Data on the Web Best Practices Working Group Teleconference

<hadleybeeman> PWinstanley: now, though I'm afraid I won't be joining the group

<PWinstanley> ok

<BernadetteLoscio> hello!

hi!

<antoine> hello

<RiccardoAlbertoni> Hello!

<PWinstanley> hi

<PWinstanley> no audio!

<RiccardoAlbertoni> following the instruction on the agenda page..

<BernadetteLoscio> https://www.w3.org/2013/dwbp/wiki/Status_of_comments_about_the_last_call_working_draft

<RiccardoAlbertoni> trackbot, start meeting

<trackbot> Meeting: Data on the Web Best Practices Working Group Teleconference

<trackbot> Date: 29 July 2016

<scribe> scribe: annette_g

<BernadetteLoscio> no :)

<BernadetteLoscio> we are here!

BernadetteLoscio: let's discuss the suggestions from the I18n group

annette_g: comment 11 looks easy, just a spelling correction

BernadetteLoscio: look at 13
... https://www.w3.org/2013/dwbp/wiki/Status_of_comments_about_the_last_call_working_draft
... annette, can you send a message to the I18n group to clarify?

annette_g: yes

BernadetteLoscio: comments 15 and 16 are also about locale parameters

<BernadetteLoscio> The WG feels that BP3 should include a reference or recommendation to consistently use BCP47 as the standard for language and locale identification and, informatively, to CLDR as the source for both representing specific localized formats and as a reference for specific locale data values.

BernadetteLoscio: for 15, we can include a recommendation to reference BCP47 in the implementation section

PROPOSED: that we add a reference to IETF BCP 47 to BP 3 in the implementation section.

<Caroline> +1

+1

<BernadetteLoscio> +1

<laufer> +1

<EricKauz> +1

<RiccardoAlbertoni> +1

<Makx> yes I am trying to get the right windows open on my screen

<antoine> +0 - need to see if there are relations with the other comment

<Makx> can someone send URL of BP3 please

<BernadetteLoscio> https://www.w3.org/TR/dwbp/#metadata

<Makx> OK..

antoine: my concern is that there may be something in the BP that conflicts with this.

<Makx> I disagree that language should be indicated by langauge codes

Caroline: we can address issues that arise in a later meeting.

antoine: I wouldn't want to have an example in example 15 that conflicts with the one in example 3.

<Makx> DCAT for example indicates language by URI that represents the language so the example is correct ...

BernadetteLoscio: we are consistent right now in example 3 and example 15.

<Makx> Example 15 should read: "The Library of Congress publishes URIs for languages (see [ISO639-1-LOC]" These are not language codes but URIs for languages -- which make a difference in Linked Data.

BernadetteLoscio: I'm going to send a message to the I18N asking if what they are suggesting is okay with our examples. We also need to ask them to clarify the above.

RESOLUTION: that we add a reference to IETF BCP 47 to BP 3 in the implementation section.

BernadetteLoscio: we will also get clarification from the I18N group as well.

<Makx> Please note that ISO 639-1 defines itself as "Codes for the representation of *names* of languages", not "Codes for the representation of languages"

BernadetteLoscio: next, comment 16
... this is about the need to tag natural language text with a natural language tag.

<Makx> Comment 16 is about direction of scripts

annette_g: I think we could add that to the implementation section

<Makx> The ask for acknowledging that this is a known problem without a good solution

<Makx> Let's acknolwledge that.

BernadetteLoscio: they say there is currently no good mechanism to indicate base direction.

<RiccardoAlbertoni> +1 to acknowledge the problem

Caroline: I think they want us to just mention that this is a problem.

annette_g: they are still saying that we should recommend people tag the language itself.

BernadetteLoscio: so should we just mention that this is an unsolved problem? or that there is no standard way to solve it?

annette_g: I think we can do that along with saying that you can tag the language.

BernadetteLoscio: can you write a paragraph that we can include?

annette_g: I can give that a try

<Makx> I think we should say all text need to be tagged. I always do that in specs.

BernadetteLoscio: it always helps to give them a concrete suggestion.

<Caroline> action to annette_g to write a message to i18n group to clarify the comment 13

<trackbot> Error finding 'to'. You can review and register nicknames at <http://www.w3.org/2013/dwbp/track/users>.

<Caroline> action to annette to write a message to i18n group to clarify the comment 13

<trackbot> Error finding 'to'. You can review and register nicknames at <http://www.w3.org/2013/dwbp/track/users>.

<Caroline> action to annette_g write a message to i18n group to clarify the comment 13

<trackbot> Error finding 'to'. You can review and register nicknames at <http://www.w3.org/2013/dwbp/track/users>.

<Caroline> action to BernadetteLoscio update the document according to comment 15 and send a message to i18n goup

<trackbot> Error finding 'to'. You can review and register nicknames at <http://www.w3.org/2013/dwbp/track/users>.

<Caroline> action BernadetteLoscio update the document according to comment 15 and send a message to i18n goup

<trackbot> Created ACTION-291 - Update the document according to comment 15 and send a message to i18n goup [on Bernadette Farias Loscio - due 2016-08-05].

<Caroline> action annette_g write a message to i18n group to clarify the comment 13

<trackbot> Created ACTION-292 - Write a message to i18n group to clarify the comment 13 [on Annette Greiner - due 2016-08-05].

<Caroline> action annette_g update the document according to comment 16 - explain that the text should be tagged

<trackbot> Created ACTION-293 - Update the document according to comment 16 - explain that the text should be tagged [on Annette Greiner - due 2016-08-05].

BernadetteLoscio: Antoine, what do you think about comment 17?

<Makx> +1 to Antoine

antoine: I suggested two options and I suggest we go with the simpler one. Replace the word country by language.

BernadetteLoscio: this is example 15

Makx: what the library of congress publishes is URIs for languages, so my proposal is to say the LoC publishes URIs for languages.
... we should avoid opening the can of worms about the names of languages versus the languages themselves.

<antoine> suggested text: "The Library of Congress publishes lists of ISO 639 languages as Linked Data (see [ISO639-1-LOC] for two-letter codes):"

<Makx> +1

<Caroline> +1

<laufer> +1

PROPOSED: that we use Makx's suggested text for example 15, of "The Library of Congress publishes lists of ISO 639 languages as Linked Data (see [ISO639-1-LOC] for two-letter codes".

+1

<Caroline> +1

<BernadetteLoscio> +1

<laufer> +1

<antoine> +1

<RiccardoAlbertoni> +1

<Makx> +1

RESOLUTION: that we use Makx's suggested text for example 15, of "The Library of Congress publishes lists of ISO 639 languages as Linked Data (see [ISO639-1-LOC] for two-letter codes".

<Makx> It was Antoine's proposal!

BernadetteLoscio: let's look at comment 12. Should we include a new benefit or just mention it in the introduction?

annette_g: I feel like it should be a minor mention

BernadetteLoscio: I think we should put it in the "why" section. what do you think?

annette_g: I agree.

<Caroline> +1

PROPOSED: That we add a mention in the Why section to BP13 that most machine-readable standardized formats are also locale neutral.

<Caroline> +1

<BernadetteLoscio> +1

<antoine> +1

+1

<laufer> +1

RESOLUTION: That we add a mention in the Why section to BP13 that most machine-readable standardized formats are also locale neutral.

<Makx> we can add that.

<BernadetteLoscio> https://www.w3.org/TR/dwbp/#metadata

BernadetteLoscio: last comment is comment 18. They say that in BPs 1, 2, and 3 we say metadata must be provided for both humans and computers, but we don't say that it should be provided in multiple languages. I think we should add this in section 8.2

<Makx> +q

<laufer> but only metadata?

annette_g: we can't tell people to use all the languages of the world.

<laufer> We do not say that the datsets should be provided in multiple languages...

BernadetteLoscio: we can say multiple languages and include the language of the user

<Caroline> I think laufer has a point that "We do not say that the datsets should be provided in multiple languages."

antoine: I think we should do only the first part

<laufer> +1 to makx

Makx: I think we should do both

BernadetteLoscio: can you send a message with a proposal?

<laufer> bye all... nice to meet you again...

Makx okay, I will

<laufer> how to record the minutes

<RiccardoAlbertoni> good week end!

<BernadetteLoscio> thank you all!