15:02:20 <RRSAgent> RRSAgent has joined #personalization
15:02:20 <RRSAgent> logging to https://www.w3.org/2022/02/07-personalization-irc
15:02:29 <Matthew_Atkinson> present+
15:02:30 <CharlesL> present+
15:02:30 <Lionel_Wolberger> present+
15:02:30 <janina> present+
15:02:34 <janina> chair: lionel
15:02:40 <janina> date: 07 feb 2022
15:02:40 <Matthew_Atkinson> rrsagent, make minutes
15:02:40 <RRSAgent> I have made the request to generate https://www.w3.org/2022/02/07-personalization-minutes.html Matthew_Atkinson
15:02:49 <sharon> sharon has joined #personalization
15:02:50 <janina> Meeting: Personalization Task Force
15:03:08 <sharon> present+
15:03:56 <becky> becky has joined #personalization
15:04:16 <janina> rrsagent, make log public
15:04:39 <Matthew_Atkinson> rrsagent, make minutes
15:04:39 <RRSAgent> I have made the request to generate https://www.w3.org/2022/02/07-personalization-minutes.html Matthew_Atkinson
15:04:43 <Lionel_Wolberger> Lionel is getting a cup of tea-- one minute
15:06:01 <becky> present+
15:07:10 <Lionel_Wolberger> present?
15:08:32 <Lionel_Wolberger> chair: Lionel
15:08:45 <becky> scribe: becky
15:08:57 <Lionel_Wolberger> agenda?
15:09:08 <becky> zakim, next item
15:09:08 <Zakim> agendum 1 -- Content Module Implementations Status (Follow-up on i18n issue #144) -- taken up [from Lionel_Wolberger]
15:10:21 <Matthew_Atkinson> q+
15:10:42 <becky> Lionel_Wolberger : thanks Matthew for all of the hard work on this; Rendering makes the point that right to left works; seems that some of the bliss symbols are pairs - like reading numbers in hebrew - have to read them as left to right; concerned that i18n will be concerned about symbol pairs
15:10:52 <janina> q+
15:11:38 <Lionel_Wolberger> q?
15:11:49 <becky> Matthew_Atkinson : do we expect symbols to be flipped?
15:13:01 <becky> Lionel_Wolberger : some of bliss symbols look like two combined symbols - but these don't seem to flip in right to left, should they or are the symbols units as a whole
15:13:08 <Lionel_Wolberger> q?
15:13:11 <Lionel_Wolberger> ack Matthew_Atkinson
15:13:24 <Matthew_Atkinson> q+
15:14:28 <becky> janina : I don't think we have to worry about this; I18N is concerned the right to left will work. But doubling or flipping of symbols is an implementation detail; We have proved the main point of right to left; other details can be resolved within CR
15:14:28 <Lionel_Wolberger> ack janina
15:15:52 <becky> Matthew_Atkinson: 3 quetions: do the symbols used make sense - we don't have enough info to answer; 2: should the symbols be flipped, again we don't have enough info but Janina pointed out that may not need to be immediately resolved
15:16:42 <becky> Matthew_Atkinson : symbols on the line beginning with "drain" are different between the two versions; that may be okay since the hebrew wording may be different - need to double check that
15:17:03 <Lionel_Wolberger> q?
15:17:07 <Lionel_Wolberger> ack Matthew_Atkinson
15:17:25 <becky> Matthew_Atkinson : if we are reasonably happy with the current version can we share / present with the chairs of other groups
15:18:16 <becky> Lionel_Wolberger : shares screen and discusses symbols
15:19:06 <CharlesL> q+
15:19:23 <Matthew_Atkinson> ack Matthew_Atkinson
15:19:40 <becky> Lionel_Wolberger : ingredients symbol is a triangle glyph and a glyph of a circle with a line under - we don't know if this is one symbol made of two glyphs or two separate symbols
15:20:35 <becky> Lionel_Wolberger : discusses 2 cans symbols which is also two glyphs or symbols and same issue with garbanzo beans
15:21:17 <janina> q+
15:21:50 <Matthew_Atkinson> Here's the markup for the "2 cans..." line: <span>2 </span><span data-symbol="22590">cans</span><span> of garbanzo </span><span data-symbol="24021">beans</span>
15:22:46 <becky> Lionel_Wolberger : one symbol, 22590 is what is generating the 2glyphs on the screen
15:24:41 <Lionel_Wolberger> q?
15:24:54 <becky> Lionel_Wolberger : looks at hebrew version: sees the same triangle and circle floating over the line - in the same order. What to understand if these glyphs should be in reverse order? Concerned that I18N may be raise a concern about that.  Believe we would be better served to create an example that uses symbols that are made up of only singular glyphs
15:25:46 <Matthew_Atkinson> Per CharlesL, here is the markup for the garlic line: <span>2 cloves </span><span data-symbol="14423">garlic, </span><span data-symbol="15190">peeled</span>
15:26:00 <Matthew_Atkinson> (This markup is the same as in issue 144)
15:26:29 <becky> CharlesL: noted that the "domino" symbol is representing an actual value of 2 or if it is meant to be a plural symbol
15:26:32 <Lionel_Wolberger> q?
15:27:02 <Lionel_Wolberger> ack CharlesL
15:27:52 <Lionel_Wolberger> +1 to Charles' box idea, but it sounds like development work
15:27:54 <Matthew_Atkinson> q+
15:28:17 <becky> CharlesL : it might be useful to put boxes around each single symbol to help distinguish them. That would help to specify that the order of the symbols is responding to direction (not the direction within a symbol made up of multiple glyphs)
15:28:20 <becky> jana
15:28:51 <Lionel_Wolberger> Lionel realizes - we could add the red boxes in an image, and share an image and not HTML
15:29:57 <becky> janina : if translation of text requires a different metaphor / terms are different - symbols are tracking the text
15:29:59 <Lionel_Wolberger> q?
15:30:00 <becky> Matthew_Atkinson
15:30:02 <Lionel_Wolberger> ack janina
15:30:39 <janina> q+
15:31:20 <Lionel_Wolberger> ack Matthew_Atkinson
15:31:26 <becky> Matthew_Atkinson : boxes is a good idea, probably easier to implement than to find different symbols; we really need to get clarification on whether the glyphs within a symbol should match the direction of the language (ie be flipped)
15:32:14 <Lionel_Wolberger> q?
15:33:48 <becky> Lionel_Wolberger : fear that we need to resolve the symbol / glyph flipping issue to avoid i18n confusion
15:34:12 <Lionel_Wolberger> q?
15:34:18 <becky> Lionel_Wolberger : explains how the word hummus is used in hebrew
15:34:48 <becky> Matthew_Atkinson : there are symbols for numbers but they were not included in the translation to symbols
15:34:52 <Matthew_Atkinson> ack Matthew_Atkinson
15:35:39 <becky> janina : validate the bliss is concatenive - should ask Steve to validate
15:35:44 <Lionel_Wolberger> q?
15:35:57 <Lionel_Wolberger> ack janina
15:38:46 <becky> janina: we need to verify with Steve whether bliss is concatenative - individual symbols are made up of multiple glyphs
15:39:37 <becky> sharon: from looking at the bliss library there are several examples of concatenated symbols; pharmacy is one example
15:41:19 <becky> Matthew_Atkinson : we are not translating the language; just using the correct symbols for the existing words
15:42:54 <becky> Matthew_Atkinson: will send a screen grab to Steve with questions
15:43:35 <Lionel_Wolberger> q+
15:44:25 <becky> Matthew_Atkinson : do the symbols used make sense; should the symbols be flipped horizontally in the hebrew version; verify the symbols reflect the text directly (question about why symbols for line starting with drain are different); does addition of borders help or hinder?
15:44:50 <becky> janina : we are discussing use of borders for non-experts to understand how the symbols are built
15:46:16 <Matthew_Atkinson> https://github.com/w3c/personalization-semantics/issues/144
15:46:59 <becky> Lionel_Wolberger : believe we satisfied the question of whether or not the symbols should flip;  There is an example online, I spoke to Dr E A Draffan about this and she verified that symbols do not flip
15:47:58 <becky> Lionel_Wolberger : but looking at the online example, some do flip - confusing!
15:48:50 <becky> CharlesL : some flip and some don't;
15:49:10 <Lionel_Wolberger> q?
15:49:41 <becky> Matthew_Atkinson : these symbols are more like images; the bliss symbols are more abstract; I don't see an obvious concatenated synbol within this example
15:50:44 <Lionel_Wolberger> q?
15:53:17 <becky> Lionel_Wolberger :  in conversations with other, I learned that bliss is not well used, the more image like symbols are more common. However using bliss as the normative set is useful
15:53:25 <Lionel_Wolberger> q?
15:53:34 <Matthew_Atkinson> q+
15:53:38 <Matthew_Atkinson> ack Lionel_Wolberger
15:53:45 <Lionel_Wolberger> ack Lionel_Wolberger
15:55:57 <becky> Matthew_Atkinson : using bliss symbols in the example because that is what works. This is a live example that we can share; We can add the boxes around each symbol; We can add any of the questions that i18n may ask
15:56:18 <Lionel_Wolberger> Lionel laughs
15:56:45 <becky> zakim, topic issue #202
15:56:45 <Zakim> I don't understand 'topic issue #202', becky
15:56:46 <janina> https://github.com/w3c/personalization-semantics/issues/
15:56:56 <becky> Topic: issue 202
15:56:59 <janina> https://github.com/w3c/personalization-semantics/issues/
15:57:28 <becky> janina : came from chairs as a11y working group, wants to use our example as a technique
15:57:38 <Matthew_Atkinson> q+
15:57:45 <becky> janina : in 2.2
15:58:00 <becky> janina : techique for WCAG 2.2
16:00:31 <becky> janina: clarifying that this is not about the AAC example
16:00:39 <becky> rrsagent, make minutes
16:00:39 <RRSAgent> I have made the request to generate https://www.w3.org/2022/02/07-personalization-minutes.html becky
16:02:04 <becky> zakim, bye
16:02:04 <Zakim> leaving.  As of this point the attendees have been CharlesL, sharon, Roy, Matthew_Atkinson, LisaSeemanKest_, JF, janina, mike_beganyi, becky, Lionel_Wolberger, LisaSeemanKest,
16:02:04 <Zakim> Zakim has left #personalization
16:02:07 <Zakim> ... janina_, bgibson, .75
16:02:35 <becky> rrsagent, bye
16:02:35 <RRSAgent> I see no action items