IRC log of ldp on 2013-10-28

Timestamps are in UTC.

13:58:56 [RRSAgent]

RRSAgent has joined #ldp

13:58:56 [RRSAgent]

logging to http://www.w3.org/2013/10/28-ldp-irc

13:58:58 [trackbot]

RRSAgent, make logs public

13:58:58 [Zakim]

Zakim has joined #ldp

13:59:00 [trackbot]

Zakim, this will be LDP

13:59:00 [Zakim]

ok, trackbot; I see SW_LDP()10:00AM scheduled to start in 1 minute

13:59:01 [trackbot]

Meeting: Linked Data Platform (LDP) Working Group Teleconference

13:59:01 [trackbot]

Date: 28 October 2013

14:00:31 [Zakim]

SW_LDP()10:00AM has now started

14:00:42 [Zakim]

+[IPcaller]

14:00:54 [Zakim]

+Arnaud

14:01:01 [codyburleson]

ZAKIM, IPcaller IS ME

14:01:02 [Zakim]

+codyburleson; got it

14:01:13 [Zakim]

+Ashok_Malhotra

14:02:09 [JohnArwe]

JohnArwe has joined #ldp

14:02:17 [Zakim]

+JohnArwe

14:02:54 [Zakim]

+??P19

14:03:35 [stevebattle6]

zakim, ??P19 is me

14:03:35 [Zakim]

+stevebattle6; got it

14:03:40 [Zakim]

+Sandro

14:04:39 [Zakim]

+OpenLink_Software

14:04:49 [TallTed]

Zakim, OpenLink_Software is temporarily me

14:04:49 [Zakim]

+TallTed; got it

14:05:11 [Zakim]

+[IBM]

14:05:20 [SteveS]

zakim, [IBM] is me

14:05:20 [Zakim]

+SteveS; got it

14:05:58 [stevebattle6]

topic approval of minutes

14:06:10 [stevebattle6]

topic: approval of minutes

14:06:54 [stevebattle6]

arnaud: minutes approved

14:07:08 [stevebattle6]

arnaud: meeting same time next week

14:07:25 [stevebattle6]

topic: action & issues

14:08:25 [stevebattle6]

JohnArwe: I asked EricP to look at paging.

14:09:19 [stevebattle6]

JohnArwe: We were looking at first/next - re ambuguity in proposal

14:10:15 [stevebattle6]

Arnaud: The proposal was to move from RDF to link headers.

14:10:29 [stevebattle6]

Arnaud: I'm happy to close this

14:11:51 [stevebattle6]

JohnArwe: Some remaining editorial issues (in red).

14:12:10 [stevebattle6]

Arnaud: People - please take a look at paging.

14:12:31 [JohnArwe]

also note (in paging) that the 209 mockup is still in here, since that is still open

14:12:34 [stevebattle6]

Arnaud: I'm suggesting that we close all remaining actions.

14:12:46 [stevebattle6]

Arnaud: No objections.

14:13:14 [stevebattle6]

JohnArwe: We've caught up on the normative issues.

14:14:04 [stevebattle6]

Arnaud: There are actions relating to other documents that are open:- e.g. primer

14:14:28 [stevebattle6]

topic: confusing predicate names

14:15:18 [stevebattle6]

Arnaud: This week the plan was to choose the most popular polled solution.

14:15:21 [TallTed]

q+

14:15:34 [Arnaud]

ack TallTed

14:15:47 [stevebattle6]

TallTed: My poll input didn't save, this changes the rankings.

14:17:09 [stevebattle6]

Arnaud: Now JohnArwe's proposal is in first place. I'm confused about use of case in this.

14:17:26 [stevebattle6]

s/confused/concerned/

14:18:46 [stevebattle6]

Arnaud: We could combine proposals

14:19:21 [stevebattle6]

Sandro: Standard practice is to use 'has' instead of 'link' that we now have.

14:19:39 [stevebattle6]

Arnaud: We should make a decision about using John's proposal as a basis.

14:20:26 [Arnaud]

Proposal: regarding ISSUE-81 Part I: Confusing predicate names, accept John's renaming proposal as the basis moving forward

14:20:31 [JohnArwe]

I don't think we have 'link' today Sandro; that was me suggesting off that cuff that we could use ldp:containerLink instead of ldp:container (lower case c), but I ldp:hasContainer certainly works fine for me

14:20:54 [TallTed]

+1

14:20:54 [JohnArwe]

+1

14:20:57 [SteveS]

+1

14:20:57 [stevebattle6]

+1

14:21:31 [Ashok]

+1

14:21:40 [stevebattle6]

Arnaud: We could use containerID from Ted's proposal?

14:21:51 [Arnaud]

Resolved: regarding ISSUE-81 Part I: Confusing predicate names, accept John's renaming proposal as the basis moving forward

14:21:56 [stevebattle6]

Sandro: Link isn't recommended either.

14:22:04 [stevebattle6]

s/Link/ID/

14:22:28 [stevebattle6]

Arnaud: For next week: propose modifications to John's proposal.

14:22:59 [stevebattle6]

JohnArwe: I'm happy to propose 'hasContainer' right now.

14:23:41 [Arnaud]

Proposal: amend John's naming scheme by changing ldp:container to ldp:hasContainer

14:23:59 [stevebattle6]

Arnaud: The proposal is moot

14:24:17 [stevebattle6]

Arnaud: Other proposals are welcomed

14:24:22 [Ashok]

+1 to hasContainer

14:24:42 [stevebattle6]

(oh we are voting!)

14:24:45 [TallTed]

-.5 hasContainer

14:24:53 [JohnArwe]

+0

14:24:55 [stevebattle6]

0

14:25:04 [SteveS]

+0 think we can do better

14:27:15 [stevebattle6]

Arnaud: There were two points, the second is to address the predicate names. The aim is to limit the number of things you have to specify.

14:27:50 [stevebattle6]

Arnaud: Henry raised the point that properties may be grouped under a blank node.

14:28:15 [stevebattle6]

Arnaud: Lets address this next week with Henry on th ecall.

14:28:31 [stevebattle6]

topic : Paging

14:30:23 [stevebattle6]

JohnArwe: As soon as you have >1 resource the notion of completeness breaks down.

14:30:49 [stevebattle6]

JohnArwe: The client never knows the complete state.

14:32:05 [Ashok]

q+

14:32:33 [stevebattle6]

Arnaud: The current status - if the client fetches a resource and the server initiated paging, the client has to do another get to get the FIRST page.

14:33:08 [stevebattle6]

Arnaud: This introduces an extra fetch. Avoid the round trip by creating a new status code?

14:33:34 [stevebattle6]

Arnaud: Why not just return 200 with the first page - as John describes.

14:34:00 [SteveS]

SteveS has joined #ldp

14:34:12 [stevebattle6]

Arnaud: It was previously felt that 200 was misleading because the client doesn't get exactly the resource they were expecting.

14:34:43 [stevebattle6]

Sandro: If the client thinks they have the full resource how can they use PUT to update it?

14:35:04 [stevebattle6]

Sandro: Even on a small resource this is a problem.

14:35:17 [JohnArwe]

Erik also said (although this is not directly relevant to paging, it is relevant to TimBL's other use for 209) that the 209 MIGHT make sense for the general Linked Data 303 (non-info resource).

14:35:34 [stevebattle6]

Arnaud: In the current spec the redirect tells you this is is the first page.

14:36:10 [stevebattle6]

Arnaud: If we have the 200 with the first page you can't rely on the status page, you have to check the link header to know this is not the full resource.

14:36:58 [stevebattle6]

Arnaud: This would mean we don't have to create a new status code.

14:37:09 [stevebattle6]

JohnArwe: and we don't need the redirect.

14:37:35 [stevebattle6]

SteveS: And timbl will be convinced? He wasn't satisfied with the 200 status code.

14:38:09 [stevebattle6]

JphnArwe: This is a general linked data problem.

14:39:19 [Arnaud]

ack Ashok

14:39:24 [stevebattle6]

JohnArwe: Problem with the 'location' code (to inform that this is 1st page) is that we're redefining it.

14:39:36 [JohnArwe]

s/JphnArwe/JohnArwe/

14:40:08 [stevebattle6]

ashok: The problem is clients with limited capability (eg. mobile clients). Do we get 200 in both cases?

14:40:58 [stevebattle6]

s/redefining/potentially redefining/

14:41:52 [stevebattle6]

Sandro: If it's just a browser its not going to do anything different. We're interested in automated agents that need more information.

14:42:27 [stevebattle6]

Arnaud: Checking is relatively cheap, so this shouldn't be a problem.

14:43:06 [stevebattle6]

Arnaud: Please think about this issue and if you can live with your conscience.

14:43:19 [JohnArwe]

q+ to ask about issue-81 part 2

14:43:31 [stevebattle6]

topic: put/create

14:44:04 [stevebattle6]

Arnaud: Can clients CREATE resources with PUT.

14:44:32 [stevebattle6]

Arnaud: timbl wanted to have this capability.

14:45:01 [stevebattle6]

Arnaud: Using the filesystem analogy

14:45:05 [JohnArwe]

TimBL wanted something very specific, I re-read his comments: writable LDPRs (no collections reqd), and yes the name hint would be within some URI space owned by the server.

14:45:49 [stevebattle6]

JohnArwe: He doesn't need collections.

14:46:11 [stevebattle6]

JohnArwe: You must be able to PUT/UPDATE a container

14:46:35 [stevebattle6]

JohnArwe: If you don't need collections its a lot less problematic.

14:47:36 [stevebattle6]

SteveS: We could spend a lot of time to get this right.

14:48:48 [stevebattle6]

JohnArwe: We could add a URI template and a link to create a space of implementations

14:49:41 [stevebattle6]

Arnaud: Even if we had a hook for URI templates, the server needs to guarantee that it'll honour the request.

14:50:01 [stevebattle6]

Sandro: Where would that link come from?

14:51:24 [stevebattle6]

Arnaud: I think Tim has a use case in mind where he doesn't use a container at all.

14:51:50 [stevebattle6]

SteveS: timbl mentioned a workspace ontology he was working on.

14:52:29 [stevebattle6]

Sandro: What other inputs do we have on this?

14:53:42 [stevebattle6]

Arnaud: This may have to be more fleshed out before we add it.

14:54:20 [stevebattle6]

JohnArwe: Henry has argued for this (or something like it) in the past - but using containers.

14:54:58 [JohnArwe]

q?

14:55:10 [Arnaud]

ack JohnArwe

14:55:10 [Zakim]

JohnArwe, you wanted to ask about issue-81 part 2

14:55:24 [stevebattle6]

JohnArwe: We didn't get to the bottom of issue-81

14:56:11 [stevebattle6]

Arnaud: Postpone this until next week.

14:56:32 [stevebattle6]

topic: resolving comments

14:56:40 [stevebattle6]

Arnaud: Are we on track?

14:57:17 [stevebattle6]

Arnaud: Is it true that the resolution system sends the message for you? (Yes)

14:58:21 [Zakim]

+EricP

14:59:00 [stevebattle6]

Arnaud: The number of outstanding comments has gone down.

14:59:56 [stevebattle6]

SteveS: I completed some last week (the system may/may not have sent the responses).

15:01:07 [stevebattle6]

Arnaud: Make sure that the status is reflected in the tracker.

15:01:16 [Zakim]

-Ashok_Malhotra

15:01:17 [stevebattle6]

Arnaud: meeting adjourned.

15:01:29 [Zakim]

-SteveS

15:01:31 [stevebattle6]

scribe: stevebattle6

15:01:32 [Zakim]

-Sandro

15:01:32 [Zakim]

-EricP

15:01:33 [Zakim]

-TallTed

15:01:35 [Zakim]

-JohnArwe

15:01:37 [Zakim]

-Arnaud

15:01:46 [Zakim]

-stevebattle6

15:02:03 [Zakim]

-codyburleson

15:02:04 [Zakim]

SW_LDP()10:00AM has ended

15:02:04 [Zakim]

Attendees were Arnaud, codyburleson, Ashok_Malhotra, JohnArwe, stevebattle6, Sandro, TallTed, SteveS, EricP

15:06:59 [stevebattle6]

stevebattle6 has joined #ldp

15:30:50 [codyburleson]

codyburleson has joined #ldp

17:10:25 [jmvanel]

jmvanel has joined #ldp

17:11:22 [Zakim]

Zakim has left #ldp

17:57:28 [deiu]

deiu has joined #ldp

18:33:39 [SteveS]

SteveS has joined #ldp

19:48:20 [stevebattle6]

stevebattle6 has joined #ldp

19:58:44 [jmvanel]

jmvanel has joined #ldp

21:00:29 [gavinc]

gavinc has joined #ldp