IRC log of ldp on 2013-05-20

Timestamps are in UTC.

14:00:25 [RRSAgent]
RRSAgent has joined #ldp
14:00:25 [RRSAgent]
logging to
14:00:27 [trackbot]
RRSAgent, make logs public
14:00:27 [Zakim]
Zakim has joined #ldp
14:00:29 [trackbot]
Zakim, this will be LDP
14:00:29 [Zakim]
ok, trackbot, I see SW_LDP()10:00AM already started
14:00:30 [trackbot]
Meeting: Linked Data Platform (LDP) Working Group Teleconference
14:00:30 [trackbot]
Date: 20 May 2013
14:00:43 [Zakim]
14:00:55 [cygri]
cygri has joined #ldp
14:00:58 [SteveS]
zakim, [IBM] is me
14:01:02 [Zakim]
+SteveS; got it
14:01:14 [Arnaud]
zakim, who's on the phone?
14:01:14 [Zakim]
On the phone I see SteveBattle, JohnArwe, Arnaud, SteveS
14:01:22 [Zakim]
14:01:40 [stevebattle]
Arnaud, I'm happy to scribe today.
14:02:19 [stevebattle]
who said I'm happy?
14:03:01 [stevebattle]
scribe: stevebattle
14:03:29 [Zakim]
+ +44.208.5.aaaa
14:03:44 [roger]
roger has joined #ldp
14:03:51 [JohnArwe]
zakim, aaaa is Roger
14:03:51 [Zakim]
+Roger; got it
14:04:11 [Arnaud]
zakim, who's on the phone?
14:04:11 [Zakim]
On the phone I see SteveBattle, JohnArwe, Arnaud, SteveS, Ashok_Malhotra, Roger
14:05:21 [nmihindu]
nmihindu has joined #ldp
14:05:42 [Zakim]
14:05:48 [stevebattle]
chair: Arnaud
14:06:07 [bblfish]
14:06:41 [Zakim]
14:07:08 [BartvanLeeuwen]
BartvanLeeuwen has joined #ldp
14:07:17 [stevebattle]
roger: next week is a public holiday in the UK
14:07:31 [Arnaud]
nmihindu are you going to join the call?
14:07:39 [stevebattle]
SteveS: next week is also a public holiday in the US
14:07:54 [Arnaud]
what about you bart?
14:08:12 [Zakim]
14:08:22 [BartvanLeeuwen]
Zakim, ??P8 is me
14:08:22 [Zakim]
+BartvanLeeuwen; got it
14:08:28 [krp]
krp has joined #ldp
14:09:06 [Zakim]
14:09:38 [stevebattle]
Arnaud: The minutes for last week are not available so approval will be deferred until next week.
14:09:49 [bblfish]
we did progress though...
14:09:51 [nmihindu]
Arnaud, yes I will join in a minute. I have to connect to another network for VoIP
14:10:10 [Zakim]
14:10:36 [stevebattle]
Next week is also ESWC.
14:10:41 [nmihindu]
Zakim, ??P10 is me
14:10:41 [Zakim]
+nmihindu; got it
14:11:15 [SteveS]
I won't be able to attend May 27th
14:11:20 [Zakim]
14:11:47 [stevebattle]
Arnaud: We will skip the meeting next Monday, next meeting June 3rd.
14:12:06 [stevebattle]
resultion: next meeting June 3rd
14:12:18 [stevebattle]
resolution: next meeting June 3rd.
14:12:39 [JohnArwe]
14:12:41 [krp]
zakim, ??P11 is me
14:12:41 [Zakim]
+krp; got it
14:13:49 [stevebattle]
ericP: Are we meeting at TPAC?
14:13:49 [bblfish]
14:14:34 [stevebattle]
Arnaud: Has seen no message regarding ericP's hallucination.
14:15:31 [stevebattle]
topic: actions
14:16:26 [JohnArwe]
pending review: 45, 59, 60, 65
14:16:53 [stevebattle]
Arnaud: Proposes we close these actions.
14:17:02 [SteveS]
14:17:08 [stevebattle]
14:17:11 [JohnArwe]
14:17:12 [nmihindu]
14:17:19 [krp]
14:17:36 [bblfish]
Arnaud: undocumented action for veryone to object to the minutes if closing these issues is problematic
14:17:47 [stevebattle]
resolution: close actions 45, 59, 60, 65
14:18:11 [JohnArwe]
close action-45
14:18:11 [trackbot]
Closed ACTION-45 [EDITOR] Adopting with the property name ldp:membershipPredicateInverse.
14:18:19 [JohnArwe]
close action-59
14:18:19 [trackbot]
Closed ACTION-59 [EDITOR] Deployment guide, best practice is to make predicates used in LDPRs be dereferencable, per resolution of issue-9.
14:18:26 [JohnArwe]
close action-60
14:18:26 [trackbot]
Closed ACTION-60 [EDITOR] LDPCs may contain non-LDPRs + updating membership triples and inlined members, per resolution of issue-13.
14:18:28 [stevebattle]
Arnaud: There are outstanding actions to review UC&R
14:18:33 [JohnArwe]
close action-65
14:18:33 [trackbot]
Closed ACTION-65 [EDITOR] Remove section 4.1.4 canonical uri from the spec, per resolution of issue-49.
14:18:42 [nmihindu]
Miguel is working on the UCR review and he will send it soon
14:18:53 [Kalpa]
Kalpa has joined #ldp
14:18:55 [SteveS]
+1 to closing 52
14:19:02 [stevebattle]
Arnaud: resolution close action 52
14:19:07 [bblfish]
14:19:07 [trackbot]
ACTION-52 -- Roger Menday to create a wish list wiki page with issue-38 -- due 2013-03-21 -- OPEN
14:19:07 [trackbot]
14:19:33 [stevebattle]
resolution: close action-52
14:19:51 [JohnArwe]
close action-52
14:19:52 [trackbot]
Closed ACTION-52 Create a wish list wiki page with issue-38.
14:20:45 [stevebattle]
close action-50
14:20:45 [trackbot]
Closed ACTION-50 And rgarcia to come up with a revised proposal for the test suite framework.
14:20:45 [bblfish]
14:20:45 [trackbot]
ACTION-50 -- Eric Prud'hommeaux to and rgarcia to come up with a revised proposal for the test suite framework -- due 2013-03-21 -- CLOSED
14:20:46 [trackbot]
14:20:56 [bblfish]
14:20:56 [trackbot]
ACTION-48 -- Ashok Malhotra to take a first crack at the requirements of Access Control note -- due 2013-03-21 -- OPEN
14:20:56 [trackbot]
14:21:28 [nmihindu]
14:22:20 [Zakim]
14:22:47 [stevebattle]
close action-48
14:22:47 [trackbot]
Closed ACTION-48 Take a first crack at the requirements of Access Control note.
14:23:08 [stevebattle]
topic: Issues
14:23:51 [bblfish]
14:23:51 [trackbot]
ISSUE-62 -- Creating containers associated with LDPRs -- raised
14:23:51 [trackbot]
14:24:20 [stevebattle]
rogerp: I revised the text of the issue for clarity.
14:25:16 [stevebattle]
What does issue-62 add?
14:25:32 [stevebattle]
14:26:11 [stevebattle]
roger: The difference is that you don't necessarily have a POSTable endpoint with a container.
14:26:41 [SteveS]
Perhaps we can cover with ACTION-55, which was to cover ISSUE-36
14:26:41 [BartvanLeeuwen]
14:27:04 [stevebattle]
open issue-62
14:27:22 [bblfish]
14:27:26 [bblfish]
14:27:36 [Arnaud]
ack bart
14:27:52 [BartvanLeeuwen]
14:28:04 [bblfish]
14:28:10 [stevebattle]
bartvanleeuwen: ???
14:29:14 [BartvanLeeuwen]
does this allow us to 'bootstrap' a ldp server ??
14:29:34 [BartvanLeeuwen]
so no containers at all, and we can create them
14:29:59 [JohnArwe]
if by bootstrap you mean go from an LDPR to an LDPC, sounds like yes.
14:30:00 [BartvanLeeuwen]
14:30:08 [stevebattle]
Arnaud: We can always bootstrap containers using PUT
14:30:14 [Arnaud]
reopen issue-62
14:30:14 [trackbot]
Re-opened ISSUE-62 Creating containers associated with LDPRs.
14:30:23 [bblfish]
but how yould you know where to PUT, you'd have to PUT in a container or something.
14:30:39 [stevebattle]
14:30:39 [trackbot]
ISSUE-70 -- simple LDPCs -- raised
14:30:39 [trackbot]
14:32:04 [stevebattle]
bblfish: Some parts of the spec may not be absolutely necessary. We should review them, simplifying the spec.
14:32:44 [SteveS]
Didn't we cover this beginning of the year when we created the UC&R doc?
14:34:04 [stevebattle]
Arnaud: This may be too disruptive at this stage; we need to come up with specific proposals.
14:34:17 [Ashok]
maybe this goes in the primer?
14:34:37 [JohnArwe]
henry what do you mean by "property links"?
14:34:50 [roger]
membershipPredicate and membershipSubject
14:34:51 [roger]
I think
14:36:20 [stevebattle]
Arnaud: The existing features are justified by many use-cases.
14:37:25 [stevebattle]
Arnaud: There are, of course, alternative ways to achive the same results.
14:37:27 [roger]
14:37:50 [Zakim]
14:37:54 [stevebattle]
bblfish: But are there ways to achive the same results in a *simpler* way?
14:38:05 [stevebattle]
14:38:45 [Zakim]
14:38:46 [stevebattle]
14:38:52 [BartvanLeeuwen]
Zakim, ??P8 is me
14:38:52 [Zakim]
+BartvanLeeuwen; got it
14:39:15 [Arnaud]
ack roger
14:39:16 [stevebattle]
Arnaud: Opening these issue may present a road-block to last call
14:39:27 [stevebattle]
14:40:21 [stevebattle]
roger: Do you want to have a solution that uses only rdfs:member to associate containers with members?
14:40:26 [ericP]
14:40:27 [ericP]
14:40:38 [Arnaud]
ack ericp
14:40:54 [Zakim]
14:41:23 [stevebattle]
ericp: You could rally the troops to come up with a counter-proposal, add this to the list, then use this as the basis of discussion.
14:42:34 [stevebattle]
Arnaud: This is good but we cannot have open-ended issues.
14:43:20 [stevebattle]
close issue-70
14:43:20 [trackbot]
Closed ISSUE-70 simple LDPCs.
14:44:12 [stevebattle]
topic: open issues
14:44:19 [bblfish]
14:44:19 [trackbot]
ISSUE-65 -- FirstPage and Hypermedia as the Engine of Application State (HATEOAS) Compliance -- open
14:44:19 [trackbot]
14:44:59 [stevebattle]
Arnaud: The issue was first raised by James Leigh.
14:45:31 [bblfish]
Issue-70 is closed for reasons of not precise enough wording.
14:45:49 [stevebattle]
arnaud: We lose the ability for the client to initiate paging.
14:46:18 [SteveS]
ACTION-62 hasn't been done which edits in resource pagination from resolution of ISSUE-33
14:46:26 [stevebattle]
arnaud: A link header can be used to supply the first page URL
14:47:09 [SteveS]
14:47:16 [JohnArwe]
+1 to removing firstPage, do think we need link header
14:47:36 [bblfish]
The headers seem like a good idea. I have not implemented this part of the spec yet.
14:47:38 [stevebattle]
+1 to removal, link header should be optional.
14:47:40 [JohnArwe]
...our mobile clients will want to keep response sizes manageable
14:48:03 [SteveS]
+1 to Arnaud's all 3 proposals, including having link header optional
14:48:29 [BartvanLeeuwen]
14:48:30 [Arnaud]
PROPOSAL: Close ISSUE-65: FirstPage HATEOAS Compliance per Arnaud's proposal, with link header optional
14:48:39 [stevebattle]
14:48:41 [Ashok]
14:48:53 [nmihindu]
14:48:53 [roger]
14:48:54 [krp]
14:48:57 [Kalpa]
14:49:00 [SteveS]
14:49:15 [Arnaud]
Resolved: Close ISSUE-65: FirstPage HATEOAS Compliance per Arnaud's proposal, with link header optional
14:49:30 [stevebattle]
14:49:30 [trackbot]
ISSUE-62 -- Creating containers associated with LDPRs -- open
14:49:30 [trackbot]
14:51:15 [stevebattle]
Arnaud: action-55 adds clarification. Does this address issue-62 fully?
14:52:15 [stevebattle]
Arnaud: We'll postpone until we have the text of issue-55.
14:52:40 [bblfish]
14:52:40 [trackbot]
ISSUE-58 -- Property for asserting that complete description of members is included in LDPC representation -- open
14:52:40 [trackbot]
14:52:51 [JohnArwe]
close issue-65
14:52:51 [trackbot]
Closed ISSUE-65 FirstPage and Hypermedia as the Engine of Application State (HATEOAS) Compliance.
14:54:15 [bblfish]
14:54:26 [SteveS]
Here is the email summary of options:
14:54:31 [Arnaud]
ack bblfish
14:55:10 [bblfish]
14:55:12 [stevebattle]
bblfish: This is a very bad idea. This makes it possible to create inconsistent graphs and is unnecessary.
14:56:05 [JohnArwe]
14:56:13 [Ashok]
14:57:11 [stevebattle]
arnaud: I don't see how inlining introduces the problem. The inconsistency arises out of the data.
14:57:22 [Arnaud]
ack john
14:57:55 [stevebattle]
johnarwe: The arguments expose different assumptions that people hold.
14:59:09 [Arnaud]
ack ashok
15:01:25 [Arnaud]
15:02:02 [stevebattle]
arnaud: The option on the table is to close with option E
15:02:11 [nmihindu]
Option E: A combination of both Options A & B, in other words people wanted both.
15:02:24 [bblfish]
So since then I move to -1 on all
15:02:25 [JohnArwe]
...with proviso that the subject is the Page not the Resource
15:02:34 [stevebattle]
15:03:16 [JohnArwe]
need to drop
15:03:17 [stevebattle]
ashok: We should explore Henry's idea just a little further.
15:03:30 [bblfish]
My proposal is to have a content relation to a literal
15:03:51 [JohnArwe]
Henry proposed a concrete alternative.
15:03:56 [Zakim]
15:04:04 [stevebattle]
ashok: I asked Henry for a simple example that was added last night which the working group should provide feedback on.
15:04:06 [SteveS]
+0.5 on option E
15:04:54 [bblfish]
15:04:54 [trackbot]
ISSUE-32 -- How can clients discover that a resource is an LDPR or LDPC, and what features are supported? -- open
15:04:54 [trackbot]
15:05:13 [stevebattle]
arnaud: I don't want to give this issue too much agenda time.
15:05:29 [SteveS]
regrets from me for May 27 and June 3
15:05:33 [stevebattle]
15:05:35 [Zakim]
15:05:36 [BartvanLeeuwen]
thx bye
15:05:36 [Zakim]
15:05:37 [bblfish]
15:05:39 [Zakim]
15:05:40 [Zakim]
15:05:41 [Zakim]
15:05:42 [Zakim]
15:05:42 [Zakim]
15:05:43 [Zakim]
15:05:44 [Zakim]
15:05:47 [Zakim]
15:05:57 [Kalpa]
Kalpa has left #ldp
15:06:35 [bblfish]
Arnaud: please look up my last mail and see if you agree that there is a problem with the logical inconsistency problem being so easy to build
15:07:19 [Arnaud]
Henry, I read that email
15:07:37 [Arnaud]
I don't see how the inlining really changes anything
15:07:50 [Arnaud]
the inconsistency is already there
15:07:58 [Arnaud]
it only makes it more evident
15:08:02 [bblfish]
but in a different graph
15:08:11 [Arnaud]
15:08:24 [Arnaud]
at the application level it would make no difference to me
15:08:24 [bblfish]
what if the inlined content has relations about the LDPC , say chaing the state of it
15:09:14 [Arnaud]
I agree that it can be used in a bad way
15:09:41 [Arnaud]
in our case the server wouldn't allow you to do that
15:09:49 [Arnaud]
so the problem never occurs
15:10:10 [Arnaud]
that's why I said in my email that I can see how a vanilla implementation may not be able to do that easily
15:10:30 [bblfish]
another argument is that inlining means that a client that POSTs needs to have not just the overview of the page, but the overview of all of the content of the LDPC because I suppose page orderings can change, etc... Otherwise when POSTing you can easily make things inconsistent
15:10:47 [Zakim]
disconnecting the lone participant, bblfish, in SW_LDP()10:00AM
15:10:48 [Zakim]
SW_LDP()10:00AM has ended
15:10:48 [Zakim]
Attendees were SteveBattle, JohnArwe, Arnaud, SteveS, Ashok_Malhotra, +44.208.5.aaaa, Roger, bblfish, ericP, BartvanLeeuwen, nmihindu, krp, Kalpa
15:11:44 [Arnaud]
again you're assuming the client can post anything it wants
15:12:11 [Arnaud]
in an application specific case the server wouldn't support that
15:12:14 [bblfish]
we don't have any vocabulary to restrict the client yet to what it should post
15:12:19 [Arnaud]
it would allow a client to create specific data
15:12:22 [Arnaud]
in a controlled way
15:12:30 [Arnaud]
well, we do :)
15:12:47 [Arnaud]
btw, did you see the announcement for the RDF validation workshop?
15:12:52 [Arnaud]
this is what it is about
15:13:03 [Arnaud]
as far as we are concerned
15:13:08 [bblfish]
I did not see that yet, no
15:13:12 [Arnaud]
hold on
15:13:39 [Arnaud]
15:13:42 [bblfish]
But even with RDF validation local validation is not enough: in the simple bug example: I managed to get an inconsistency just because of different points of views being expressed about what the cause of the bug is
15:13:46 [Arnaud]
btw, IBM didn't initiate this
15:13:59 [Arnaud]
eric was working on setting it up when I learned about it
15:14:10 [Arnaud]
so there are other people wanting this
15:14:12 [bblfish]
that's ok, I don't have anything against IBM. :-)
15:14:27 [bblfish]
being supporters of Linux is cool :-)
15:14:32 [Arnaud]
15:15:05 [Arnaud]
we have products using this functionality and we haven't had problems with that
15:15:21 [bblfish]
Anyway, my point is just that I think this issue arises because we're just trying to put too much into the LDPC. We could just shift this problem to an LDPR,
15:16:01 [Arnaud]
I don't see that
15:16:16 [bblfish]
yes, Arnaud, but you control the whole stack of your product, and you employ all the engineers. If you control all aspects there are things you won't see that people in the wild will see with an open world.
15:17:10 [bblfish]
Well I am not sure about that claim yet. But my feeling is that if one had simple LDPCs then we'd move the rest of the work to dealing with LDPRs
15:17:11 [Arnaud]
well, we integrate with products from other vendors and customers so that's not exactly true
15:17:31 [bblfish]
well I don't want to speak much about your product as I don't know it
15:18:13 [Arnaud] gives you the list of organizations involved
15:18:19 [Arnaud]
it's hardly just IBM
15:18:59 [Arnaud]
but I agree that we do exercise some control in what the server accepts and without it things can go bad
15:19:15 [Arnaud]
that comes with the flexibility of linked data
15:19:22 [Arnaud]
the fact that anyone can assert anything
15:19:37 [bblfish]
that is why it is important to distinguish creation of documents using POST and content of documents ie. <#uri> and mergers of graphs
15:20:07 [bblfish]
an LDPC can have inconsistent members - that's ok if those members are documents.
15:20:25 [Arnaud]
but if you got an inconsistent graph from inlining couldn't you go back and check what each resource contains and figure out where the inconsistency comes from?
15:20:37 [bblfish]
I mean the log:semantics of the LDPC members can be inconsistent - without problem as long as rdf:member <member> the <member> is a document
15:20:47 [Arnaud]
at least you would only have to fetch every resource individually when things go bad rather than all the time
15:21:22 [bblfish]
I am just wondering why you want to merge all that info in the LDPC?
15:21:45 [Arnaud]
for performance + convenience
15:21:57 [Arnaud]
you do one get and you're all set
15:21:58 [bblfish]
ok so why not use quoted content
15:22:12 [bblfish]
with quoted content you get it all, and it is safe
15:22:20 [bblfish]
the client can decide if he wishes to unquote
15:22:37 [bblfish]
ie to believe the content
15:23:55 [Arnaud]
that seems reasonable but it doesn't work with turtle, does it?
15:24:23 [bblfish]
not well no. But then you could as you did above point to future specs like N3 or Trig
15:24:31 [bblfish]
N3 it works very well
15:24:39 [Arnaud]
yeah, I read you said that
15:24:56 [bblfish]
<member> log:semantics { <#joe> a foaf:Person }
15:25:02 [Arnaud]
right now we are requiring support for turtle
15:25:17 [bblfish]
15:25:35 [bblfish]
yes so there one would need to have a way to put Turtle in quotes.
15:25:46 [bblfish]
"..."^^lang:Turtle .
15:25:59 [bblfish]
it's just a question of getting some URLs for those languages
15:26:06 [Arnaud]
"just" :)
15:26:25 [bblfish]
well that's not a very big task. People could see the point of that relatively easily
15:26:35 [Arnaud]
that's not the problem
15:26:43 [Arnaud]
the problem is that turtle is in last call
15:26:43 [bblfish]
but I think it's a good spur for N3. TimBl would love to have that standardised I am sure.
15:26:53 [Arnaud]
you know what it takes to add something like that
15:27:18 [Arnaud]
maybe he would but again it's not even in the works right now
15:27:24 [bblfish]
I can ask on the Turlte list if they have a standard way of doing that
15:28:15 [bblfish]
but you end up with a lot less issues than if you merge the content
15:29:09 [bblfish]
If you merge content you could end up in the wild with a bunch of jokers messing up your collections.
15:29:43 [bblfish]
and also be able to argue their way out of the problem by saying they were not conscious of the consistency requirements
15:30:51 [Arnaud]
I understand, that again assumes people are allowed to do all sorts of stuff
15:31:07 [Arnaud]
this is definitely not the case in an enterprise application
15:31:33 [Arnaud]
that's a big difference with the web at large
15:33:29 [bblfish]
Even enterprise applications will have the problem. If I can prove that you need local context to exchange information which you have not made explicit, then you are not doing semantic web.
15:33:50 [bblfish]
semantic web is about building a system that does not require local contextual knowledge
15:34:10 [bblfish]
IT is about building worldwide platform.
15:34:12 [betehess]
log:semantics translates to plain RDF using reification
15:35:22 [bblfish]
what I'd need is <member> log:semantics "<> a <"^^lang:Turtle .
15:36:58 [betehess]
pretty much the same, isn't it?
15:37:36 [bblfish]
same as what?
15:37:44 [bblfish]
as a bunch of reified statements?
15:38:00 [betehess]
I guess
15:38:30 [bblfish]
should be the same but a lot more readable than reification.
15:39:20 [bblfish]
ok I sent out an initial mail to see what the situation is.
15:40:09 [bblfish]
Mhh pitty we did not log the above conversation :-)
15:40:55 [Arnaud]
bblfish: for what it's worth I never said we are doing semantic web :)
15:41:10 [Arnaud]
I spend quite a bit of time actually saying the opposite!
15:41:33 [betehess]
semantic web :-)
15:43:49 [bblfish]
yes but the ldp group is about that, and the w3c is about it.
15:45:11 [bblfish]
The problem is that it is difficult for large companies to think about the semantic web ( I worked at sun) because they tend to not notice implicit assumptions they are using. One really needs to make these things explicit for it to count as linked data.
15:45:55 [Arnaud]
I don't agree with that assertion "the ldp group is about that, and the w3c is about it." actually
15:46:09 [Arnaud]
and the discussions I have had with the w3c staff confirms that
15:46:41 [bblfish]
( I think its not just large companies btw that have trouble with the explicitness required )
15:47:01 [bblfish]
well I am pretty sure that is a key criterion for ldp to be ldp
15:47:14 [betehess]
well, I still don't know what semantic web means...
15:47:35 [bblfish]
still it's not well expressed above so we may be interpreting that statement differently
15:47:48 [betehess]
but I think I understand Linked Data, as a concept and as a well-defined technology: LDP
15:48:19 [bblfish]
I need to be able to read your data and not know that you are IBM to interpret it and interact with it - I may need to know you are IBM to TRUST it but not to interpret it
15:49:35 [bblfish]
betehess: we are looking for criterion for LDP here. We can't just say whatever LDP comes up with is good. LDP has to be part of the larger semantic web stack
15:50:02 [bblfish]
and the criterion has to be as defined above: ie "I need to be able to read your data and not know that you are IBM to interpret it and interact with it - I may need to know you are IBM to TRUST it but not to interpret it"
15:52:20 [bblfish]
I may not be able to interpret fully all the relations because they are not accessible to me, but that's something else.
15:55:29 [bblfish]
Some of this is just application of Web Arch. The data has to be accessible from any resource: you should not assume someone is coming from a certain position in the web.
15:56:02 [bblfish]
that's essentially why RDF exists: as a maxiumum decontextualisation of information.
15:56:54 [bblfish]
( Context still exists: you have a certain position in the web: I can publish something and IBM the same thing, but they will have different positions in the web. People may trust IBM and not me even if we say the same thing )
15:57:25 [bblfish]
That is why WebID works.
17:21:09 [Zakim]
Zakim has left #ldp
17:50:37 [jmvanel]
jmvanel has joined #ldp
18:39:31 [AndyS]
AndyS has joined #ldp
18:54:56 [jmvanel]
jmvanel has joined #ldp
19:29:49 [SteveS]
SteveS has joined #ldp
20:38:12 [gavinc]
gavinc has joined #ldp
21:33:33 [bblfish_]
bblfish_ has joined #ldp
21:51:59 [bblfish]
bblfish has joined #ldp
22:16:56 [jmvanel]
jmvanel has joined #ldp