Arabic Layout TF telecon

28 Jun 2016

See also: IRC log


Shervin, r12a, Mostafa, Najib


<scribe> ScribeNick: mostafa

<najib> https://github.com/w3c/alreq/wiki/Draft-for-%E2%80%9C2.3-Direction%E2%80%9D

Shervin: In the wiki page for the “2.3 direction’ draft I’m seeing this draft.

Mostafa: I didn’t go into fine details and edge cases in this draft.
... It’s just a big picture overview of how the Arabic text is rendered.

<Shervin> http://unicode.org/cldr/utility/bidi.jsp?a=%DA%A9%D9%84%D9%85%D9%87+word&p=Auto

Richard: The image for the logical/visual order can be a little confusing.
... An arrow could help.

Shervin: Another option is to add position numbers.

<najib> s/Heh/Teh/

<najib> s/U+0647/U+0629/

Richard: We can use “display order” instead of “visual order” to make it more clear.

Shervin: We can also use something like “memory order” for “logical order”.
... Or “reordered string” instead of “visual order”.
... So, do you want me to merge this PR?

Mostafa: No, I have to update this using your feedbacks.


Shervin: On the justification, there are two schools of thought.
... One is discussed in the article “Justify Just or Just Jusify”.
... The other is writen about by Thomas Milo.

Richard: Azzedine’s approach is based on adding more kashida.
... Tom’s approach also uses extra spaces between words and inside words.

Shervin: In reality there should be a mix of kashida and inter-word spaces.

Najib: Kashida and inter-word spaces are the main justification mechanisms used in type-writers.
... Justification using ligatures is more complicated and not easy to achieve using mechanics.

Richard: In InDesign they do allow you to use ligatures and swash kaafs in the end of the words, which is not actually kashida.
... Tom’s application, Tasmeem, is used in InDesign.
... But he also thinks that you have to use more spaces than what Azzedine suggests.
... In type-writers a tatweel is used, not a kashida. They are different.
... These are the tools used for streching the lines for justification.
... And there are the rules for each writing style and each tool.

Shervin: I agree that it’s not black or white.
... It’s also totally dependant on the environment.
... More complex environments can have, for example, longer kaaf.

<Shervin> Shervin-

Shervin: Plain text environments are more limited and can probably only use tatweels and extra spaces.
... We just have to cover the nuances of using these models.

Mostafa: We can only cover all these tools and solutions, and mention that there are different approaches to mixing these tools. We don’t have to make a decision ourselve.s

Shervin: +1

Richard: At a later step, we can also cover how browsers implement Arabic justification.
... I suggest to use Tom Milo’s definitions for tatweel and kashida.
... When he says “tatweel”, he means the Unicode character.
... But when he says kashida he means extending the horizontal line of the letters.

Shervin: Behnam calls that elongation.


scribe: Moving on to Najib’s new text.

Najib: I want to cover different families of numbers, what are their origins, and some of the issues regarding the number families like fonts and keyboards.

<najib> Writing and reading numbers, bidi with numbers

Shervin: I suggest avoiding Arabic digits term for the U+0030—U+0039.
... It’s defined in the glossary, we can use that.

Najib: I have a concern about pronounciation and how the Arab read their numbers
... What about Persian?

Shervin: I’m not sure about that section. We are talking about text, not how different languages read their texts.

Richard: +1

Najib: Action-37 is done.

Mostafa: Action-55 is still ongoing.

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.144 (CVS log)
$Date: 2016/06/28 17:35:16 $