<LisaSeemanKestenbaum> genda+ should we merge the first three attributes. This included issues in merging them, and if we do, how we change the names to avoid errors? See https://raw.githack.com/w3c/personalization-semantics/rewrite-prototype/content/index.html

<LisaSeemanKestenbaum> https://www.w3.org/wiki/TPAC/2019 16-20 September 2019 in Fukuoka, Japan.

webapps follow up

<LisaSeemanKestenbaum> scribe: charlesL

scrive: CharlesL

<Roy> https://www.w3.org/2017/08/telecon-info_personalization

<Roy> resent+

lisa: only do 3 attributes unless its not a high price thing. got more feedback from WebApps and waiting for Leonie

<LisaSeemanKestenbaum> charles is sending it

Leonie: cochare of webplatform WG
... , still wg where W3C looks after html, but most work being done in whatWG
... , at TPAC this group met with webplat took an action to create a vocabulary which you did. thanks
... , primary concern creating new attributes is expensive. A there is a use case, and B evidence for that use case, and C that this will address this usecase
... , question Marcus C. web components could be a solution. I think this won't let us meet the use case. we want authors to be able to identify an interactive element so that some 3rd party can come in and change some portion. Now most of the time we don't want any changes made but we want for those to be able to perosnalize this to be able to change it seemlessly.
... , you can't recreate the link exactly as a web component.
... , I think a call between the WebPlat and this taskforce if we can get him and other browser engineers would be a good idea for a next step.
... , there is some things to consider. keep attributes as small as possible. 2. very useful to much out in the wild evidence will be extremely important. the TF needs to pick our best option and then go create some scripts to show these methods work.
... , early days of ARIA we had to find partners to help us take a leap of faith and implement some solutions to demonstraight this can work.

John: My concern: 3 basic types of data. what kind of assurance we have a solution and they say no you can't do it that way. we come forward with proposals and they say no yo9u can't do it that way… and we are back to square 1

Leonie: results of using this new attribute will be visible: LongDesc was an issue because it wasn't visible. I think we can minimize the risk by keeping up the engagement which is why another call would be a good idea.
... , keep talking with them an show them some examples in the wild.

Lisa: we have 2 groups so far to implement and we told them that the spec could change and then they left. the other group has 1 more year of funding. BBC is interested. we have issues lthe longer we take to get something stable by waiting for WebApps. it will be harder if we wait for webapps… what does the browsers want .. we had originally when it was in ARIA we had 2 programmers making prototypes but they messed up when

we tried to do 1 attribute, so then we separated them out and then no one make mistakes. SO if we go back to 1 attributes programmers will make mistakes.

Leonie: first list of the different groupings there were 9 or 10 attriburtes (action, destination, attribute etc… they won't implement 9/10/15 attributes…. if we can find 3 attributes, URI, string, token.

<JF> 3 types/attributes: token, string, or URI

Lisa: when they messed it up it was the tokens,
... , alternative for people who doesn't do numbers, and another needs a simplier desciption.
... , even if browser doesn't implement this do we really need them?

Leonie. if you pick # of attrivbutes and create extentions and then implement external extentions.

scribe: , people who need these personalized content will end up doing with our without native browser implementation which would be a nice touch.
... , if we get adoption.

Lisa: it won't be the entire BBS but small implementations.

Leonie: but you need to do this in 12 months.
... , bits of the BBC is a good start. sounds like the partners you have is a great springboard .

Janinia: I think we also want to semantically markup the content as well. We agree Browsers won't solve this it will be polyfills or web components, extentions. So the reason to have to conversation early is to pick the path which is least resistance.
... , there is a grant clock running, passing the URI can work, passing the transformation to symbols, and how do we structure the one string attribute.

Leonie: do we need to do all 3 things at the same time. not suggesting that any is less important and save the more difficult one for later.

Janina: do the URI/Symbols first and make the text 2.0

JF: my concern is if we get into prefixed attributes we will have a solution that no one supports. we need a robust for serious adoption.

Leonie: I suggest to use data- attribute it also means data-action if we get to the point that convinces the browsers to just loose the data- attribute afterwarsds.

Janina: suggestion is don't think aui- consume the data- instead.

JF: aui- is already being used from conversations at TPAC. data- is not my preferred solution but this is a way forward.

Janina: we want to avoid the clashes

Leonie: run a grep on a wide variety of web pages to see if you have any collisions. so doing a search would be a good step. 10's of thousands of pages would give you a good idea.
... , by using data- you won't get any errors for conformance.

<JF> data-ACTION: https://magento.stackexchange.com/questions/22124/what-is-data-action-attribute-is-added-for-in-1-9-0-0

Leonie: , if you find any data-action then you may find folks doing things which you are trying to do.
... , it was the way HTML intended this kind of work to be done

<JF> data-ACTION: https://www.npmjs.com/package/data-action

Janinia: we need a wright up on this and this sounds like the right path.

<JF> +

Janinia: , we still need to look at data- if its used in the wild and it clashes we need analysis this is the next bit of evidence once we show success we can look at alpha, beta, gamma

JF: the URL I pasted data-action its some kind of NPM its a data action to bind to click events. It might do what we are trying to do.

Leonie: proves that the data- to pollyfill even if this won't work but would be a great way forward.

JF: change find/replace aui- to data- is a lift but hopefully won't be a

q_

scribe: , I like this way forward.

Lisa: whats nice about this I want to close on something sooner.

Leonie: Marcos is in Melborn so it will be in the middle of the night for him. I can point him to these minutes and does the data- sound like a good way forward.

Lisa: Yes thanks
... , thanks Leonie

charles: cant we use data-aui-action?

Leonie: I would not silo this. I would avoid that.

Janina: may get tim his semantic web.

Becky: I like the idea of data. we still want to limit it, we don't want too many. and I don't think data-action we may come up with some different but similar enough.

Lisa: if they are doing the same thing then action for controls as apposed to fields or direct links to a page we would want to use the same, we could be just an additional use case.

Leonie: make sure its a conflict in name only.

Janina: thanks this has been great.

Leonie: I will take this and see what Marco thinks.

Lisa: yes thanks Leonie.
... that was very helpful. anything more?

Leonie: do we need to look through our vocabulary and see if there is a conflict.

S/Leonie/Lisa/

scribe: , can someone take an action to look these to see if there is any conflicts.

Sharon: I can look at this do a web search related to data-action

<LisaSeemanKestenbaum> ACTION: sharon see if data-action. data-function-function data-desitnation have conflicts and arent compatable

<trackbot> Created ACTION-7 - See if data-action. data-function-function data-desitnation have conflicts and arent compatable [on Sharon Snider - due 2019-03-04].

Becky: not sure but the way it is written, onload when the page its the action of the element not a lot of detail in his … its also 5 years old. no github repo either.

Lisa: if it hasn't been updated in 5 years then it should be defunct at this point.

<LisaSeemanKestenbaum> https://www.w3.org/wiki/TPAC/2019

tpac

<LisaSeemanKestenbaum> 16-20 September 2019 in Fukuoka, Japan.

Lisa: I think we should pencil it in. we should have stable working drafts. Coga is thinking of going as well. how do other people think?

Janina: I think its good to talk but it is expensive and may be too early to commit.
... , it will be valuable but lets keep talking about it.

Lisa: thanks all very productive.

This is scribe.perl Revision: 1.154  of Date: 2018/09/25 16:35:56  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Present: CharlesL janina stevelee Becka11y Roy LisaSeemanKestenbaum sharon present thaddeus JF MichaelC
Found Scribe: charlesL
Inferring ScribeNick: CharlesL

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth

Found Date: 25 Feb 2019
People with action items: sharon

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)