IRC log of dig on 2013-08-12

Timestamps are in UTC.

02:37:07 [timbl]
timbl has quit (Ping timeout: 246 seconds)
02:37:09 [timbl_]
timbl_ (~timbl@184.151.61.52) has joined #dig
04:23:48 [timbl_]
timbl_ has quit (Quit: timbl_)
08:00:41 [deiu]
deiu (~andrei@unaffiliated/deiu) has joined #dig
08:07:11 [cheater__]
cheater__ has quit (Ping timeout: 246 seconds)
08:08:50 [cheater__]
cheater__ (~cheater@p57AE9430.dip0.t-ipconnect.de) has joined #dig
11:34:10 [Ralph]
Ralph (rswick@w3cvpn1.w3.org) has joined #dig
12:24:59 [Ralph]
Ralph is now known as RalphS
12:26:00 [DIGlogger]
DIGlogger (~dig-logge@groups.csail.mit.edu) has joined #dig
12:26:00 [morgan.freenode.net]
topic is: Decentralized Information Group @ MIT http://dig.csail.mit.edu/
12:26:00 [morgan.freenode.net]
Users on #dig: DIGlogger RalphS cheater__ deiu bblfish melvster Yudai zuzak presbrey ericP sandro mattl bergi tyteen4a03
12:39:17 [scor]
scor (scor@drupal.org/user/52142/view) has joined #dig
12:54:24 [timbl]
timbl (~timbl@huntsville-dialup-209-91-149-66.vianet.ca) has joined #dig
12:59:18 [timbl_]
timbl_ (~timbl@huntsville-dialup-209-91-149-66.vianet.ca) has joined #dig
13:00:15 [bblfish]
odd cwm breaks on sioc ontology even though the validator says its fine
13:00:20 [bblfish]
cwm --rdf http://rdfs.org/sioc/ns --n3
13:01:04 [timbl]
timbl has quit (Ping timeout: 264 seconds)
13:01:04 [timbl_]
timbl_ is now known as timbl
13:01:34 [bblfish]
https://gist.github.com/anonymous/9dead3e42e8d4de953bd
13:01:52 [bblfish]
that's the error I get. Did an update of cwm just in case
13:10:32 [betehess]
betehess (~betehess@2001:470:8b2d:804:823f:5dff:fe08:6375) has joined #dig
13:15:57 [Yudai]
Yudai has quit (Quit: Tiarra 0.1+svn-36726: SIGTERM received; exit)
13:19:55 [Yudai]
Yudai (~Yudai@nttkyo218001.tkyo.nt.ngn2.ppp.infoweb.ne.jp) has joined #dig
13:25:43 [deiu]
deiu has quit (Quit: Leaving)
14:11:58 [bblfish_]
bblfish_ (~bblfish@AAubervilliers-651-1-323-225.w83-200.abo.wanadoo.fr) has joined #dig
14:14:48 [bblfish]
bblfish has quit (Ping timeout: 276 seconds)
15:02:53 [timbl]
timbl has quit (Ping timeout: 256 seconds)
15:56:05 [deiu]
deiu (~andrei@unaffiliated/deiu) has joined #dig
16:12:53 [sandro]
sandro has quit (Ping timeout: 248 seconds)
16:13:23 [sandro]
sandro (~sandro@ssh.w3.org) has joined #dig
16:24:53 [timbl]
timbl (~timbl@184.151.61.4) has joined #dig
16:38:34 [bblfish_]
hi betehess? are you back from holidays?
16:56:01 [bblfish_]
bblfish_ is now known as bblfish
17:11:54 [betehess]
hey bblfish
17:12:07 [bblfish]
hi
17:12:27 [bblfish]
how are you?
17:13:28 [betehess]
good, thanks!
17:14:32 [bblfish]
Have been busy with Javascript recently, improving the rdflijb.js demo. ( will push some code there soon ). Then was at hacker conf in Holland. And this week just been helping with web2.0 code at Stample. So not much progress on banana. But was just writing a pre/post condition system for ldp - so one could easily say only allow say sioc:Post-s to be posted into a ldp:Container that is also a sioc:Container
17:17:41 [bblfish]
was wondering about ldp and event systems - perhaps you have ideas on this. One solution is to have agents post interest in receiving events on a set of resources. Then ldp actors could send some agent events to process
17:19:35 [betehess]
not sure I understand what you want to do with actors
17:19:45 [betehess]
I mean, agents
17:19:49 [betehess]
you mean Akka agents?
17:20:33 [bblfish]
package org.w3.banana.ldp
17:20:33 [bblfish]
import org.w3.banana._
17:20:33 [bblfish]
import org.openrdf.model.Graph
17:20:34 [bblfish]
/**
17:20:36 [bblfish]
* Advice to be applied before or after application of a Command
17:20:38 [bblfish]
*/
17:20:40 [bblfish]
trait CommandAdvice[Rdf<:RDF] {
17:20:42 [bblfish]
/**
17:20:44 [bblfish]
* run some pre-conditions on an ldp ldp command
17:20:46 [bblfish]
*
17:20:48 [bblfish]
* @param cmd the initial command
17:20:50 [bblfish]
* @tparam A
17:20:52 [bblfish]
* @return Some throwable if the command should not continue.
17:20:54 [bblfish]
* // a more java way would just be to throw the exception and declare that it does.
17:20:56 [bblfish]
* // does it help to just return it?
17:20:58 [bblfish]
*/
17:21:00 [bblfish]
def pre[A](cmd: LDPCommand[Rdf, LDPCommand.Script[Rdf,A]]): Option[Throwable]
17:21:02 [bblfish]
/**
17:21:05 [bblfish]
* run some actions after executing the command.
17:21:06 [bblfish]
*
17:21:08 [bblfish]
* @param cmd the initial command - todo: it may be useful to also get the resulting command, so that if needed one
17:21:10 [bblfish]
* can be flatmapped on the other.
17:21:12 [bblfish]
* @tparam A
17:21:14 [bblfish]
*/
17:21:16 [bblfish]
def post[A](cmd: LDPCommand[Rdf, LDPCommand.Script[Rdf,A]])
17:21:18 [bblfish]
}
17:21:20 [bblfish]
/**
17:21:22 [bblfish]
* An Advice Selector is a function that takes a graph, and can find CommandAdvices to apply for the given resource
17:21:24 [bblfish]
* //todo: it should be a function from LDPResource to command advice so that metadata on the resource is
17:21:26 [bblfish]
* taken into account.
17:21:28 [bblfish]
* @tparam Rdf
17:21:30 [bblfish]
*/
17:21:32 [bblfish]
trait AdviceSelector[Rdf<:RDF] extends Function1[PointedGraph[Rdf],CommandAdvice[Rdf]]
17:21:34 [bblfish]
/**
17:21:36 [bblfish]
* try building an advice selector to test the interface
17:21:38 [bblfish]
* @param ops
17:21:40 [bblfish]
* @tparam Rdf
17:21:42 [bblfish]
*/
17:21:44 [bblfish]
class TestAdviceSelector[Rdf<:RDF](implicit ops: RDFOps[Rdf]) extends AdviceSelector[Rdf] {
17:21:46 [bblfish]
import ops._
17:21:48 [bblfish]
import org.w3.banana.diesel._
17:21:50 [bblfish]
import syntax._
17:21:52 [bblfish]
val sioc = SiocPrefix[Rdf]
17:21:54 [bblfish]
def apply(resourceGraph: PointedGraph[Rdf]) = {
17:21:56 [bblfish]
for (tp <- resourceGraph / rdf.typ) yield {
17:21:58 [bblfish]
tp.pointer match {
17:21:59 [betehess]
I'm sure people here will like seeing so much Scala at once :-)
17:22:00 [bblfish]
case sioc.Container => new CommandAdvice[Rdf] {
17:22:02 [bblfish]
def pre[A](cmd: LDPCommand[Rdf, LDPCommand.Script[Rdf, A]]) = {
17:22:04 [bblfish]
cmd match {
17:22:06 [bblfish]
case CreateLDPR(container,_,graph,k) => {
17:22:08 [bblfish]
val passes = (PointedGraph(container,graph.resolveAgainst(container))/rdf.typ).exists{ pg =>
17:22:10 [bblfish]
pg.pointer == sioc.Item || pg.pointer == sioc.Post
17:22:12 [bblfish]
}
17:22:14 [bblfish]
if (!passes)
17:22:16 [bblfish]
Some(WrongExpectation("One can only Post sioc:Item and so sioc:Posts to a sioc:Container that is an ldp:Container"))
17:22:18 [bblfish]
else None
17:22:20 [bblfish]
}
17:22:22 [bblfish]
//what other things would one want to test ?
17:22:24 [bblfish]
case _ => None
17:22:26 [bblfish]
}
17:22:28 [bblfish]
}
17:22:30 [bblfish]
def post[A](cmd: LDPCommand[Rdf, LDPCommand.Script[Rdf, A]]) {
17:22:32 [bblfish]
cmd match {
17:22:34 [bblfish]
case CreateLDPR(container,_,graph,k) => {
17:22:36 [bblfish]
//we should look into the resourceGraph to check if there is a link to a list of listeners
17:22:38 [bblfish]
//and if there is we should send each of them notifications. The notifications can be sent
17:22:40 [bblfish]
//to an event bus.
17:22:42 [bblfish]
}
17:22:44 [bblfish]
}
17:22:46 [bblfish]
}
17:22:48 [bblfish]
}
17:22:50 [bblfish]
}
17:22:52 [bblfish]
}
17:22:54 [bblfish]
}
17:22:56 [bblfish]
}
17:22:58 [bblfish]
oops
17:23:00 [bblfish]
sorry meant to sent that to gist
17:23:04 [bblfish]
https://gist.github.com/anonymous/1e7d58f0f7186fb3534b
17:55:17 [deiu]
bblfish, try http://fixee.org/home/
17:57:13 [bblfish]
thanks
18:08:17 [timbl]
bblfish, Hmmm re cwm --rdf http://rdfs.org/sioc/ns --n3 …, strange
18:17:38 [timbl]
uri should be RDF_NS_URI + ln it seems
18:17:43 [timbl]
checked in new version
18:17:58 [timbl]
bizarre after all these years
18:21:11 [bblfish]
thanks
18:23:17 [bblfish]
presbrey: over at OHM we made some more changes to the rdflib demo application. https://github.com/stample/rdflib.js/blob/master/example/people/social_book.js
18:23:46 [bblfish]
it works a lot better now. the columns slide over to the right as one explores more the social graph.
18:23:51 [bblfish]
still a lot more work to do.
18:24:21 [bblfish]
but that's probably all I can get done for now.
18:26:34 [bblfish]
not sure I can do a pull request because you did not accept the changes to the directory structure
18:31:28 [deiu]
I'm close to finishing a have a RWW demo app, based on rdflib.js
18:31:54 [deiu]
however, I'm not sure how to handle authenticated requests in rdflib
18:33:05 [deiu]
it would be great to be able to add client certs to requests somehow
18:35:03 [deiu]
https://github.com/deiu/findMyLoc
18:36:29 [deiu]
you can just copy/paste the contents of index.html to a file in your own RWW data space
18:44:42 [bblfish]
you can try it out here http://stample.github.io/rdflib.js/example/people/social_book.html
18:54:28 [bblfish]
( there is some problem with my makefile.... at ohm the folks had to do make 3 times before the thing compiled.... pretty odd)
18:56:51 [danbri]
danbri (~danbri@146.90.244.38) has joined #dig
19:48:19 [timbl]
I tried out the location sharing app deiu
19:48:37 [timbl]
at https://timbl.data.fm/test2/locator/locator.html
19:48:50 [timbl]
(It thinks I am in Kendal square)
19:48:56 [deiu]
I see
19:49:07 [timbl]
Where is the data stored?
19:49:15 [deiu]
it should be stored where the app is
19:49:25 [deiu]
it creates a "location" file when you register
19:49:44 [deiu]
but it only publishes the location once you decide to share it
19:49:46 [timbl]
It made as file
19:49:54 [timbl]
but it is empty it seems
19:50:02 [timbl]
https://timbl.data.fm/test2/locator/location
19:50:22 [deiu]
yes, that's how it works
19:50:43 [deiu]
were you able to register the app? (it tries to add some triples to your profile)
19:50:49 [timbl]
I don't see how to share it
19:51:04 [timbl]
No, my public profile is not RWW
19:51:14 [deiu]
once authenticated, you should have the option to register the app
19:51:35 [deiu]
if it fails to add the triples, it should display them so you can copy/paste
19:51:48 [timbl]
well, I would be happy to have the triples stored in a pubs registry associated tithe app
19:51:57 [timbl]
it does display tem
19:52:04 [timbl]
but I am being a naive user :-)
19:52:36 [deiu]
yes, that's something I'm looking into, but you still have to link to the app somehow (from your profile)
19:53:19 [timbl]
Interestig security issue
19:53:42 [deiu]
I have published a draft vocabulary to describe web apps, but it's something to be discussed on the rww list
19:53:54 [timbl]
I want to be able to write to my public profile only when it is me and also it is a app i have decided to trust
19:54:33 [timbl]
Yes, in this case I would trust the app by loving on, and knowing it was made by someone friendly
19:54:39 [deiu]
I agree, and apps should have a manifest, explaining its usage and what access it requires from the user
19:55:12 [timbl]
Maybe I should give append-only access to my main profile
19:55:34 [timbl]
We wondered about adding append
19:55:38 [timbl]
to the ACL
19:55:48 [deiu]
yes, I would love that option in WAC!
19:56:21 [deiu]
it would allow for a "push" notification system
19:56:45 [timbl]
Ah, yes it is there
19:57:01 [deiu]
I'm feeling that we should really start a CG or a WG to standardize ACLs
19:57:12 [timbl]
The SPARQL update handler also could look at the request and check it had no delees
19:57:33 [timbl]
We need a set of tests for ACLs
19:57:35 [deiu]
ah yes, but you can't use normal HTTP verbs then
19:58:00 [deiu]
you need to make the difference between POST and PUT
19:58:12 [timbl]
In the original design, a POST of a data file ro a document would append to it
19:58:34 [timbl]
I think data.fm supported that
19:58:45 [deiu]
yes, so does rww.io
19:58:49 [timbl]
but i would prefer to use sparql update as that is what the code uses.
19:59:33 [deiu]
but if a client doesn't speak SPARQL? how can you set a rule that would only allow acl:Write for POSTs?
19:59:37 [timbl]
I have put a tracker app in there too
20:00:12 [timbl]
acl:Write means an operation which changes the file. I would not authorize that, only cal:Append
20:00:34 [deiu]
indeed, an :Append action would be great
20:00:40 [timbl]
the handler of a POST would check that either write or append were ok
20:00:53 [timbl]
Well, it is in th ontology
20:00:59 [timbl]
just code it up
20:00:59 [deiu]
Append?
20:01:19 [timbl]
Joe may have already
20:02:40 [deiu]
oh, is this change recent?
20:02:43 [timbl]
https://timbl.data.fm/test2/skin2.html is the tracker app .. not working very well but the key bit is the onboarding
20:02:54 [timbl]
append recent? no ages ago
20:02:56 [deiu]
(as in < 6 months old)
20:03:10 [deiu]
weird that I've missed it
20:03:51 [deiu]
this is great news!
20:03:56 [timbl]
1.3 (timbl 18-Jun-10): Append a rdfs:Class;
20:04:15 [deiu]
I can't explain why I haven't noticed it so far
20:04:33 [timbl]
the original ontology was april 09, appended added in Jun 2010
20:05:17 [deiu]
going back to the location tracking app, there is one important triple that is required to link the app to the user's profile
20:05:33 [deiu]
it helps with discovery
20:05:38 [timbl]
let me stick on those triples by hand
20:06:17 [deiu]
the registration is somewhat intelligent, so it should have proposed triples that match your current profile and install location
20:08:17 [deiu]
once registered, if you add my WebID to your foaf:knows list, it will display my name and picture in a list of "Friends who use this app"
20:09:21 [timbl]
your webid being ...?
20:10:09 [deiu]
https://my-profile.eu/people/deiu/card#me
20:10:32 [deiu]
sorry
20:11:50 [timbl]
card:i rdfs:seeAlso <http://dig.csail.mit.edu/2008/webdav/timbl/foaf.rdf
20:12:47 [timbl]
That is editable.
20:13:37 [deiu]
I'm not sure how the app find out about that
20:13:43 [deiu]
can find out*
20:13:47 [timbl]
I need to look at https://my-profile.eu/people/deiu/card#me in tabulator
20:13:58 [timbl]
Then I can say I know you
20:14:03 [timbl]
but I get redirected
20:14:14 [timbl]
to https://my-profile.eu/view?webid=https%253A%252F%252Fmy-profile.eu%252Fpeople%252Fdeiu%252Fcard%2523me#me
20:14:14 [deiu]
oh
20:14:26 [deiu]
that's some weird double encoding
20:15:04 [timbl]
It is an any rate my profile.eu deciding I need a HTML viewinsteadof rdf
20:16:01 [timbl]
when tabulator says it doe both and the profile is primarily a data file so you should give the data not the html
20:16:22 [timbl]
yes double encoding
20:17:37 [deiu]
it does a redirect if the Accept header does not contain turtle
20:20:00 [timbl]
Hmph
20:20:35 [RalphS]
RalphS has quit ()
20:21:56 [timbl]
I think tabulator sends text/xml,application/xml,application/xhtml+xml,text/html;q=0.5,text/plain,image/png,*/*;q=0.1,application/rdf+xml,text/n3,text/rdf+n3
20:22:13 [timbl]
So no text/turtle
20:22:26 [timbl]
That should get changed
20:22:42 [timbl]
and text/rdf+n3 removed
20:23:44 [deiu]
there's also something strange happening with myprofile
20:24:13 [deiu]
double encoding is done by apache during redirect
20:24:24 [deiu]
as the URI was already encoded
20:25:05 [deiu]
this used to work until today (update maybe?)
20:33:30 [timbl]
timbl has quit (Quit: timbl)
20:41:28 [deiu]
I get a 404 when I try to fetch my profile...weird
20:49:56 [scor]
scor has quit (Quit: scor)
21:17:47 [timbl]
timbl (~timbl@184.151.61.4) has joined #dig
21:19:11 [timbl]
wroks
21:19:15 [deiu]
great
21:21:50 [timbl]
Now the social pane says "deiu knows you (unconfirmed)." which is correct but it is missing "Confirm you know deiu" button
21:22:53 [deiu]
is there something I need to do?
21:28:43 [deiu]
oh, it shows that you are now using the app
21:29:01 [deiu]
I can see you in the list of people that I know and that are using the app
21:32:50 [deiu]
I've shared my location with you
21:35:49 [timbl]
I have some problem confirming I know you with webdav
21:38:25 [deiu]
it's ok, I'm leaving it on for the moment so you can check back later
21:39:12 [deiu]
basically, all it does is to set the ACL for the location file to the list of people with whom I chose to share my location
21:40:02 [deiu]
if I click "Stop sharing", it removes the geolocation triples from the file (as extra precaution)
21:40:52 [timbl]
Ok, cool, I hadn't seen the UI for choosing the set of people
21:43:12 [deiu]
you see that once the app has been successfully registered
21:57:18 [timbl]
(I have a problem it seems writing back using webdav, to be investigated later)
22:01:26 [timbl]
Ok, so I have added the triples to my public profilr.
22:01:41 [timbl]
Not sure how to know who I am logged in as
22:02:19 [timbl]
It is still asking me to register.
22:02:27 [timbl]
"Logout" doesn't seem to work
22:02:31 [timbl]
in locator
22:02:56 [timbl]
or maybe logs me right back in
22:04:25 [deiu]
logout is a bit of a work in progress, as it can't really guess what the application backend handles log outs
22:04:44 [deiu]
s/what/how
22:05:05 [deiu]
let me see...
22:06:49 [deiu]
ah right, so you've added it in a seeAlso
22:07:23 [timbl]
you are in France?
22:07:37 [deiu]
yes
22:07:39 [timbl]
not surprised tired
22:08:12 [deiu]
does rdflib automatically fetch seeAlso resources?
22:08:15 [timbl]
I should go get dinner and you should sleep :)
22:08:46 [deiu]
yes, I should..soon :)
22:09:00 [timbl]
dereferencing x involves automatically fetching y if it comes across x seeAlsoe y.
22:09:17 [timbl]
but not z if it comes acroess y seealso z.
22:09:38 [timbl]
so it won't folio a chain of things, only linked from x itself
22:09:47 [timbl]
(or anything sameAs x)'
22:10:08 [timbl]
It will also pull in rel=meta
22:10:22 [deiu]
so in your case, it should have fetched the webdav seeAlso, right?
22:10:55 [timbl]
yes
22:11:03 [timbl]
and it did and it knows it is editable
22:11:32 [timbl]
but it failed with an empty message when I tried to change the file
22:12:07 [timbl]
If I log in as me and look at you and click the box "you know diet".
22:12:11 [timbl]
in deiu knows you (unconfirmed).You and deiu confirm you know deiu.
22:12:11 [timbl]
You know deiu
22:12:12 [timbl]
You and deiu know 1 person found in common, Henry Story
22:12:15 [timbl]
___
22:12:29 [timbl]
that panel.
22:12:45 [deiu]
hmm
22:12:46 [timbl]
but I should go for dinner now
22:12:57 [deiu]
enjoy your dinner :)
22:13:02 [timbl]
don't worry about the webdav prob, it is probably code rot
22:13:09 [deiu]
we'll pick pick this out some other time
22:13:11 [timbl]
sleep well :)
22:13:14 [deiu]
thanks!
22:13:28 [deiu]
s/out/up
22:13:47 [deiu]
deiu has quit (Quit: Leaving)
22:23:11 [timbl]
timbl has quit (Ping timeout: 256 seconds)
22:50:52 [danbri]
danbri has quit (Ping timeout: 264 seconds)
22:56:37 [danbri]
danbri (~danbri@146.90.244.38) has joined #dig
23:25:08 [timbl]
timbl (~timbl@184.151.61.4) has joined #dig