Meeting minutes
Scribe: Gschwend, Adrian
<pfps> minutes look fine
Approval of last week's minutes: 1
ora: any comments on the minutes?
<ora> proposal: Approve last week's minutes
<pfps> +1
+1
<ora> +1
<Dominik_T> +1
<gkellogg> +1
<pchampin> +1
<Timothe> +1
<Tpt> +1
<rubensworks> +1
<TallTed> +1
RESOLUTION: Approve last week's minutes
Impressions from LDBC Technical User Community meeting in Seattle (Ora)
ora: Just for general information, I attended the SIGMOD conference, there was LDBC community meeting. I was there with some other folks, Olaf was there.
… Discussions were mostly around PGs. Fair amount of discussion about PG schema, paper won the best paper award at SIGMOD.
… List of authors reads like the who is who of PGs. I want to mention it because it's a small, elegant schema language.
… I gave a talk about Onegraph project. Many people came to me and asked about RDF Star.
… Some also asked why are you doing that, nobody is using RDF. I have reasonably strong arguments to counter that.
Ora: Dominik_T what did you think of it?
<TallTed> link to the winning paper?
Dominik_T: interesting, many good papers.
… I'm one of the authors of PG schema.
<ora> https://
<Dominik_T> https://
ktk: the word 'schema' means different things for different people
ktk: How does that PG schema compare to the definitions of what RDF people understand under the word "schema".
… how does this paper compare to JSON-Schema or RDF-Schemas
Dominik_T: The paper shows at the end how it compares to other schemas, including JSON schema.
olaf: it's like a database schema language, where you can define nodes, address, properties, constraints, etc.
Dominik_T: I would like to add that in the paper we do not discuss properties. Nodes, address and constraints are there.
ora: I find it a nice design.
Dominik_T: yes it took four years of discussions to get there.
Review of open actions, available at 2
pchampin: Action #67, permissions on the use-case wiki. Should now be solved.
<ghurlbot> Action 67 ensure that anyone in the WG can edit the UC wiki / make PR (on pchampin) due 29 Jun 2023
pchampin: Wiki is now public, anybody with a GH account can contribute. Pull-requests on the wiki are not supported by GH but you can make one on the UC repo.
… I propose to close it.
… #62 is about SPARQL 1.2 CWG. It turns out that it is possible to completely rename a CWG.
<ghurlbot> Action 62 check with sys team to have a redirect from community/sparql-next to community/sparql-12 (on pchampin) due 1 Jun 2023
pchampin: I discussed with W3C, they don't do it often but it's technically possible. As long as we or the CWG decides on a new name, I can forward it to the right person.
… The mailing list has to be closed & a new one to be opened. The people would be migrated but people have to go from one archive to the other.
… Everything would be consistent in the end.
pfps: All 3 of my actions are done.
ktk: I don't see a problem with the ML archive for SPARQL CWG as most of the action happens in the issues.
AndyS: we can ask the chairs (AndyS & Jerven Bolleman)
… it's good that links to old messages stay stable.
Review of pull requests, available at 3
ora: Pull requests?
gkellogg: regarding short names, we need to be sure we can get back to the old specs. We might have to do some aliases to make sure that works properly.
ora: I was browsing some RDF specs and it did jump to the latest ones, I could not get to the old ones.
gkellogg: RDF Schema is an issue for example.
ora: where are we with the mobile phone issue.
pchampin: I can report on it. I contacted someone of the reference people for CSS in the W3C team. His advice is to avoid reducing the font size. Reducing this or aggressive word wrapping should be last resort. He has some alternatives, I made a draft pull request with it.
<Dominik_T> w3c/
pchampin: pfps commented on that, it's not optimal. One of the things needed is to set the width as a fixed one. We cannot detect it in the browser.
… YMMW how it looks like in other peoples browser for the moment.
<pchampin> https://
ora: I like the idea of stacking rows.
<pchampin> example visible at https://
pfps: Update on a preamble. Restricted view-ports create all sort of problems. Question is what is the best of the worst options. The problem I see is that CSS does not provide any tools for solving this properly from what I can see. So we cannot solve it. pchampin proposal makes it use much more space, which make the result much worse.
<Zakim> TallTed, you wanted to assert that "rotating one's phone" should be acceptably different from "scrolling horizontally"
TallTed: I agree with pfps take here. I don't understand why it is that important to make this document work on tiny screens.
… We did what is reasonable.
… when you add screenshots to the issue please put them side to side so they are comparable.
… but I don't think people will read these documents on very small screens.
… Reading them on a reader or tablet is ok with these documentss.
Dominik_T: I would like to disagree, I think it is important to make it ready for mobile phones. I tested on many models & screen sizes and it is unreadable on all I tested.
<pfps> I would like to see screen shots of the unreadable situations.
TallTed: I can read this document on my iphone if I turn it sideways. It is not great but ok.
<pchampin> FTR, I agree that we are probably spending too much time on this -- and I am personally happy with how the spec renders on my own phone (including in portait mode)
Dominik_T: It is unreadable in the table, the rest is ok.
<pfps> XML Schema datatypes has a much worse problem on small viewports
Dominik_T: I read other specs on mobile phone and there is no problem with them.
pchampin: I feel we spend too much time on this. Unless there is a quick consensus on some improvements we should not go down that rabbit hole.
… Let's take this problem the other way. Could Dominik_T create some screenshots of specific problems.
<pfps> on my phone - an old middle-of-the-road phone - the document is readable in both portrait and landscape mode
ora: I worked 17 years for a large mobile phone company and I find it hard reading such documents on mobile phones. But I understand that a large population reads documents on the web exclusively on mobile phone.
… maybe we can have a look at specific problems and we can try to address them. But we should at the same time not go down that rabbit hole.
… I see myself trying to get into this. I suggest we move on to other things now.
<pfps> as far as I can tell the biggest problem is that there is a very long identifier
<TallTed> I imagine a new CG, if there isn't such already, focused on adjusting the W3 CSS for mobile optimization might be worthwhile
pchampin: stop
<gtw> oops
you killed the zoom
we all did
gkellogg: the i18n reviews w3c/rdf-concepts#48, they will come back on it.
<gkellogg> w3c/
<TallTed> i18n == internationalization
<AndyS> s/IETN/i18n/
<gkellogg> w3c/
ora: do we wait until they discussed it?
gkellogg: yes
ora: what about the editorial ones?
gkellogg: concepts, PR 50 can be merged now
Dominik_T: PR 97 is also ready
rubensworks: We can also merge the CSV/TSV one
Tpt: PR 85 can also be closed
… SPARQL query
… and maybe also SPARQL query 93
olaf: I pushed two pull requests today, to SPARQL-query. They do not show up in the dashboard.
<olaf> w3c/
olaf: It is 109
<gkellogg> The build error is "Error: Bad credentials"
olaf: This automatic checking was not successfull. Is there anything I could change?
gkellogg: I noticed this earlier. If you look on the action tab, there is an action "Add PR to project". It stales due to "bad credentials". pchampin needs to look into it.
AndyS: it's not your credentials, it's the one that grants W3C update permissions.
gkellogg: we need to look at the repository and look where these "stale actions" are so it will properly run again.
AndyS: did you do them at the same time?
olaf: no same day but hours differences
gkellogg: not a pull request but something in issues.
Any Other Business (AOB), time permitting
gkellogg: I want to point out that there are 4 issues that are in "proposed closing" for some time. I don't think we need to discuss them.
… I don't think I'm the one to close them but I think all the work is done.
ora: If all the work is done you have every bit of authority to close them.
gkellogg: ok I will do so.
gkellogg: We could say if something is marked closing & no one does it it can simply be done.
ora: We trust each other and can do that.
AndyS: About tests, somebody submitted small tests for the RDF Star CWG.
AndyS: I would at least like to acknowledge that work.
ora: do we want to do that? Is the CG still functioning? What is the rationale?
AndyS: I don't see why we would want to drop this. We have tests in the CG and we don't have in the WG.
gkellogg: there is a CG that works on test. The right thing to do is to adopt the test repository and come up with a mechanism to add RDF 1.2 tests.
gkellogg: there is a repository that we inherit that carries all the tests.
… it makes sense to take this over at some point. We might want discuss how we do that.
ora: I propose to make this an agenda item for the next week.
AndyS: should I go ahead and merge this contribution?
ora: I am not against merging this.
… so people understand we appreciate their contribution.
ora: Let us discuss this test repository issue in two weeks.
<gkellogg> regrets for next week.