W3C

– DRAFT –
Personalization Task Force Teleconference

14 June 2021

Attendees

Present
becky, CharlesL, JF, Matthew_Atkinson
Regrets
-
Chair
-
Scribe
becky, JF, Lionel_Wolberger

Meeting minutes

Editor's note on action-destination-purpose (Matthew and John) -- Matthew sent the proposed editor's note

MA: we've been working on this - refocus on providing the deliverables as expected and this is a work in progress.
… seeking feedback and comments

This note outlines the different approaches, with code samples. Also includes a matrix with all of the options in one place (view)

<becky> https://github.com/w3c/personalization-semantics/wiki/Action-Destination-Purpose-Research-Questions

<becky> https://lists.w3.org/Archives/Public/public-personalization-tf/2021Jun/0011.html

LW: are these really Research Questions? Isn't this more seeking feedback?

MA: sure, what we call is less important than what we are seeking, which is more feedback

JS: likes "Request for Feedback" - implies that we want that feedback

BG: this is also to capture history, right? Historical data

LW: agree with Becky - this is a presentation of the topic

CL: fine with request for feedback. Want to note that changing the title changes the URL (wiki)

CL: this is the first time I've seen "meaning" - is this new?

MA: when I did the demo page, this was one of the options. This is a proposal that would apply to a change of approach

so, "meaning" was a new concept - I added the name but open to suggestions

it's a step on the way - a less drastic approach (merging action and destination) - there is also another proposal that re-uses attributes that are in common

CL: fine if this is just for illustration

MA: we can choose a different name if this is confusing. But we can certainly change it

BG: don't want to go down a rabbit hole.. if we already have purpose, why would we need meaning? So why (we will need an answer for that)

CL: the MathML working group is doing work on a proposed attribute "@intent"

to be used in MathML - "what is this for in the mathematical equation?"

LW: introducing a new word has kicked off new discussion - suggest instead "action-destination" instead of "meaning"

MA: 1) this is a placeholder, 2) contemplated that but it makes it a long attribute name

MA: to Charles... re: @intent - interesting. Considered it, but it is a common term already, so we need to be mindful of that.

JS: Noted that MathML was working on @intent as well - interesting.

to the extent that we can re-use... that may be useful. Is MathML looking at a prefixed set of attributes?

CL: no it would be part of the MathML spec

<Lionel_Wolberger> q/

JF: -1 to "@action-destination" - too long. We could note that @meaning = @action+@destination

CL: how will we cross link this? Note that it's just a wiki page today

MA: the idea is to link to the wiki page from the Editor's Note
… tried to keep this as neutral as possible - without showing a preference to one or the other

but I do understand that this may actually introduce more confusion

if @meaning isn't working we can perhaps come up with something else?

<janina> Proposed Ed Note: https://lists.w3.org/Archives/Public/public-personalization-tf/2021Jun/0011.html

<Lionel_Wolberger> q/

BG: only concern is I don't recall us deciding to merge this. Feel like we haven't discussed this as a group

<janina> +1 tro Becky that this is new

LW: we can take that as a consideration, but we have discussed this

JF: we have discussed this before. As I recall, the idea is to continue with the original 3, and note in the Working Draft that these are some open quesions

LW: what are next steps, and should we have this as a public conversation. There is some resistance to continued discussion

LW: looking to create some concensus. we could add a note that ActionDestination (or some other notation) is an open idea

Don;t want to get hung up on terms

LW: understand why @meaning becomes a problem, so perhaps using camelcase or hyphenated term

BG: concern is that we are going to CR - if we have 3 different proposals, we're going to lose time

we had 2 proposals, now it seems we have 3

BG: we were going to have 1 or 3, now this is a new proposal (2)

suggests that we haven't discussed this a lot

but this seems to be opening up the discussion even more - root of concern

BG: so, are we introducing more confusion?

this seems we're not making it clear - it was pick one or the other, now it's pick one of three

BG: less concerned on what we call it - the concern is that there is now 3 different approaches

JF: this was driven by a realization that action and destination were often confused, or not reliably distinguishable

<JF> JS: want to underscore the context - if we are just updating the Working Draft not too bad. But if this is for CR then that introduces more concern

<JF> JS: now hearing a question - what is being proposed here? 2 alternative approaches or 3?

<JF> worry that we add those questions to a CR document - maybe just update the working draft for now

<JF> JS: important that we are clear on this before we go to CR

<JF> MA: not looking to delay things. Just want to ask that once final edits are completed, really implore that you look at the table (matrix) at the bottom

<JF> shows the implications (as we understand them today) of the different approaches

<JF> MA: trying to make it clear that we propose to continue as is, but also want to anticipate some feedback

<JF> MA: review of the matrix will help

<JF> LW: clarification of edits

<JF> LW: We have a useful document, we are engaging with this

<JF> JF: suggests moving matrix to top of document (the TL;DR as it were)

<JF> BG: noting scrolling issue in Chrome

<JF> [discussion: the table/matrix is an embedded content with markdown]

<JF> LW: 3 attributes (Action/Destination/Purpose) or 1, now we might have a 3rd option (Action+Destination/Purpose)

<JF> LW: we need a 'name' for the combined Action/Destination - can use hyphen, CamelCase, or something else

+1

Lionel agrees with Becky, adding meaning is confusing

<Matthew_Atkinson> I'm OK with changing my "@purpose (for action-destination) and @meaning (for form controls)" example to "@action-destination and @purpose" as that's clearer.

<JF> [discussion]

<JF> LW; we can have Mathew do some cleanup - confusion around @meaning (and resultant discussion)

<JF> LW: so we need to have the author(s) do some edits

<JF> JF: proposes that Matthew makes some edits, shares the changes via email (and some other cleanup ideas)

+1 table on top

<Zakim> Matthew_Atkinson, you wanted to confirm revisions needed (there are three I think)

<JF> LW: so Matthew will make the edits, we should all read that once he's done (will send out an email) so that we can finalize next week

<JF> LW: [discussion on scrolling table in wiki]

<JF> BG: what if we do it in markdown

<JF> +1 BEcky

<JF> BG: wil take a stab at it

<JF> BG: it's a tedious process

<JF> MA: will try doing this via a conversion process - will communicate with Becky if he needs help

<JF> BG: it's only the one table that fails

<JF> MA: rendering the data in a table is useful

<JF> -1 to an appendix to a linked note

<JF> JS: we will need to continue the discussion of the name of the page - important!

LW: want to change the name sooner rather than later but Janina believes we should discuss the name next week

Janina: want to make sure the name is set before share with the public

Minutes manually created (not a transcript), formatted by scribe.perl version 136 (Thu May 27 13:50:24 2021 UTC).

Diagnostics

Succeeded: s/do it in HTML markup/do it in markdown

Maybe present: BG, CL, Janina, JS, LW, MA