Chatlog 2012-11-19

From Linked Data Platform
Jump to: navigation, search

See original RRSAgent log or preview nicely formatted version.

Please justify/explain all edits to this page, in your "edit summary" text.

14:59:58 <RRSAgent> RRSAgent has joined #ldp
14:59:58 <RRSAgent> logging to
15:00:00 <trackbot> RRSAgent, make logs public
15:00:00 <Zakim> Zakim has joined #ldp
15:00:02 <trackbot> Zakim, this will be LDP
15:00:02 <Zakim> ok, trackbot, I see SW_LDP()10:00AM already started
15:00:03 <trackbot> Meeting: Linked Data Platform (LDP) Working Group Teleconference
15:00:03 <trackbot> Date: 19 November 2012
15:00:11 <MacTed> Zakim, who's here?
15:00:11 <Zakim> On the phone I see SteveBattle, OpenLink_Software
15:00:12 <Zakim> On IRC I see RRSAgent, AshokMalhotra, dret, MacTed, ghard, deiu, betehess, SteveS, stevebattle, bblfish, oberger, nathanr, Arnaud, trackbot, Yves, sandro, ericP
15:00:16 <Zakim> +[IBM]
15:00:18 <AshokMalhotra> zakim, code?
15:00:18 <Zakim> the conference code is 53794 (tel:+1.617.761.6200, AshokMalhotra
15:00:19 <MacTed> Zakim, OpenLink_Software is temporarily me
15:00:20 <Zakim> +MacTed; got it
15:00:21 <MacTed> Zakim, mute me
15:00:22 <Zakim> MacTed should now be muted
15:00:28 <SteveS> Zakim, [IBM] is me
15:00:29 <Zakim> +SteveS; got it
15:00:30 <Zakim> + +1.510.206.aaaa
15:00:34 <Zakim> +??P31
15:00:38 <Zakim> +??P32
15:00:42 <Zakim> +Arnaud
15:00:43 <deiu> Zakim, ??P31 is me
15:00:43 <Zakim> +deiu; got it
15:00:58 <JohnArwe> JohnArwe has joined #ldp
15:01:01 <dret> zakim, who's here?
15:01:01 <Zakim> On the phone I see SteveBattle, MacTed (muted), SteveS, +1.510.206.aaaa, deiu, ??P32, Arnaud
15:01:01 <deiu> Zakim, mute me please
15:01:03 <Zakim> On IRC I see JohnArwe, Zakim, RRSAgent, AshokMalhotra, dret, MacTed, ghard, deiu, betehess, SteveS, stevebattle, bblfish, oberger, nathanr, Arnaud, trackbot, Yves, sandro, ericP
15:01:03 <Zakim> deiu should now be muted
15:01:06 <Zakim> +JohnArwe
15:01:10 <Zakim> +Ashok_Malhotra
15:01:12 <deiu> scribe: Andrei Sambra
15:01:16 <deiu> scribenick: deiu
15:01:17 <antonis> antonis has joined #ldp
15:01:20 <dret> zakim, +1.510.206.aaaa is me
15:01:21 <Zakim> +dret; got it
15:02:23 <Zakim> +[GVoice]
15:02:33 <ericP> Zakim, [GVoice] is me
15:02:33 <Zakim> +ericP; got it
15:02:36 <cygri> cygri has joined #ldp
15:02:50 <Zakim> +??P49
15:03:00 <oberger> Zakim, ??P49 is me
15:03:00 <Zakim> +oberger; got it
15:03:06 <oberger> Zakim, mute me
15:03:06 <Zakim> oberger should now be muted
15:03:16 <deiu> I am
15:03:53 <nmihindu> nmihindu has joined #ldp
15:03:59 <Zakim> +Yves
15:03:59 <deiu> Chair: Arnaud
15:04:11 <deiu> Topic: Minutes from last week
15:04:18 <raul> raul has joined #ldp
15:04:19 <SteveS> has not read or looked yet
15:04:31 <dret> looked and they looked ok
15:04:56 <deiu> RESOLVED: Minutes of Nov 12th approved
15:05:40 <krp> krp has joined #ldp
15:05:40 <oberger> Topic: Tracking of Actions & Issues
15:05:44 <Zakim> +??P60
<deiu> Subtopic: Actions
15:05:46 <deiu> Arnaud: no actions pending review, moving to open actions
15:05:46 <deiu> ... anyone wanting to declare victory on an action?
<deiu> ... Hearing none, let's move to issues
<deiu> Subtopic: Raised issues
15:05:56 <deiu> ... We have several issues
15:06:04 <raul> zakim, ??P60 is me
15:06:04 <Zakim> +raul; got it
15:06:42 <Zakim> +??P62
15:06:45 <rogerm> rogerm has joined #ldp
15:06:56 <oberger>
15:07:07 <Zakim> +MHausenblas
15:07:11 <nmihindu> zakim, ??P62 is me
15:07:11 <Zakim> +nmihindu; got it
15:07:15 <cygri> zakim, i'm with mhausenblas
15:07:15 <Zakim> +cygri; got it
15:07:33 <Zakim> +??P61
15:07:42 <stevebattle> q+
15:07:43 <ghard> zakim, ??P61 is me
15:07:46 <Zakim> +ghard; got it
15:07:50 <Zakim> +??P64
15:07:50 <Zakim> + +44.208.573.aabb
15:08:04 <krp> zakim, ??P64 is me
15:08:05 <Zakim> +krp; got it
15:08:19 <deiu> Arnaud: the problem we had with use cases was that it seems unreasonable for people to add use cases and requirements, so we asked people to open issues as a buffer, to wait for issues to be added to the document
15:08:30 <ghard> zakim, mute me
15:08:30 <Zakim> ghard should now be muted
15:08:39 <deiu> ... we should open issue 30
15:08:51 <deiu> ISSUE30?
15:08:56 <stevebattle> q-
15:09:00 <oberger> Issue-30?
15:09:00 <trackbot> ISSUE-30 -- Hierarchical bugtracking service -- open
15:09:00 <trackbot>
15:09:13 <deiu> RESOLVED: Open ISSUE-30
15:09:26 <Zakim> +??P67
15:09:42 <deiu> Issue-36?
15:09:42 <trackbot> ISSUE-36 -- Can applications create new containers? -- raised
15:09:42 <trackbot>
15:09:49 <antonis> Zakim, ??p67 is me
15:09:49 <Zakim> +antonis; got it
15:10:11 <dret> q+
15:10:20 <Arnaud> ack dret
15:10:25 <deiu> Arnaud: people think this is an important issue
15:10:30 <AndyS> AndyS has joined #ldp
15:10:53 <deiu> ericP: this is a sub-issue of the model issue (ISSUE-37)
15:11:21 <Zakim> -krp
15:11:32 <deiu> Arnaud: you are right; should we then raise it as a different issue or close ISSUE-36 and fold into ISSUE-37?
15:11:44 <cygri> specific issues are better than generic issues
15:11:53 <Zakim> +bblfish
15:11:53 <SteveS> q+
15:12:07 <Arnaud> ack steves
15:12:08 <deiu> ack SteveS
15:12:11 <bblfish> yes, that was me
15:12:21 <deiu> SteveS: we should leave it there and not close it
15:12:27 <Zakim> +??P64
15:12:35 <deiu> ... what do we mean by application? client or server?
15:12:36 <AndyS> zakim, ??P64 is me
15:12:36 <Zakim> +AndyS; got it
15:12:40 <dret> yes, i think it talks about the protocol
15:12:40 <rogerm> who am i
15:12:56 <cygri>
15:12:56 <Zakim> +??P72
15:13:03 <dret> the server is opaque magic, as far as REST is concerned
15:13:13 <krp> zakim, ??P72 is me
15:13:13 <Zakim> +krp; got it
15:13:41 <deiu> AndyS: the application is in the client, and servers are deployed with an application framework
15:14:11 <Kalpa> Kalpa has joined #ldp
15:14:20 <rogerm> zakim, aabb is me
15:14:20 <Zakim> +rogerm; got it
15:14:34 <deiu> Arnaud: we will open ISSUE-36 and then close it later once we have addressed the bigger question of the model
15:14:43 <AndyS> ... another model is servers are neutral to the application design, and all logic is in the client
15:14:50 <dret> the REST view would be: servers create a media type (i.e., they can engage in conversation supporting this data model and associated behavior), and clients can then engage in conversations.
15:15:12 <oberger> issue-37?
15:15:12 <trackbot> ISSUE-37 -- What is the LDP data model and the LDP interaction model? -- raised
15:15:12 <trackbot>
15:15:13 <Arnaud> ok
15:15:24 <Zakim> -krp
15:15:29 <Arnaud> RESOLVED: open ISSUE-36
15:15:30 <Zakim> +Kalpa
15:15:40 <Arnaud> reopen ISSUE-36
15:15:40 <trackbot> ISSUE-36 Can applications create new containers? re-opened
15:15:49 <deiu> ericP: need clarification to say what it is that we do, what are the resources that we use, the hyperlinks to them
15:16:01 <deiu> ... now everything is implicit and not defined in one place
15:16:10 <Arnaud> s/ericP/dret/
15:16:10 <deiu> ... maybe we can have an introductory section
15:16:29 <deiu> ... these questions should be answered in that section
15:17:01 <deiu> ... that section could be structured similar to what has been described in the email 127
15:17:15 <deiu> ... we can fill it in in whatever way we see fit
15:17:30 <deiu> ... the only feedback was that we should definitely have that in the spec
15:17:49 <deiu> Arnaud: that will be a useful thing to have, giving us a general guideline
15:18:00 <stevebattle> +1
15:18:16 <dret> i'd volunteer to write such a section, if that help to drive things forward.
15:18:18 <deiu> ... general questions: who can create LDPC/R; who can delete them, etc.
15:18:22 <deiu> ... we need a general description
15:18:35 <deiu> ... no objections
15:18:36 <oberger> There has been already quite some support indeed
15:18:38 <Arnaud> RESOLVED: open ISSUE-37
15:18:46 <Arnaud> reopen ISSUE-37
15:18:46 <trackbot> ISSUE-37 What is the LDP data model and the LDP interaction model? re-opened
15:18:58 <deiu> ISSUE-38?
15:18:58 <trackbot> ISSUE-38 -- filtered representations and inlining -- raised
15:18:58 <trackbot>
15:19:03 <AndyS> Good idea to explicitly state the formal design - seems to help some people.
15:19:11 <deiu> AndyS, +1
15:19:39 <JohnArwe> zakim, who is speaking?
15:19:50 <Zakim> JohnArwe, listening for 10 seconds I could not identify any sounds
15:20:22 <deiu> [rogerm is describing the issue from his email]
15:20:25 <cygri> q+
15:20:31 <deiu> Arnaud: this is a reasonable question to ask
15:20:37 <cygri> q+ to suggest it can be subsumed under ISSUE-32
15:20:43 <Arnaud> ack cygri
15:20:43 <Zakim> cygri, you wanted to suggest it can be subsumed under ISSUE-32
15:20:45 <cygri> ISSUE-32
15:20:45 <deiu> ... but where do we draw the line?
15:20:48 <cygri> ISSUE-32?
15:20:48 <trackbot> ISSUE-32 -- How can clients discover that a resource is an LDPR or LDPC, and what features are supported? -- open
15:20:48 <trackbot>
15:21:51 <dret> regarding ISSUE-32: in AtomPub, it's up to the client to either embed the resource contents or link to it. then the server simply manages it like that, and always returns it like it was created.
15:21:52 <deiu> Richard: maybe we can update the ISSUE-32 description to mention what ISSUE-38 is trying to achieve
15:21:56 <stevebattle> I don't think this is covered under 32 - 32 is about descriptions of affordances.
15:22:13 <dret> s/ISSUE-32/ISSUE-38/
15:22:17 <oberger> adding pointers to eachother's in 32 and 38
15:22:33 <Zakim> +??P66
15:22:36 <oberger> deiu, rogerm
15:22:48 <deiu> s/Richard/rogerm
15:22:56 <Arnaud> q?
15:23:06 <SteveS> q+
15:23:08 <stevebattle> q+
15:23:21 <Arnaud> ack steveb
15:23:23 <oberger> the 2 Steves again ;)
15:23:30 <deiu> stevebattle: the issues look different (32 vs 38)
15:23:46 <dret> it seems like ISSUE-38 is yet another question that should be addressed in ISSUE-37
15:24:19 <stevebattle> q+
15:24:30 <deiu> rogerm: I read ISSUE-38 as making resources discoverable, similar to ISSUE-32
15:24:37 <Arnaud> ack steves
15:24:50 <dret> discoverability should be factored out of all issues, it needs to be designed uniformally
15:24:58 <krp_> krp_ has joined #ldp
15:25:19 <dret> +1 to what cygri is saying
15:25:20 <AndyS> Move the discovery aspec of 38 to 32 but leave the filtering as 38
15:25:23 <Zakim> -??P66
15:25:36 <Zakim> +??P66
15:25:42 <deiu> ... should there be separate issue on how to make resources discoverable? There is a concern about discoverability. I don't understand why do we need two issues at this point.
15:25:47 <krp_> zakim, ??P66 is me
15:25:49 <Zakim> +krp_; got it
15:26:24 <dret> maybe adjust ISSUE-38 so that it refers to ISSUE-32?
15:26:29 <deiu> Arnaud: there are two issues: the filtering aspect and discovery
15:26:42 <deiu> ... ISSUE-38 focuses on the filtering aspect
15:26:46 <MacTed> once the feature is defined (#38), then it needs to be discoverable (#32)
15:26:52 <Arnaud> q?
15:27:02 <MacTed> two issues.  open #38.
15:27:04 <Arnaud> ack steveb
15:27:11 <bblfish> Issue-38?
15:27:11 <trackbot> ISSUE-38 -- filtered representations and inlining -- raised
15:27:11 <trackbot>
15:27:17 <deiu> ... by making something discoverable it doesn't mean it gets on the filtering system
15:27:32 <Arnaud> RESOLVED: open ISSUE-38
15:27:41 <Arnaud> reopen ISSUE-38
15:27:41 <trackbot> ISSUE-38 filtered representations and inlining re-opened
15:28:09 <deiu> ... we could edit ISSUE-38 to point to ISSUE-32 for discoverability aspects
15:28:13 <deiu> ISSUE-39?
15:28:13 <trackbot> ISSUE-39 -- HTTP status codes used for creation -- raised
15:28:13 <trackbot>
15:28:39 <deiu> Zakim, who is speaking?
15:28:50 <Zakim> deiu, listening for 10 seconds I could not identify any sounds
15:28:59 <stevebattle> Makes sense to me.
15:29:06 <Arnaud> q?
15:29:08 <oberger> 202 Accepted vs 201 Created
15:29:14 <deiu> stevebattle: there's a motivating case for it
15:29:23 <Zakim> -krp_
15:29:47 <SteveS> s/stevebattle: there's/SteveS: if there's/
15:30:09 <deiu> rogerm, RESTful representation of "turning a light bulb on", in slow motion; or the lifetime of a virtual machine
15:30:13 <mhausenblas> mhausenblas has joined #ldp
15:30:22 <Zakim> +??P66
15:30:29 <Zakim> -??P32
15:30:37 <stevebattle> Thanks Roger.
15:30:45 <bblfish> yes
15:30:47 <oberger> let's open it yes
15:30:48 <bblfish> sounds good
15:30:51 <SteveS> no objections with opening
15:30:55 <deiu> ... there is a need for cases where there is a process (multiple steps)
15:30:57 <Arnaud> RESOLVED: open ISSUE-39
15:31:04 <Arnaud> reopen ISSUE-39
15:31:04 <trackbot> ISSUE-39 HTTP status codes used for creation re-opened
15:31:09 <dret> we're just "defining HTTP" here, there is no need for us to define these things. this is how resources can behave according to HTTP's uniform interface.
15:31:27 <krp_> zakim, ??P66 is me
15:31:27 <Zakim> +krp_; got it
15:31:37 <deiu> Arnaud: moving on to issues pending review
15:31:40 <oberger> Topic: Use Cases and Requirements
15:31:50 <SteveS>
15:32:27 <deiu> SteveS: ongoing work on changing the document to have a more readable narrative style
15:32:44 <SteveS> s/SteveS/stevebattle/
15:32:48 <deiu> ... today it is a lot more readable
15:33:23 <Zakim> -krp_
15:33:29 <deiu> ... we have user stories that may not fit, so they need to be removed
15:33:52 <Zakim> +??P32
15:34:05 <deiu> Arnaud: there was a point where we had to freeze the document; are we reworking the structure?
15:34:35 <SteveS> q+
15:34:48 <cygri> q+
15:34:53 <Arnaud> ack steves
15:34:55 <deiu> ack SteveS
15:35:02 <cygri> q-
15:35:43 <deiu> SteveS: we talked about how much effort to put into making it into the final form; review the base sections to see what stays in; improving the current sections
15:36:11 <deiu> ... this will be the first thing to review, the workgroup today
15:36:19 <bblfish> Hi, I notice that there is no use-case for making a binary resource
15:36:28 <bblfish> so what would that be?
15:36:29 <bblfish> q+
15:36:31 <deiu> Arnaud: should we invite the group to start reviewing the document at this point?
15:36:49 <cygri> q+
15:36:50 <Arnaud> ack bblfish
15:37:01 <deiu> stevebattle: starting next meeting we should invite people that have use cases
15:37:02 <bblfish> hi
15:37:04 <SteveS> Isn't it at
15:37:15 <bblfish> so there is a story for it
15:37:24 <Zakim> -??P32
15:37:25 <bblfish> but no use case. What should one do ?
15:37:32 <stevebattle> yes deiu
15:37:34 <oberger> ack cygri
15:37:38 <Arnaud> ack cygri
15:37:46 <bblfish> +1 for cygri that was my question
15:37:54 <Zakim> +??P32
15:38:03 <SteveS> q+
15:38:04 <oberger> create issues...
15:38:09 <bhyland> bhyland has joined #ldp
15:38:15 <SteveS> q-
15:38:56 <bblfish> but if there is a user-story should there not be another use-case going with it?
15:39:11 <ericP> q?
15:39:42 <ericP> q+ to ask how i should edit the buglist use case
15:39:51 <bblfish> ok
15:39:53 <Arnaud> ack ericP
15:39:53 <Zakim> ericP, you wanted to ask how i should edit the buglist use case
15:40:23 <SteveS> q+
15:40:29 <Arnaud> ack steves
15:40:33 <deiu> ericP: wonder where to polish the use case for bug tracker
15:40:42 <deiu> ... will be on the mailing list
15:41:02 <deiu> SteveS: what will be the process for use cases?
15:41:07 <ericP> stevebattle, could you give me your favorite template use case from the doc for me to emulate?
15:41:24 <Zakim> -??P32
15:41:39 <deiu> Arnaud: we have two weeks during which people should review the use cases and raise issues if they don't get to an agreement
15:41:41 <Zakim> +??P41
15:41:58 <SteveS> To be clear, we are only proposing "Use Stories" as defined in the document….stevebattle can correct me if Use Cases are being asked for now
15:42:05 <stevebattle> Example scenario (for EricP)
15:42:19 <Arnaud> q?
15:42:42 <oberger> user stories
15:42:44 <deiu> Arnaud: by next week, people should have a look at their own use cases
15:42:52 <stevebattle> To recap: user-story authors should review their user-stories
15:43:17 <deiu> ... we don't have a plan to see how we can progress on this
15:44:07 <deiu> stevebattle: at the moment we adding examples to the user stories, and we're hoping to be done by next week
15:44:12 <stevebattle> yes
15:44:22 <deiu> Arnaud: people should add examples to the user stories
15:44:31 <oberger> Arnaud, aha ;)
15:44:40 <stevebattle> we're basing example scenarios on the user-stories.
15:45:07 <stevebattle> We're not adding examples to the user-stories.
15:45:10 <oberger> which year ? ;)
15:45:24 <Zakim> -??P41
15:45:27 <MacTed> Zakim, unmute me
15:45:27 <Zakim> MacTed should no longer be muted
15:45:30 <stevebattle> The spec had a head-start :)
15:45:32 <MacTed> Zakim, mute me
15:45:32 <Zakim> MacTed should now be muted
15:45:36 <MacTed> +1 stevebattle
15:45:42 <Zakim> +??P81
15:45:48 <oberger> topic: LDP Specification
15:46:44 <SteveS> ISSUE-25
15:46:55 <deiu> ISSUE-25?
15:46:55 <trackbot> ISSUE-25 -- Weak aggregation and strong composition in containers -- closed
15:46:55 <trackbot>
<deiu> Steves: sent email to the list on what changes will be made to the spec with regard to issue-25
<deiu> ... will now proceed to updating the spec
15:47:30 <stevebattle> Do we have a process for closing issues?
15:47:30 <deiu> Arnaud: we need to start closing issues
15:47:30 <oberger> topic: Open Issues
15:47:35 <MacTed> Zakim, unmute me
15:47:35 <Zakim> MacTed should no longer be muted
15:47:59 <deiu> MacTed: member submission was not focused on use cases
15:48:31 <stevebattle> q+
15:48:41 <Arnaud> ack steveb
15:48:42 <deiu> ... the process we're following here is bound to create a lot of issues
15:48:56 <deiu> stevebattle: do we have a process for closing issues? (see f2f style)
15:48:57 <cygri> q+
15:49:02 <SteveS> q+
15:49:09 <deiu> ... it would be good to have a process
15:49:24 <Zakim> -??P81
15:49:31 <deiu> Arnaud: ideally the process is to use this teleconf, going through the issues one by one
15:49:33 <oberger> when/if we stop filling new ones, the call will deal with closing them hopefully
15:49:45 <Zakim> +??P41
15:49:46 <Zakim> -ericP
15:49:53 <deiu> ... it's not easy to close issues on the mailing list
15:50:10 <cygri> q?
15:50:13 <deiu> ... if we get close to closing the issues on the mailing list, we can officially close them during calls
15:50:18 <Arnaud> ack cygri
15:50:50 <deiu> cygri: from my experience, I find this can be helpful for closing issues to have properly formulated proposal sent by email before calls
15:50:59 <bblfish> +1
15:51:04 <deiu> ... they become the baseline for discussions during the calls
15:51:12 <AshokMalhotra> +1 to wording change proposals for issues
15:51:19 <Arnaud> ack steves
15:52:02 <deiu> SteveS: we can plan the agenda to avoid pushing this process towards the end of the call
15:52:47 <deiu> Arnaud: when we started, we decided to allow the people who opened the issues to decide how they want to close them
15:52:49 <bblfish> perhaps also in e-mail people who send a mail that contains a soluton should put it in the header
15:52:52 <stevebattle> q+
15:53:06 <deiu> ... when people care about an issue, they should make concrete proposals and highlight them on the mailing list
15:53:21 <AndyS> Not sure a new issue can always have a proposed resolution at this stage -- e..g the overall design issue.
15:53:24 <Zakim> -??P41
15:53:24 <deiu> ... if it's a small issue, we can leave it open for discussion
15:53:29 <Arnaud> ack stevebattle
15:53:45 <Zakim> +??P41
15:53:51 <deiu> stevebattle: I agree with cygri; the resolution text will be different from the proposal text
15:53:54 <AndyS> Only as we get a more complete set of docs does the requirement of resolution work.
15:53:58 <cygri> q+
15:54:11 <Arnaud> ack cygri
15:54:47 <deiu> cygri: I want to agree with that; you will notice that people care about it and they will eventually write up a concrete proposal for closing the issue
15:55:00 <deiu> ... if I don't propose a solution, someone else will do it
15:55:38 <oberger> 5 mins left :-/
15:55:45 <deiu> Arnaud: is there an issue someone wants to discuss at this point?
15:56:01 <bblfish> I wanted to know if the RDF people agreed with the notion that a form is like a query
15:56:03 <AndyS> actions for the overall design text?
15:56:16 <stevebattle> I'm not hearing anything on the phone - ah I've been disconnected.
15:56:16 <bblfish> and how one can get ahead with doing forms in rdf
15:56:37 <cygri> ISSUE-37?
15:56:37 <trackbot> ISSUE-37 -- What is the LDP data model and the LDP interaction model? -- open
15:56:37 <trackbot>
15:56:44 <deiu> AndyS: there was an issue about the overall design text
15:57:07 <bblfish> AndyS: how does one make a model?
15:57:14 <dret> proposal is to use email 127 as starting point, and take it from there.
15:57:17 <bblfish> q+
15:57:23 <bblfish> ow does one make a model?
15:57:25 <Zakim> -??P41
#15:57:28 <Zakim> +SteveBattle.a
15:57:29 <Arnaud> ack bblfish
15:57:37 <stevebattle> it was me
15:57:39 <bblfish> how does one make a model?
15:57:56 <deiu> bblfish, just talk, I'll try to scribe
15:58:12 <AndyS> dret -- record an action?
15:58:19 <deiu> bblfish: is there a notion about how to make an interaction model?
15:58:20 <JohnArwe> we use interaction diagrams
15:58:22 <Zakim> +??P54
15:58:52 <deiu> AndyS: there is no formal language for it, it's about agreement
15:59:32 <deiu> ... I would volunteer to create that section of the spec; my expectation is to make it short (1-2 pages), add statements about how clients are supposed to interact with them
15:59:34 <bblfish> that sounds nearly like the spec.
15:59:44 <bblfish> q+
15:59:54 <Arnaud> ack bblfish
16:00:11 <Zakim> -Ashok_Malhotra
16:00:11 <bblfish> should I open an issue?
16:00:14 <deiu> bblfish: was there anything mentioned about RDF and forms?
16:00:37 <deiu> Arnaud: if you think that's a valid issue, then you may create one
16:00:53 <deiu> ... it has to be related to the spec, and to identify what the real problem is
16:01:08 <deiu> bblfish: it's about getting "views" of a collection: page 1, 2, etc.
16:01:25 <Zakim> -??P54
16:02:07 <deiu> Arnaud: people need to make concrete proposals on the mailing list first, to make a better use of teleconf time
16:02:14 <cygri> bblfish, i don't think there's an open issue for RDF forms. there's an issue for pagination.
16:02:24 <oberger> giving thanks to you
16:02:25 <SteveS> suggest we actually assign owners to each issue, I'll grab some that I'm interested
16:02:26 <dret> thanks, arnaud!
16:02:28 <stevebattle> bye
16:02:31 <bblfish> yes, I was thinking of using forms for pagination cygri
16:02:33 <Zakim> -rogerm
16:02:34 <ghard> Bye
16:02:34 <Zakim> -SteveS
16:02:34 <Zakim> -MHausenblas
16:02:36 <Zakim> -Yves
16:02:36 <Zakim> -oberger
16:02:36 <rgarcia> bye
16:02:37 <deiu> Arnaud: MEETING ADJOURNED
16:02:38 <Zakim> -MacTed
16:02:38 <antonis> bye
16:02:39 <Zakim> -AndyS
16:02:40 <Zakim> -JohnArwe
16:02:42 <Zakim> -Kalpa
16:02:42 <Zakim> -bblfish
16:02:42 <Zakim> -Arnaud
#16:02:43 <Zakim> -SteveBattle.a
16:02:43 <Zakim> -raul
16:02:44 <Zakim> -antonis
16:02:44 <Zakim> -dret
16:02:49 <Zakim> -ghard
16:03:01 <deiu> RRSAgent, draft minutes
16:03:01 <RRSAgent> I have made the request to generate deiu
16:03:11 <deiu> RRSAgent, make minutes public
16:03:11 <RRSAgent> I'm logging. I don't understand 'make minutes public', deiu.  Try /msg RRSAgent help
16:03:21 <deiu> RRSAgent, make log public
16:03:21 <bblfish> I proposed using SPARQL for forms
16:03:36 <Zakim> -deiu
16:03:51 <bblfish> but that was just a high level way of doing things
16:04:22 <dret> unfortunately, there's no concept of "issue ownership" in our tracker, afaict. that would be useful and pretty much all trackers have it, just not ours...
16:04:46 <deiu> bblfish, from my perspective (implementing LDP), I see all collections as SPARQL endpoints
16:05:25 <deiu> I wonder if that makes any sense
16:05:40 <dret> the implementation may use a SPARQL, but the interactions are limited to whatever the service surface makes available as interactions.
16:06:37 <deiu> it doesn't have to be a real (public) endpoint, though it can pass requests to the local (main) endpoint
16:06:51 <dret> if we say an LDP service *is a SPARQL endpoint*, then we're just doing GSP, aren't we?
16:07:03 <deiu> I'm not saying that :-)
16:07:08 <dret> good ;-)
16:07:21 <deiu> I'm saying that from an implementation p.o.v. it makes sense
16:07:21 <Zakim> -nmihindu
16:07:27 <AndyS> Another (not the only) view is the whole server is one SPARQL endpoint.  Both can be made to work.
16:08:46 <bblfish> deiu:  the forms are not for the server, but for the client
16:09:00 <bblfish> they are ways for the server to specify a way for the client to answer questions
16:09:32 <dret> we're doijng what pretty much every web-facing service on earth does: we're accepting requests where, let's say, URI parameters represent the specific way in which clients ask for service (give me page 42 and 42 entries per page: ...?pagesize=42&page=42), and the the implementation maps this service surface to whatever it uses as its implementation, retrieves the required data from its backend, and then generates whatever representation is required by the s[CUT]
16:09:34 <SteveS> dret, right I'd say that SPARQL is not required but one a possible implementation of LDP
16:10:22 <dret> no suprises here at all, standard web engineering, just against a RDF back-end, and with an RDF service surface (which is the tricky part, because of the links).
16:11:32 <dret> +1 to SteveS
16:12:21 <Zakim> disconnecting the lone participant, SteveBattle, in SW_LDP()10:00AM
16:12:23 <Zakim> SW_LDP()10:00AM has ended
16:12:23 <Zakim> Attendees were SteveBattle, MacTed, SteveS, Arnaud, deiu, JohnArwe, Ashok_Malhotra, dret, ericP, oberger, Yves, raul, MHausenblas, nmihindu, cygri, ghard, +44.208.573.aabb, krp,
16:12:23 <Zakim> ... antonis, bblfish, AndyS, rogerm, Kalpa, krp_
16:13:23 <AndyS> AndyS has left #ldp
16:54:07 <gavinc> gavinc has joined #ldp
17:00:45 <SteveS_> SteveS_ has joined #ldp
17:46:27 <SteveS> SteveS has joined #ldp
17:46:45 <SteveS> SteveS has joined #ldp
18:14:53 <Zakim> Zakim has left #ldp
18:23:08 <cgueret> cgueret has joined #ldp