See also: IRC log
<Shervin> https://www.w3.org/International/groups/arabic-layout/track/actions/open
Najib: Re action-7, I’m waiting
for feedback on my PR.
... I’m waiting for receiving another file and updating my
text.
https://github.com/w3c/alreq/issues/31
Mostafa: I updated our scripts to
add CLDR data too, but there still a bunch of missing
characters.
... There is a list of characters from a national standard, and
I have a lis of characters from there that are missing from our
tables.
Shervin: Some control characters
are already in the root.
... Characters in the root are also available for every other
language too.
... So, if you add root, a lot of the first characters in your
listt will be added.
Richard: Thinking about the purposes of these lists in our documents, maybe adding these root characters may not be relevant to Persian or Arabic.
Shervin: These national standards
are not extensive. They are not being updated regularly.
... The root of CLDR has the positive point over that.
Richard: We can assume that ASCII characters are already there.
Shervin: +1, and that reduces this list.
Richard: So, the list we need in
our document is the characters we need for working with
Persian.
... And assuming that ASCII is already there we don’t have to
put ASCII characters in the list.
... And so we can have the characters from ICU and CLDR and
ISIRI 6219, excluding the ASCII ones.
... Mostafa, do you agree with that?
Mostafa: Yes. The only issue I have with that standard is being old.
Shervin: For what’s wrong in these standards, we can ammend them, like the isolated characters.
Richard: Yes, I have to say that isolated ones are very important and we should have them.
Mostafa: So, is there a similar solution for Arabic?
Shervin: Question for Naib and Richard: does standards for computation in Arabic exist?
Najib: Not that I’m aware of.
Richard: Me neither.
Shervin: There are encodings and keboard layouts in the operating systems, so there might be sources somewhere.
Shervin: Najib, can you ask about it?
Najib: I can search about it.
Richard: Can we use keyboards as a source for this?
Shervin: For Persian, these
standars of 6219 is inter-related with the keyboard
standards.
... I don’t think we have any characters in the keyboards that
doesn’t exist in 6219. Do you agree, Mostafa?
Mostafa: Yes, but I can look, to make sure.
<r12a> http://r12a.github.io/apps/encodings/
<r12a> show all the characters in this encoding:
Richard: For encodings, you can use this link and see a list of all the characters in an encoding.
Shervin: That’s a good start. Is there a document for these encodings?
Richard: We can see about that.
Shervin: Mostafa, what do you think?
Mostafa: I rememeber these old os-based encodings as being buggy. Are we sure these lists are free of bugs?
Richard: No.
Shervin: I understand, but in
lack of a better standard we can rely on that and if there are
missing characters there, we might have other sources for
that.
... This is still ongoing, then.
Richard: Action-40 has not been started.
Mostafa: Action-42 is kind of on-hold. I’m putting my focus on the charts.
Shervin: That makes sense.
... Re action-49 I have to see if all the required control
characters exist in ICU and CLDR.
... Re action-50, UN transliteration is pretty good. We can
stay with that.
Mostafa: Action-51 is done.
Najib: I can take some more actions.
Shervin: It would be great if you could take action-37.
<Shervin> https://docs.google.com/spreadsheets/d/1llcjYqJnwQfN2fx2-T3CeoZgAMhCPDwWiqhia9NdQGs/edit
Shervin: And I want to talk about
it a little bit.
... And I’m sharing a document about it.
<Shervin> https://docs.google.com/spreadsheets/d/1pebcAbf9UOL_jRL4zCUQidE7BBj-6S_mA4yLXMf-7yw/edit?usp=sharing
Shervin: This is controversial
topic: which numeral sets should we use in Arabic
countries?
... Roozbeh and I put together this data of what is being
actually used in all these countries.
... These data are based on bank notes, coins, newspapers,
…
... This is still an onging survery, expanding on public
education, traffic signs, etc.
... I don’t have a preference here, but if people use something
because of software problems, should we assumg that is the
common practice?
... I like to hear what Najib or Richard say about this.
Najib: Some newspapers use a
mix.
... I need to look deeper into it.
Shervin: You can use this survery and I’m interested and can contribute to it.
Richard: We don’t need to be prescriptive about what Arabic users should do. We can be descriptive.
<najib> +1
Richard: Say which set of numerals is being used in each locale.
Shervin: Certainly.
+1
https://github.com/w3c/alreq/pulls
Shervin: Let’s start with the pull requests.
<Shervin> http://shervinafshar.github.io/alreq/
<Shervin> http://shervinafshar.github.io/alreq/#glossary
Shervin: PR #35 is for me. This is about adding transliterations to the tables.
<Shervin> https://github.com/shervinafshar/alreq/blob/533d20d94c5e9339695a1945188589ac6be6ad98/index.html
Shervin: This is what has
changed.
... As Richard recommended, I changed the direction using CSS,
rather than HTML.
... I looked into RSpec, but was not sure how to check
compatibility with it. Richard, can elaborate on that?
<r12a> https://www.w3.org/respec/
Richard: s/RSpec/ReSpec/
... ReSpec has definitions for some items, and we can use that
instead of redefining them.
... We’d better apply direction and language in the HTML. That
makes it easier to read the document without CSS.
... Langauge codes for transliterations mention Latin, but this
is Arabic, just not written in Arabic script.
<Shervin> "und-Latn-t-und-cyrl: The content is in the Latin script, transformed from the Cyrillic script." https://tools.ietf.org/html/rfc6497
Shervin: This example is from an
RFC we are working on.
... I guess you are right. I’ll pass it to someone more
familiar with these tags.
<Shervin> https://github.com/w3c/alreq/pull/34/files
Shervin: I can merge PR #34 if no
one has any objection with it, because it’s only the code for
the script.
... PR #35 I’ll reject, because the direction should not go in
the CSS.
... I can merge PR #33, but, Najib, you have to update the HTML
code with Richard’s comments and send another one.
... The other one is for Behnam, PR #30. I can merge that,
too.
... Isn’t there any objection here?
<Shervin> https://github.com/w3c/alreq/labels/drafting
Shervin: So I’ll merge #33, #34,
and #35.
... Issue #32 is for Najib to provide some text for.
<Shervin> ACTION Najib to produce text Github issue 32 re: deprecated Arabic presentation forms
<Shervin> ACTION: Najib to produce text Github issue 32 re: deprecated Arabic presentation forms [recorded in http://www.w3.org/2016/02/23-alreq-minutes.html#action01]
Mostafa: Issue #28 is ongoing: we need text for explaining the bi-di algorithm concisely and refer to Richard’s extensive explanation for more.
Shervin: Issue #22 has no action.
Is there anyone interseted in this?
... Since it’s supplemental, we can leave it for future.
... We used Azzedine’s list and actually merged it into our
list. We need to give some credit here.
Richard: You need beside having him as an editor?
<Shervin> s/Assedine/Azzeddine/
Shervin: Since he’s not actively drafting the document, can we have him as an editor or acknowledge him for his contribution.
<Shervin> ACTION: Shervin to add acknowledgement for Azzeddine's contribution to the glossary. [recorded in http://www.w3.org/2016/02/23-alreq-minutes.html#action02]
Richard: Since he is not actually an editor of the document, we can acknowledge him for his glossary work.