Data on the Web Best Practices Working Group Teleconference -- 11 Nov 2016

<deirdrelee> trackbot start meeting

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

<trackbot> Date: 11 November 2016

<deirdrelee> chair: deirdrelee

<deirdrelee> scribe: car

<deirdrelee> scribe: Caroline_

<phila> agenda: https://www.w3.org/2013/dwbp/wiki/Meetings:Telecon20161111

<phila> scribeNick: Caroline_

<phila> Least meeting minutes

<deirdrelee> PROPOSED: Resolve minutes of last week'ls meeting https://www.w3.org/2016/10/28-dwbp-minutes

<annette_g> +1

<phila> +1

<hadleybeeman> +1

<newton> +1

<ericstephan> +0

<ericstephan> wasn't here

RESOLUTION: Resolve minutes of last week's meeting https://www.w3.org/2016/10/28-dwbp-minutes

<laufer> 0 (not present)

<Eric_Kauz> +0

BP implementations

https://docs.google.com/spreadsheets/d/1RhMGyG0ZYb73RkteYr39Xqt7f5xi0BUQmSiDqjkHOSA/edit#gid=853876221

<deirdrelee> Caroline_: thank you for helping to gather evidence

<deirdrelee> .. we have put this form together

<deirdrelee> ... metadata BP, it is interesting to see that BP1 has high number of evidence, 26 now

<PWinstanley> last meeting minutes - I am not included in the minutes but was there (see log) - so can it be amended?

<deirdrelee> ... for the others in metadata, we also have good number

<deirdrelee> ... licence bps has a lot of evidence

<deirdrelee> ... as does provenance, quality and versioning

<deirdrelee> ... for identifiers, we only have 5

<deirdrelee> ... we wanted to flag this one for discussion

<deirdrelee> ... format and vocabulary are okay

<deirdrelee> ... data access is also BPs we have flagged, low number of evidence

<newton> Bill's comment: "Too difficult to test in a meaningful way. In this system, no datasets have yet been taken offline, so the archiving process has not been developed."

<deirdrelee> ... then the one we are most concerned about - data preservation, it only has the minimum number of evidence

<deirdrelee> ... newton highlighted a comment from bill in relation to this

<deirdrelee> ... would like to discuss with the group

<deirdrelee> ... just to finish, we have enough evidence for remainder BPs

<deirdrelee> ... thanks for all your help

deirdrelee: data access and data preservation have low evidences numbers
... does anyone have any question?
... maybe for data preservation we should be targeting different people
... publishers who are in the archiving business

<ericstephan> good point deirdrelee +1

deirdrelee: they might be different stakeholders that might help us getting evidences
... do you know people that deal with archives?

phila: the people we know are the people we are already reaching, such as Christopher
... he was before he went to BBC
... as Bill said it is hard because people are not doing that yet
... it is not just numbers of how many evidences do we need
... it is also about quality
... Caroline_ pointed out that the BP we have the most evidences is metadata

<annette_g> not everything was evaluated for every BP

phila: it is frightening how many we still don't have
... some of the organizations we looked at are very important and gave us good evidences
... we should have at least one of the organizations the W3C Director mentioned by name
... the spreeadsheet we have is good

<newton> we do have wikidata and schema.org. I agree with the need for collect evidences from dbpedia and imdb

phila: I am not worried about having a couple lower
... about numbers I think we are okay
... I am worried about the quality, so we need to put more organizations

<deirdrelee> deirdrelee: what is next step plans?

<deirdrelee> Caroline_: herbert will come back with feedback on preservation by nov 21st. we had planned to have all evidence by yesterday, but if the group agrees to keep the process open, we can have a final sprint

<deirdrelee> ... we can then provide the final implementation report in html

<deirdrelee> phila: don't get hung up on the html, if you can generate it easily, fine. but don't spend whole day on it

https://www.w3.org/TR/prov-implementations/

<deirdrelee> hadleybeeman: I have taken wiki pages as implementation report

<deirdrelee> phila: need to have confidence it will still be there in 5 yrs timem

<hadleybeeman> https://www.w3.org/2011/gld/wiki/ORG_Implementations

hadleybeeman: this link is much less formal

<deirdrelee> phila: that is a good example, but it is for multiple docs, so don't feel compelled for it to be that detailed

phila: I asked Martin's help as well

<ericstephan> phila working on real time as well...

deirdrelee: if you agree we can wait until the 21st and finish the report on the 23rd

phila: if we have the implementation report ready on November 23rd we can have a call on the 25th and then have the transition
... we will have another call with the director and based on the evidence we have he might proceed as a candidate recommendation
... AC member will review and we must respond them
... hopeful they will be only editorial comments
... the problem is that takes us to Christmas
... no one here can predict what the proposed recommentation review by the membership will be
... it is unusual to don't have any comments

<hadleybeeman> PR: http://www.w3.org/2015/Process-20150901/#rec-pr

hadleybeeman: PR is 28 days long

phila: the Western hemisfere shuts from Dec 16 until January 2nd?
... another thing I want to hightlight is that once you get the PR is done
... riccardoAlbertoni, Antoine and ericstephan mentioned we would have another version of the vocabs
... if you do have time I strongly suggest we vote on the updated version of the 2 vocabs

<riccardoAlbertoni> sorry i did not get when is the last chance to vote for vocabularies, would you mind repeating it?

phila: Friday the 25th is okay to have the report finished and then have the call with the director on the 1st week of December
... at this stage let's keep it

<deirdrelee> ak newton

<ericstephan> just an fyi, 25th is a US holiday... I can be on but best also check with annette_g

phila: you can only make editorial changes now
... not content changes
... it is the last chance the group has to make any comments
... once you have the review from the membership if there are only editorial changes to be made it will be a recommendation

deirdrelee: we can start working on the implementation report right way

hadleybeeman: it took us a while to organize the call with the director

phila: I will send a message to him
... if we have comments during the CR we have to make sure to answer them
... we also need to record the comments
... what you did before on the wiki is fine

action to phila to get some dates to the director call

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

action phila to get some dates to the director call

<trackbot> Created ACTION-299 - Get some dates to the director call [on Phil Archer - due 2016-11-18].

action Caroline_ editors will address the comments we will receive during PR

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

action Caroline editors will address the comments we will receive during PR

<trackbot> Created ACTION-300 - Editors will address the comments we will receive during pr [on Caroline Burle - due 2016-11-18].

action Newton editors to prepare the Implementation Report

<trackbot> Created ACTION-301 - Editors to prepare the implementation report [on Newton Calegari - due 2016-11-18].

action phila adding high profile evidences

<trackbot> Created ACTION-302 - Adding high profile evidences [on Phil Archer - due 2016-11-18].

<annette_g> Twitter API is actually very good

newton: about the Twitter evidence
... instead of analizyng all the set annette_g analized the APIs

phila: what it looks lke is that it doesn't even provide metadata
... don't leave like this
... at the moment it looks like it fails most of the BPs

newton: there are more datasets analized the same way

<ericstephan> I focused on specific parts I did not analyze the entire site in many cases

newton: the Government of Alagoas
... Eric's datasets as well

annette_g: eerything I put on I went sraight to the BP

<ericstephan> I did the same, just very targetted dataset or resource against a BP

deirdrelee: would make send to make a colum to show that that dataset was to analyse that BP

annette_g: I think the table we present can be misinterpreted that we are trying to evaluate the websites
... the point is to get evidences to the BPs

<ericstephan> +1 annette_g

PWinstanley_: I agree

ericstephan: I agree with annette_g and that is the approach I have been taking
... it is easier to look for specific BPs we need to show evidence

<ericstephan> what I found was interesting was the WAYS implementations actually carried out the best practice

<hadleybeeman> +1 to Annette's idea

Caroline_: how to show the table or the evidences that doesn't seem that we are evaluating the datasets, but actually getting evidences that our BPs are actually good

annette_g: changing the table, maybe creating a list

deirdrelee: write an exaplnation about the methodology we used

phila: if you have a object to each BP or a list of implementation you can write a script to make it easier
... you can use a selection of things, which would not necessary a table

<ericstephan> +1 phila to the json idea

phila: if you do a litle bit of hacking :)

newton: I will use tools to do that

<phila> Czech Republic

phila: that page lists all the BPs

newton: we will discuss among the editors and try to base the Implementation Report on something like that

<newton> we'll point on the Implementation Report about the methodology as well

annette_g: if we have 30 implementations of a given BP is there any value of listing all those?

hadleybeeman: it helps to demonstrate to the director that they have value
... I think it is better to include as much implementations as possible

<ericstephan> which sampling is biased toward the order of the list :-)

annette_g: since the form was so large it became dificult

hadleybeeman: we can use as an explanation if needed

deirdrelee: it is a good point

<Zakim> riccardoAlbertoni, you wanted to ask if i got it right: the 25th nov is the last chance date for voting vocabularies

riccardoAlbertoni: did I got it right: the 25th nov is the last chance date for voting vocabularies?

phila: it is right
... this meeting in 2 weeks time will be the last meeting we will have
... we can vote to the transition for the BP document and the changes on the vocabs documents
... if you make any changes I suggest it have to be done very soon so the WG can have a look

<annette_g> I just added four memento examples for BP 11 to my spreadsheet

thank you annette_g :)

riccardoAlbertoni: there are comments that we received
... would be a good idea to consider them

ericstephan: question to the BP editors
... the evidences I found was based on a target to getting evidences

newton: it would be very much helpful

deirdrelee: could you talk about the Workshop to be held in Amsterdan?

SDSVoc https://www.w3.org/2016/11/sdsvoc/agenda

phila: the link of the agenda is above
... the motivation for this workshop there are multiple

<laufer> Unfortunately, I won´t go... :(

phila: there are a lot of evidences that DCAT needs a WG
... one of the reasons for teh Workshop is because it should start a WG and that is a continuation of this WG
... more interestinly form a eng point of view is that we dont'have a agreed way to provide a machine readable format
... to see if you can link automaticaly
... datasets
... I'd expect to point stuff that already exists
... you should be able to request data, particularly metadata
... "can you give me the metadata record using DCAT, or CKAN, etc"
... the workshop is a success WG to this
... in terms of the agenda day 1 is a lot about DCAT
... day 2 is about content negotiation
... and what we will do in a new WG

<phila> Attendee list

deirdrelee: a lot for 2 days!

<laufer> ruben, ivan...

<ericstephan> okay, take care all!

deirdrelee: thank you everyone for attending today!

<riccardoAlbertoni> bye, good weekend!

<laufer> bye all... good sun for all...

deirdrelee: we will have another meeting on November 25th

<annette_g> bye!

- DRAFT -

Data on the Web Best Practices Working Group Teleconference

11 Nov 2016

Attendees

Contents

BP implementations

SDSVoc https://www.w3.org/2016/11/sdsvoc/agenda

Summary of Action Items

Summary of Resolutions

Scribe.perl diagnostic output