Meeting minutes
<PWinstanley> https://
<PWinstanley> proposed: accept https://
PROPOSED: accept minutes of last meeting
<riccardoAlbertoni> +1
<PWinstanley> +1
<DaveBrowning> +1
0 wasn't there
RESOLUTION: accept https://
<AndreaPerego> 0 (was not there)
PWinstanley: welcome new member Nobumichi Ogura
(group introductions follow)
Nobu_OGURA_DSA: I'm located in Japan, belong to Data Society Association in Japan, the main association in japan related to data trading. They have 100 something affiliates.
He's not a specialist in data activities, background more in industrial engineering, has been working on ISOs, etc. Wanted to come in to work more in international aspects of data.
DSA has been using version 1, is converting to version 2.
would like to learn more of the background on DCAT so that he can contribute.
PWinstanley reviews our charter
PWinstanley: two main deliverables, DCAT and profile negotiation. Each has its own meetings in addition to the plenary.
PWinstanley: We are transitioning from extensive chartering of a working group with expectation for a 3 year period or so, and meetings with other working groups at W3C's own plenary meetings.
… We're moving toward a bit of an agile development process for recommendations.
W3C is moving toward having shorter cycles. We also need our W3C representative, Phillippe, as guide through the process.
Pre-covid, we would have face-to-face meetings every year or so, but those are not on the horizon for a long time.
We're moving meetings to every two weeks, since we have a light workload at the moment and it's a big time investment at difficult hours of the day.
You may want to talk to others in a smaller group to get to know people.
We have a mailing list, and you may want to join the DCAT subgroup.
any questions?
Nobu_OGURA_DSA: I was able to access the DCAT subgroup, what is the other one?
PWinstanley: you can use content negotiation to get data with a range of options, like media type and range. The server delivers the highest precedent item it can. We're working to enable that kind of negotiation with profiles.
… the working subgroup has also been working with IETF.
… They've had to wait quite a bit for responses from IETF. One of the challenges of the plenary is to make some significant decisions for the future of that.
… Maybe we would want to do a full recommendation, or just a white paper.
DCAT updates
PWinstanley: we would need more people to do the first agenda item.
… Riccardo, can you give an update?
riccardoAlbertoni: good news is the publication of the working draft! Just before Christmas, we had a meeting about feedback we've received.
<riccardoAlbertoni> https://
riccardoAlbertoni: We created that issue^.
Mackx was reporting some concern about modeling of series of datasets. Mackx thinks the current solution is elegant and sound. He just noted some folks are bound to an implementation with CKAN and can't change their implementation as a result.
… we didn't reach an editorial solution. We have left the ability to use a dataset like a distribution, and Mackx was concerned about having two possibilities of how the terms are used. He would suggest being more prescriptive and using the solution with dataset as a member of a series.
… we understand the legacy implementations might not conform to that.
… The question is how prescriptive we want to be and if our handling of legacy implentations is in agreement with W3C rules.
… We can try to define DCAT 3 as superseding DCAT 2 without that difference.
PWinstanley: the engagement strategy is to get something back from publication of the working draft. What are people doing to ensure we get feedback?
riccardoAlbertoni: I am sharing the poster on LinkedIn. A lot of the people I would have emailed already saw that on LinkedIn.
<PWinstanley> https://
DaveBrowning: I'm doing much the same as riccardoAlbertoni.
PWinstanley: are people giving talks of any sort?
nope
PWinstanley: Alejandra and I have a talk coming up with a group related to the FAIR principles and some of the work the DDI/CDI are doing.
PWinstanley: How long do you think we leave it before we say we've given everybody a chance?
… I guess what that's also looking at is when do we think we'll have version 3 completed?
AndreaPerego: I think we have to make a plan. We should try to go out within the org in June, we need to schedule a new working draft and identify what we want to address.
… Usually riccardoAlbertoni has a good idea.
riccardoAlbertoni: before christmas we planned to have another working draft at the beginning of March, then try to get a candidate rec by June.
The other working draft is delayed a bit, so that may change.
A subset of editors agreed on working on DCAT until June, then people will need to understand what will happen after that.
The part about series versioning should be addressed by June.
I agree with Andrea.
We just spread the word about feedback and are waiting, but we can't wait in order to have progress. I would try to keep people working and keep an eye on feedback.
AndreaPerego: If we decide that we try to finalize version 3, probably we should also take steps to validate what we have done so far.
We should collect feedback about implementations.
PWinstanley: what we were doing with version 2 was finding situations where people using application profiles used what we were doing in the new version. This kind of thing will be more challenging with DCAT.
mackx's point centers around a number of people who use CKAN. What connection do we have with Open Knowledge Foundation so that we can ascertain their appetite for updating CKAN.
At one time we had connections with people building their own plugins for metadata.
AndreaPerego: to my knowledge, CKAN has an official extension for DCAT. I'm not sure how recently it's been updated.
… It's been updated but based on DCAT-AP 1. I don't know to what extent they support DCAT2.
… I also have the impression that people are using other types of applications for catalogs besides the ones like CKAN that have been around for quite a while. People are looking for more lightweight solutions.
In other domains there are still a lotof users, like geospatial.
What they'd be doing depend on the requests they get form users. People might be not using series because it's not supported, though.
Figshare assigns different versions in an approach that's similar to what we have. They aren't implementing it in DCAT.
Knowing that we have to get feedback on certain features would be helpful.
We may discover that there is some interest in doing the job.
PWinstanley: We should be having discussions with people like Xenodo. Just as Nobu_OGURA_DSA is moving DSA to version 2, Simpson Garfinkel and others associated with data.gov are doing the same in the U.S. That makes me think we should be making some targeted approach to Socrata.
If data.gov is using versions 3, then cities and what not will be looking for the easiest root to do the export in the same sort of way Mackx was describing for people using CKAN.
kcoyle joins!
riccardoAlbertoni: About the charter, we cannot discuss the chartering now, but we would like to have a working version so that we can analyze what we have and propose changes.
PWinstanley: I think there will need to be discussions with other W3C member organizations.
… That's something that I'll take forward with Philllippe and Caroline.
PWinstanley: thanks all! we'll have another meeting in a couple of weeks, and the DCAT meeting in between.