a mixed-orientation environment, what would be the correct output of
'ls *.c'?

-Jack

(just musing)

Re: [9fans] yet another font package converted to plan9

[boyd, rounin]

> If someone were to tackle some facility for right-to-left text, would it
> make sense to try to do top-to-bottom at the same time ...

yes, for the japanese; top-to-bottom then right-to-left.

Re: [9fans] yet another font package converted to plan9

[Skip Tavakkolian]

> If someone were to tackle some facility for right-to-left text, would it
> make sense to try to do top-to-bottom at the same time (i.e. arbitrary
> text orientation?), or do you think it would be better to have a system
> that was entirely one orientation or another?
>
> I haven't been in an environment with languages in multiple
> orientations.  Would it make sense to read something on the order of 'rm
> c.hsabta' or 'rm hsabta.c' where hsabta is atbash in Hebrew glyphs?  Or
> would it be relatively easier to just read 'c.hsabta mr' from right to left?
>
> In a mixed-orientation environment, what would be the correct output of
> 'ls *.c'?

Unicode has a report on this very subject. I confess I've
done a cursory review of it and can't judge if it is reasonable or
not.

http://www.unicode.org/unicode/reports/tr9/

Re: [9fans] tfel-ot-thgir support

[Joel Salomon]

> Has anyone done any critical thinking on this?

Would it be terribly inappropriate to mention Unicode Annex #9
http://www.unicode.org/reports/tr9/tr9-11.html ?

If anybody thinks so then I won't.

--Joel

Re: [9fans] tfel-ot-thgir support

[boyd, rounin]

> Would it be terribly inappropriate to mention Unicode Annex #9
> http://www.unicode.org/reports/tr9/tr9-11.html ?

iirc unicode has some 'typesetting' directives in it.  this is overkill.
it should just be a set of glyphs.

Re: [9fans] yet another font package converted to plan9

[andrey mirtchovski]

On Mon, 4 Aug 2003, Jack Johnson wrote:

> If someone were to tackle some facility for right-to-left text, would it
> make sense to try to do top-to-bottom at the same time (i.e. arbitrary
> text orientation?), or do you think it would be better to have a system
> that was entirely one orientation or another?
>

Rob Pike may possibly disagree, but since all text displayed in Plan 9 is
(arguably) the same as any other image you see on the screen, switching from
left-to-right to tfel-ot-thgir _should_ be as simple as changing Dx() to
-Dx().

top to bottom would be: Dy() = height + C; Dx() = 0; :)

related topic: I've heard 'freetype' mentioned on this list before.
does Charles Forsyth's port allow for antialiased fonts text?

http://groups.google.com/groups?q=freetype+group:comp.os.plan9&hl=en&lr=&ie=UTF-8&oe=UTF-8&group=comp.os.plan9&selm=ec836bd9.0206161203.634832f0%40posting.google.com&rnum=1

I know it's going to be in Inferno R4, those fonts aren't going to be
available for Plan 9? There are GPL TTFs now, can we use those?

andrey

Re: [9fans] tfel-ot-thgir support

[Skip Tavakkolian]

> Would it be terribly inappropriate to mention Unicode Annex #9 
> http://www.unicode.org/reports/tr9/tr9-11.html ?

I just did too ☺

It is worth mentioning that there is an opensource implementation
ot the algorithms at:

http://fribidi.sourceforge.net/

I downloaded it several weeks ago, and that is as far as I got.

Re: [9fans] tfel-ot-thgir support

[William Ahern]

On Tue, Aug 05, 2003 at 12:50:58AM +0200, boyd, rounin wrote:
> > Would it be terribly inappropriate to mention Unicode Annex #9
> > http://www.unicode.org/reports/tr9/tr9-11.html ?
>
> iirc unicode has some 'typesetting' directives in it.  this is overkill.
> it should just be a set of glyphs.

	"However, in the case of bidirectional text, there are circumstances
	where an implicit bidirectional ordering is not sufficient to
	produce comprehensible text. To deal with these cases, a minimal set
	of directional formatting codes is defined to control the ordering
	of characters when rendered."

Re: [9fans] tfel-ot-thgir support

[boyd, rounin]

> "However, in the case of bidirectional text, there are circumstances
> where an implicit bidirectional ordering is not sufficient to
> produce comprehensible text. To deal with these cases, a minimal set
> of directional formatting codes is defined to control the ordering
> of characters when rendered."

blah, blah, blah.  glyphs != page layout.

does it solve the general problem of bidirectional text in two dimensions?

Re: [9fans] tfel-ot-thgir support

[William Ahern]

On Tue, Aug 05, 2003 at 01:02:03AM +0200, boyd, rounin wrote:
> > "However, in the case of bidirectional text, there are circumstances
> > where an implicit bidirectional ordering is not sufficient to
> > produce comprehensible text. To deal with these cases, a minimal set
> > of directional formatting codes is defined to control the ordering
> > of characters when rendered."
>
> blah, blah, blah.  glyphs != page layout.
>
> does it solve the general problem of bidirectional text in two dimensions?
>

hmmmm... good point

Re: [9fans] tfel-ot-thgir support

[Joel Salomon]

> does it solve the general problem of bidirectional text in two dimensions?

from http://www.unicode.org/faq/middleeast.html :

> Q: Are any scripts written vertically?
>
> A:
<snip>
> Unlike the bidirectional case, the choice of vertical layout is usually
> treated as a formatting style; therefore, the Unicode Standard does not
> define default rendering behavior for vertical text nor provide
> directionality controls designed to override such behavior.

Unicode also does not provide formatting codes for bottom-to-top or
boustrophedon scripts. The RTL and LTR are for multiple levels of embedding
(LTR within RTL within LTR etc.) so that a sentence like ("logical"
order, ALL CAPS are RTL as in unicode examples):
 AVRAHAM SAID "the first two books of the bible are BERESHITH and SHEMOTH."
will print as:
 "the first two books of the bible are HTIHSEREB and HTOMEHS." DIAS MAHARVA
instead of the result of pure algorithmic bidi:
 ".HTOMEHS and HTIHSEREB the first two books of the bible are" DIAS MAHARVA

(Example from "Unicode Demystified", Richard Gillam, Addisson Wesley, 2003)

--Joel

Re: [9fans] tfel-ot-thgir support

[boyd, rounin]

ssem yltsahg

Re: [9fans] yet another font package converted to plan9

[Jim Choate]

On Tue, 5 Aug 2003, boyd, rounin wrote:

> > If someone were to tackle some facility for right-to-left text, would it
> > make sense to try to do top-to-bottom at the same time ...
>
> yes, for the japanese; top-to-bottom then right-to-left.

Don't forget they read back to front as well.


 --
    ____________________________________________________________________

      We are all interested in the future for that is where you and I
      are going to spend the rest of our lives.

                              Criswell, "Plan 9 from Outer Space"

      ravage@ssz.com                            jchoate@open-forge.org
      www.ssz.com                               www.open-forge.org
    --------------------------------------------------------------------

Re: [9fans] tfel-ot-thgir support

[Jim Choate]

On Tue, 5 Aug 2003, boyd, rounin wrote:

> does it solve the general problem of bidirectional text in two dimensions?

Make that 3D and I'll go for it.


 --
    ____________________________________________________________________

      We are all interested in the future for that is where you and I
      are going to spend the rest of our lives.

                              Criswell, "Plan 9 from Outer Space"

      ravage@ssz.com                            jchoate@open-forge.org
      www.ssz.com                               www.open-forge.org
    --------------------------------------------------------------------

Re: [9fans] yet another font package converted to plan9

[Jim Choate]

On Mon, 4 Aug 2003, Joel Salomon wrote:

> > Don't forget they read back to front as well.
> That's only a problem for troff.

Hardly. You're only thinking of static documents.


 --
    ____________________________________________________________________

      We are all interested in the future for that is where you and I
      are going to spend the rest of our lives.

                              Criswell, "Plan 9 from Outer Space"

      ravage@ssz.com                            jchoate@open-forge.org
      www.ssz.com                               www.open-forge.org
    --------------------------------------------------------------------

Re: [9fans] yet another font package converted to plan9

[Joel Salomon]

> Don't forget they read back to front as well.
That's only a problem for troff.

--Joel

Re: [9fans] yet another font package converted to plan9

[Joel Salomon]

On Mon, 4 Aug 2003, Joel Salomon wrote:

> > Don't forget they read back to front as well.
> That's only a problem for troff.
>
Correction: troff wouldn't have a problem (assumming, of course that the
basic quad-di problem had been worked out). The pm postprocessor (that
adjusted page margins) might, though, as would any program that set up
pages for hard-cover binding (i.e. many pages on one large physichal
sheet which is them cut and folded).


--Joel

Re: [9fans] yet another font package converted to plan9

[Joel Salomon]

> > > Don't forget they read back to front as well.
> > That's only a problem for troff.
>
> Hardly. You're only thinking of static documents.
>

OK, I can see the problem for bound dead trees. Documents on-line usually
present the fiction of an arbitrarily long scroll (most web pages) or can
present the pages in "logical" (reading) order. Where else would the
back-to-front problem arise?

> Make that 3D and I'll go for it.

Oh, I see now...


 --Joel

Re: [9fans] yet another font package converted to plan9

[okamoto]

> Don't forget they read back to front as well.

What does this mean?

Kenji

Re: [9fans] yet another font package converted to plan9

[matt]

okamoto@granite.cias.osakafu-u.ac.jp wrote:

>>Don't forget they read back to front as well.
>>
>>
>
>What does this mean?
>

Western books are bound with the spine is on the left

Japanese books are bound with the spine on the right

So each can say that the other reads books "back to front"

Re: [9fans] yet another font package converted to plan9

[okamoto]

> Western books are bound with the spine is on the left
>
> Japanese books are bound with the spine on the right
>
> So each can say that the other reads books "back to front"

I see.  It's only the case when the book is written in the
'formal' Japanese text.  (from top right to down and then shift to left).

Kenji