IRC log of ldp on 2013-05-06

Timestamps are in UTC.

13:57:16 [RRSAgent]
RRSAgent has joined #ldp
13:57:16 [RRSAgent]
logging to http://www.w3.org/2013/05/06-ldp-irc
13:57:18 [trackbot]
RRSAgent, make logs public
13:57:18 [Zakim]
Zakim has joined #ldp
13:57:20 [trackbot]
Zakim, this will be LDP
13:57:20 [Zakim]
ok, trackbot; I see SW_LDP()10:00AM scheduled to start in 3 minutes
13:57:21 [trackbot]
Meeting: Linked Data Platform (LDP) Working Group Teleconference
13:57:21 [trackbot]
Date: 06 May 2013
13:57:22 [pchampin]
pchampin has joined #ldp
13:58:04 [SteveS]
SteveS has joined #ldp
13:58:13 [Zakim]
SW_LDP()10:00AM has now started
13:58:22 [Zakim]
+ +329331aaaa
13:59:55 [JohnArwe]
JohnArwe has joined #ldp
14:00:03 [Zakim]
+JohnArwe
14:00:14 [mielvds1]
zakim, +329331aaaa is me
14:00:15 [Zakim]
+mielvds1; got it
14:00:35 [Zakim]
+Arnaud
14:00:47 [Zakim]
+OpenLink_Software
14:01:01 [TallTed]
Zakim, OpenLink_Software is temporarily me
14:01:01 [Zakim]
+TallTed; got it
14:01:02 [TallTed]
Zakim, mute me
14:01:02 [Zakim]
TallTed should now be muted
14:01:07 [Zakim]
+Ashok_Malhotra
14:01:10 [Zakim]
+[IBM]
14:01:15 [pchampin_]
pchampin_ has joined #ldp
14:01:18 [SteveS]
zakim, [IBM] is me
14:01:18 [Zakim]
+SteveS; got it
14:02:06 [Arnaud]
zakim, who's on the phone?
14:02:06 [Zakim]
On the phone I see mielvds1, JohnArwe, Arnaud, TallTed (muted), Ashok_Malhotra, SteveS
14:02:53 [Zakim]
+Gregg_Vanderheiden
14:02:58 [Zakim]
-Gregg_Vanderheiden
14:03:05 [cody]
cody has joined #ldp
14:03:10 [Zakim]
+Sandro
14:03:48 [Zakim]
+[IPcaller]
14:03:58 [cody]
Zakim, IPcaller is me
14:03:58 [Zakim]
+cody; got it
14:04:02 [sergio]
sergio has joined #ldp
14:04:21 [Zakim]
+[GVoice]
14:05:53 [TallTed]
Zakim, unmute me
14:05:53 [Zakim]
TallTed should no longer be muted
14:06:23 [TallTed]
Zakim, mute me
14:06:23 [Zakim]
TallTed should now be muted
14:06:28 [SteveS]
Scribe: SteveS
14:06:46 [SteveS]
Topic: Approve minutes from April 29th
14:06:57 [Zakim]
+[IPcaller]
14:07:10 [SteveS]
RESOLVED: approve minutes from April 29th
14:07:15 [Zakim]
+??P5
14:07:33 [sergio]
Zakim, [IPcaller] is me
14:07:34 [Zakim]
+sergio; got it
14:08:01 [JohnArwe]
zakim, who is here?
14:08:01 [Zakim]
On the phone I see mielvds1, JohnArwe, Arnaud, TallTed (muted), Ashok_Malhotra, SteveS, Sandro, cody, ericP, sergio, ??P5
14:08:04 [Zakim]
On IRC I see sergio, cody, pchampin_, JohnArwe, SteveS, Zakim, RRSAgent, Arnaud, mielvds1, Ashok, svillata, betehess, nmihindu, TallTed, davidwood, bblfish, Yves, sandro, trackbot,
14:08:04 [Zakim]
... cygri, jmvanel, ericP, thschee
14:08:50 [SteveS]
Topic: next meeting May 13th
14:09:23 [SteveS]
ericP: who will be in Brazil WWW2013?
14:09:46 [mielvds1]
I will, good to meet some people from the group
14:09:46 [nmihindu]
Zakim, ??P5 is me
14:09:46 [Zakim]
+nmihindu; got it
14:09:51 [SteveS]
Arnaud: will be there and may have a conflict, checking timezone clashes
14:10:31 [pchampin]
pchampin has joined #ldp
14:11:13 [SteveS]
SteveS: offered to chair if Arnaud can not attend
14:12:30 [SteveS]
Arnaud: fear of missing another meeting as we are not seeing much activity between the calls, needing to hit last call by end of June
14:12:35 [Zakim]
+??P16
14:12:44 [pchampin]
zakim, ??P16 is me
14:12:44 [Zakim]
+pchampin; got it
14:13:12 [SteveS]
…will defer decision on holding meeting on May 13th later this week
14:13:23 [SteveS]
Topic: Issues and Actions
14:13:33 [SteveS]
Arnaud: no actions pending review
14:14:37 [SteveS]
…any actions anyone want to claim progress?
14:15:25 [SteveS]
SteveS: made progress on ACTION-45 but want to do another read/rewrite on it
14:16:08 [SteveS]
Arnaud: Update on test suite? See that Raul has made progress on it
14:16:29 [SteveS]
ericP: is under it and making progress
14:16:31 [sergio]
q+
14:16:53 [Arnaud]
ack sergio
14:17:11 [SteveS]
sergio: Spoke with rgarcia and said he is making progress with ericP
14:17:13 [nmihindu]
q
14:17:28 [JohnArwe]
my email (and the corresponding agenda topic) is evidence of progress on action-51, but not done yet
14:17:33 [TallTed]
Zakim, who's noisy?
14:17:44 [Zakim]
TallTed, listening for 10 seconds I heard sound from the following: nmihindu (61%), mielvds1 (9%), JohnArwe (4%), Arnaud (33%)
14:18:12 [SteveS]
Topic: Issues pending review
14:18:41 [SteveS]
Arnaud: 1 new raised issue, Roger asked that we wait to discuss until he is available to present
14:18:51 [SteveS]
Topic: Open Issues
14:19:24 [SteveS]
http://www.w3.org/2012/ldp/wiki/Meetings:Telecon2013.05.06#Open_Issues
14:19:33 [SteveS]
Topic: ISSUE-14
14:19:39 [sandro]
issue-14
14:19:39 [trackbot]
ISSUE-14 -- Include clarifications about ordering in BPC representations -- open
14:19:39 [trackbot]
http://www.w3.org/2012/ldp/track/issues/14
14:19:57 [TallTed]
Zakim, unmute me
14:19:57 [Zakim]
TallTed should no longer be muted
14:20:29 [SteveS]
Arnaud: No new proposals available from the one proposed by Raul
14:21:37 [SteveS]
TallTed: subsequent discussions has shown that others voice the concern, need to make it clear the sorting is arbitrary by the server and the server may provide the sorted by data in the response
14:21:56 [SteveS]
ericP: what do you expect the client to do with sorting info?
14:22:07 [SteveS]
TallTed: would be dealing with paging
14:22:07 [Ashok]
q+
14:22:31 [Arnaud]
ack ashok
14:22:38 [SteveS]
ericP: a client may be able to do a binary search if it knew the sorting and paging scheme
14:23:12 [SteveS]
Ashok: concern if get 1 container it may come in different orders, no requirement that it has to be consistent
14:23:24 [pchampin]
q+
14:23:56 [SteveS]
Arnaud: notices that ISSUE-18 talks about stable paging and dealing with changing pages over time
14:24:02 [Arnaud]
ack pchampin
14:24:09 [JohnArwe]
pierre breaking up badly
14:24:18 [JohnArwe]
...every 2-3rd syllable coming through
14:24:42 [SteveS]
q+
14:25:18 [pchampin]
1. paging should be reproducible in some way
14:25:20 [SteveS]
Arnaud: would like to keep issues separate
14:25:26 [Arnaud]
ack steves
14:25:44 [pchampin]
2. I'm not sure the proposed "sortPredicate" is enough to represent all useful sorting schemes
14:27:02 [SteveS]
SteveS: to be clear ISSUE-14 is only about how to have both ascending and descending, be good to open new issues as needed
14:27:40 [SteveS]
TallTed: spec take into account localization, etc
14:27:57 [SteveS]
SteveS: spec defers to the definition of ascending and descending as within the SPARQL spec
14:28:12 [sergio]
q+
14:28:26 [Arnaud]
ack sergio
14:28:28 [pchampin]
it does not cover ascending vs. descending, it does not cover sorting agains multiple predicates
14:28:33 [SteveS]
SteveS: based on the use cases that Raul and I have put forward, the proposal is enough
14:28:50 [Ashok]
q+
14:28:54 [pchampin]
use case: sorting bugs by state, then by date
14:30:33 [Arnaud]
ack ashok
14:31:08 [SteveS]
sergio: current libraries won't be able to handle this natively as it is not based on List (at least initially)
14:31:19 [ericP]
-> http://www.w3.org/TR/sparql11-query/#modOrderBy SPARQL Order
14:31:21 [pchampin]
oops; realizing this is a *list* of predicates... so my use case above is actually handled. sorry :-(
14:32:01 [ericP]
-> http://www.w3.org/TR/xpath-functions/#func-compare XPath compare
14:32:12 [ericP]
op:numeric-less-than(A, B)
14:32:20 [SteveS]
Ashok: what are the limits of what can be sorted?
14:32:35 [SteveS]
SteveS: it is defined by SPARQL orderBy
14:33:16 [sergio]
+1 to sparql-based ordering definition
14:33:49 [SteveS]
Ashok: why are we ignoring collations?
14:35:32 [SteveS]
ericP: <added some description>
14:36:04 [SteveS]
What scenario are we saying we haven't considered to date?
14:37:51 [SteveS]
Arnaud: sounds like a separate issue that perhaps Ashok can raise and propose something regards to what he sees as missing from the spec
14:38:19 [JohnArwe]
From email http://lists.w3.org/Archives/Public/public-ldp-wg/2013Apr/0002.html
14:38:27 [JohnArwe]
The concrete proposal is the following:
14:38:27 [JohnArwe]
An LDPC server can indicate to a client the ordering of members in a
14:38:28 [JohnArwe]
container page using an ldp:containerOrder property. This property has
14:38:28 [JohnArwe]
as range a list of resources with two properties:
14:38:28 [JohnArwe]
.- ldp:containerSortPredicate, which defines the property used for sorting
14:38:28 [JohnArwe]
.- ldp:containerSortOrder, which defines the ordering (ascending or
14:38:28 [JohnArwe]
descending) and is optional
14:38:36 [JohnArwe]
[over]
14:39:03 [SteveS]
PROPOSAL: Close ISSUE-14: Include clarifications about ordering in BPC representations, per Raúl's suggestion http://lists.w3.org/Archives/Public/public-ldp-wg/2013Apr/0002.html
14:39:29 [pchampin]
would adding a 3rd property ldp:containerSortCollation make everyone happy?
14:39:40 [pchampin]
(with a sensible default value for the 3rd one?)
14:39:43 [ericP]
s/<added some description>/SPARQL ORDER BY uses the 2 parameter fn:compare(A,B) function which takes no collation parameter. this means that e.g. "ö" and "oe" don't order next to each other in german.
14:40:18 [Arnaud]
PROPOSAL: close issue-14, adding ldp:ccontainterSortOrder to allow specifying the sorting order: ldp:descending or ldp:ascending
14:41:32 [sergio]
+1
14:41:48 [TallTed]
+1
14:44:23 [JohnArwe]
...realizing Arnaud's did not capture ALL the changes in Raul's
14:44:27 [JohnArwe]
...re-drafting
14:44:34 [ericP]
-> http://www.w3.org/TR/sparql11-query/#OperatorMapping ¶3 SPARQL discussion of collations
14:44:46 [TallTed]
PROPOSAL: An LDPC server can indicate to a client the ordering of members in a container page using an ldp:containerOrder property. This property has as range a list of resources with two properties: (a) ldp:containerSortPredicate, which defines the property used for sorting; (b) ldp:containerSortOrder, which defines the ordering (ascending or descending) and is optional, (c, etc.) collation, and others
14:45:25 [pchampin]
+1
14:45:27 [SteveS]
+1
14:45:30 [JohnArwe]
+1
14:45:33 [sergio]
+1
14:45:38 [TallTed]
+1
14:45:47 [sandro]
+1
14:45:59 [nmihindu]
+1
14:46:08 [cody]
+1
14:46:21 [SteveS]
RESOLVED: Close ISSUE-14, saying an LDPC server can indicate to a client the ordering of members in a container page using an ldp:containerOrder property. This property has as range a list of resources with two properties: (a) ldp:containerSortPredicate, which defines the property used for sorting; (b) ldp:containerSortOrder, which defines the ordering (ascending or descending) and is optional, (c, etc.) collation, and others
14:46:59 [SteveS]
Topic: ISSUE-32
14:47:03 [SteveS]
ISSUE-32 ?
14:47:03 [trackbot]
ISSUE-32 -- How can clients discover that a resource is an LDPR or LDPC, and what features are supported? -- open
14:47:03 [trackbot]
http://www.w3.org/2012/ldp/track/issues/32
14:47:34 [SteveS]
Arnaud: JohnArwe has sent out an email regarding points http://lists.w3.org/Archives/Public/public-ldp-wg/2013Apr/0139.html
14:48:16 [TallTed]
http://lists.w3.org/Archives/Public/public-ldp-wg/2013Apr/att-0139/W3CIssue32.pdf
14:48:24 [betehess]
betehess has joined #ldp
14:48:54 [SteveS]
JohnArwe: illustrated in PDF, page 1 is a straw man of tasks of what clients are trying to do
14:49:35 [SteveS]
…the wiki was a fairly large list, it shows how a large number of tests may need to be done
14:50:10 [SteveS]
…consider going down "profile" route, can do high-level introspection, and define common grouping that would be common
14:51:03 [SteveS]
…3 profiles: "read-only", "strictly managed membership" and "everything"
14:51:19 [SteveS]
…looking for WG feedback on the the right grouping and right things being required
14:51:19 [pchampin]
in option 2 (managed/closed container), how can you create new ressources if containers are closed?
14:52:12 [betehess]
betehess has joined #ldp
14:52:30 [SteveS]
pchampin, what do you call option 2? This a grouping of tasks/capabilities (aka profiles) not options
14:52:32 [JohnArwe]
pierre: out of band means
14:53:36 [JohnArwe]
...actually pierre, it you want to create a new Member that is standard POST to the collection
14:53:37 [SteveS]
Arnaud: soliciting more WG feedback, doesn't current include how the affordances are communicated but looking for direction
14:54:29 [mielvds1]
if capabilities need to be discovered anyway, why profiles?
14:54:57 [SteveS]
JohnArwe: wonder if people wonder if we should do profiles at all? be good to get a feeling from the group with a straw poll
14:55:01 [pchampin]
+1 to profiles
14:55:24 [SteveS]
I see no issues with profile construct
14:56:08 [SteveS]
ericP: anyone done anything to see how many of these things are orthogonal, a checkbox list of items…it would be a good way for people to parameterize their libraries and services
14:56:41 [nmihindu]
+1 for profiles, makes sense for being compliant only offering only set of features
14:57:14 [SteveS]
JohnArwe: as example, HTTP gives a way to introspect a URI on what verbs are needed
14:57:22 [TallTed]
+1 profiles seem like a good approach. *server* profile is a starting point, *resource* profiles might differ -- but couldn't be broader than their hosting server...
14:57:39 [sergio]
+1 to explore such idea
14:57:41 [SteveS]
ericP: thinking it would be good to know what parameters you might need for reading, adding members, etc
14:57:42 [JohnArwe]
mielvds1, boils down to how complex is it for clients to code the introspection logic? if they need to look at 3 tasks (read only), do they need to code 1 piece of logic or 3?
14:58:04 [SteveS]
Topic: ISSUE-58 and options on the table
14:58:07 [ericP]
ランドMあっcえっs
14:58:09 [SteveS]
ISSUE-58?
14:58:09 [trackbot]
ISSUE-58 -- Property for asserting that complete description of members is included in LDPC representation -- open
14:58:09 [trackbot]
http://www.w3.org/2012/ldp/track/issues/58
14:58:39 [SteveS]
Arnaud: Options on the table http://lists.w3.org/Archives/Public/public-ldp-wg/2013Apr/0141.html
14:58:57 [SteveS]
…would like to get a straw poll on the 4 options
14:59:14 [Arnaud]
STRAWPOLL: which option should we go with?
15:00:40 [SteveS]
Arnaud: vote -1..+1 in order of A, B, C, D
15:00:43 [cody]
A -1, B +1, C +1, D -1
15:01:20 [sergio]
A -1, B +1, C -1, D +0 (I'm not very fan of boolean properties in rdf)
15:01:20 [TallTed]
A +1, B +1, C -0.5, D +1
15:01:22 [ericP]
A +1, B 0, C 0, D -1
15:01:41 [pchampin]
the HTTP header is probably ill-named (and yes, I did propose the name)
15:01:55 [JohnArwe]
+0.5, +0.5, +1.0, -0.5 Where the option D issue is I don't see how to make the caching work, but if I got convinced it would work then fine with it
15:02:08 [nmihindu]
-1, 0, +0, +1
15:02:13 [mielvds1]
A −1, B +1, C −1, D +1
15:02:34 [pchampin]
@JohnArwe: option D is motivated by the fact that this is a property of the representation, not of the resource/container per se
15:03:10 [SteveS]
+.5, +.4, +1 (but would like the value to be etag), +0
15:03:18 [JohnArwe]
pierre: agree and like that part; email response articulates the problems I see
15:03:25 [pchampin]
container X may inline member Y completely in *some* representations only
15:03:40 [pchampin]
sorry, didn't read the thread yet
15:03:44 [pchampin]
will right away :)
15:03:52 [JohnArwe]
np
15:04:09 [Zakim]
-Ashok_Malhotra
15:04:34 [Zakim]
-Sandro
15:04:40 [Zakim]
-JohnArwe
15:04:42 [Zakim]
-SteveS
15:04:42 [Zakim]
-nmihindu
15:04:44 [cody]
cody has left #ldp
15:04:50 [Zakim]
-cody
15:05:07 [mielvds1]
mielvds1 has left #ldp
15:05:17 [Zakim]
-mielvds1
15:05:24 [Zakim]
-Arnaud
15:09:37 [Zakim]
-sergio
15:09:41 [sergio]
sergio has left #ldp
15:12:45 [Zakim]
-TallTed
15:12:53 [Zakim]
-pchampin
15:12:54 [Zakim]
-ericP
15:12:54 [Zakim]
SW_LDP()10:00AM has ended
15:12:54 [Zakim]
Attendees were JohnArwe, mielvds1, Arnaud, TallTed, Ashok_Malhotra, SteveS, Gregg_Vanderheiden, Sandro, cody, ericP, sergio, nmihindu, pchampin
15:31:43 [bhyland]
bhyland has joined #ldp
15:38:10 [betehess]
betehess has joined #ldp
16:33:10 [Zakim]
Zakim has left #ldp
17:26:21 [jmvanel]
jmvanel has joined #ldp