IRC log of dig on 2015-01-28

Timestamps are in UTC.

01:41:20 [timbl]

timbl has quit (Quit: timbl)

02:55:52 [bblfish]

bblfish (~bblfish@p200300624776CA1669DFE9C21E884AE6.dip0.t-ipconnect.de) has joined #dig

03:02:42 [bergi]

bergi has quit (Remote host closed the connection)

03:09:59 [deiu]

deiu has quit (Quit: deiu)

03:17:15 [deiu]

deiu (~andrei@unaffiliated/deiu) has joined #dig

03:26:02 [bblfish]

bblfish has quit (Remote host closed the connection)

03:27:52 [bblfish]

bblfish (~bblfish@p200300624776CA1679B7670F77E3EE22.dip0.t-ipconnect.de) has joined #dig

03:30:36 [bblfish]

bblfish has quit (Remote host closed the connection)

04:47:16 [deiu]

deiu has quit (Quit: deiu)

05:26:13 [betehess]

betehess has quit (Ping timeout: 245 seconds)

06:25:36 [bblfish]

bblfish (~bblfish@p200300624776CA16C8C143DC9B5A365C.dip0.t-ipconnect.de) has joined #dig

06:30:18 [bblfish]

bblfish has quit (Ping timeout: 272 seconds)

07:42:24 [bblfish]

bblfish (~bblfish@p5DCBE626.dip0.t-ipconnect.de) has joined #dig

07:58:29 [bblfish]

bblfish has quit (Remote host closed the connection)

08:07:10 [bblfish]

bblfish (~bblfish@p5DCBE626.dip0.t-ipconnect.de) has joined #dig

09:08:20 [Sebastien-L]

Sebastien-L (~sebastien@2a01:e35:8b47:7ab0:b9f3:95e5:424e:4b2c) has joined #dig

12:01:47 [RalphS]

RalphS (rswick@w3cvpn1.w3.org) has joined #dig

12:06:00 [timbl]

timbl (~timbl@c-24-62-225-55.hsd1.ma.comcast.net) has joined #dig

12:16:30 [betehess]

betehess (~betehess@cpe-74-72-240-74.nj.res.rr.com) has joined #dig

12:43:42 [bblfish]

bblfish has quit (Remote host closed the connection)

12:47:09 [Sebastien-L]

Sebastien-L has quit (Ping timeout: 276 seconds)

13:17:35 [Sebastien-L]

Sebastien-L (~sebastien@2a01:e35:8b47:7ab0:b9f3:95e5:424e:4b2c) has joined #dig

13:18:24 [bblfish]

bblfish (~bblfish@p200300624776CAA9041233C5CAA2A76F.dip0.t-ipconnect.de) has joined #dig

13:53:21 [deiu]

deiu (~andrei@unaffiliated/deiu) has joined #dig

14:00:15 [betehess]

betehess has quit (Quit: Leaving)

14:15:01 [bblfish]

bblfish has quit (Remote host closed the connection)

14:16:32 [bblfish]

bblfish (~bblfish@p200300624776CAA97D093DE6D3894276.dip0.t-ipconnect.de) has joined #dig

14:21:11 [bblfish]

bblfish has quit (Ping timeout: 265 seconds)

14:23:38 [bblfish]

bblfish (~bblfish@p200300624776CAA97D093DE6D3894276.dip0.t-ipconnect.de) has joined #dig

15:02:30 [bblfish]

hi deiu

15:02:37 [bblfish]

sandro_

15:02:53 [sandro_]

Hi bblfish

15:02:58 [sandro_]

sandro_ is now known as sandro

15:03:26 [bblfish]

I sent a mail to the social_web_wg yesterday to see who would be interested in looking at an LDP API

15:03:46 [bblfish]

I suppose both of you would be :-)

15:04:07 [bblfish]

perhaps we should try to organise a bit

15:04:21 [bblfish]

pool our resources

15:04:40 [deiu]

Hi

15:04:49 [bblfish]

hi deiu

15:05:12 [sandro]

Yeah, I'm somewhat concerned that even deiu, timbl, and I wouldn't come close to agreeing on what that might look like.

15:05:24 [deiu]

I saw the email but I don’t really have the time to work on this at the moment

15:05:40 [sandro]

Have you done a swat0, bblfish ?

15:06:06 [bblfish]

We could get this done as a group I think quite easily

15:06:41 [bblfish]

But we don't need to build the swat0 to start with, we can look at how we could come to an agreement.

15:06:57 [bblfish]

the best way to do that is to meet up

15:07:38 [sandro]

in person?

15:08:20 [bblfish]

skype or some conf channel to work out what is needed should do

15:09:08 [betehess]

betehess (~betehess@static-72-69-24-242.nycmny.fios.verizon.net) has joined #dig

15:09:19 [sandro]

i'm doubtful tim would spend the time, but maybe

15:10:20 [sandro]

what do you mean about swat0? It seems like a reasonable place to start is to say how swat0 would be done with ldp. (it'd be nice to get it working, but for now I'm just saying spell it in a page of bullet points.)

15:11:11 [bblfish]

yes, I mean we don't need to build it before we get to an agreement. We can work out what we agree on and then build it :-)

15:12:23 [bblfish]

what we need to find is the group of LDP people who would be interested in working on this with us.

15:13:08 [bblfish]

We have the core team here I know. This is pretty important because if the Social Web WG moves to LDP then it's also the EU D-Cent project that moves that way, ie, we get a lot of coverage

15:13:18 [bblfish]

I think we can also get Arnaud Le Hors interested

15:13:25 [bblfish]

( I think he is interested )

15:14:45 [betehess]

betehess has quit (Quit: Leaving)

15:15:02 [betehess]

betehess (~betehess@static-72-69-24-242.nycmny.fios.verizon.net) has joined #dig

15:16:32 [bblfish]

DIGlogger: pointer?

15:16:32 [bblfish]

See http://dig.csail.mit.edu/irc/dig/2015-01-28#T15-16-32

15:19:23 [bblfish]

the e-mail I mentioned is here http://lists.w3.org/Archives/Public/public-socialweb/2015Jan/0065.html

15:19:41 [bblfish]

If we don't get together here we are going to loose a huge opportunity IMHO

15:29:24 [sandro]

I agree, bblfish - just not sure how to move forward. maybe we should each go sketch out how we'd do swat0

15:34:39 [bblfish]

one thing is that we have the Social Web API "requirements" that came from studying web 2.0 apis https://www.w3.org/wiki/Socialwg/Social_API/Requirements

15:35:08 [bblfish]

so we can fill that out with how that can be done with LDP, and what is missing

15:35:35 [sandro]

Yes, I'm currently drafting an email about it. But I think addressing that point-by-point is too much work to start with. swat0 shows much more clearly what we're trying to do.

15:36:11 [bblfish]

yes, I think we need both. The Social API gives us the big picture view, the Swat is a good use case.

15:36:20 [bblfish]

I suppose we can look at it now quickly

15:36:21 [sandro]

And my guess is if you ask you and deiu and timbl and me to do swat0, we'll get 4 completely different answers.

15:36:38 [bblfish]

http://www.w3.org/2005/Incubator/federatedsocialweb/wiki/SWAT0

15:37:04 [bblfish]

yes, of course. But that's what consensus building is about

15:37:17 [bblfish]

if you don't build consensus then you get incompatible versions

15:37:31 [bblfish]

if we try to synchronise we can come to an agreement.

15:37:45 [bblfish]

We already have LDP as a pretty strong building block

15:37:46 [betehess]

I think deiu and I would have a quite common answer ;-)

15:37:56 [timbl]

swar0?

15:37:57 [betehess]

s/common/similar/

15:38:01 [bblfish]

http://www.w3.org/2005/Incubator/federatedsocialweb/wiki/SWAT0

15:38:01 [timbl]

swat0?

15:38:38 [bblfish]

> With his phone, Dave takes a photo of Tantek and uploads it using a service

15:39:04 [bblfish]

so replace service with ldp:Container

15:39:12 [bblfish]

and uploading with POST

15:39:26 [bblfish]

then one just needs a relation to find the ldp:Container

15:39:36 [bblfish]

the phone needs to have that relation.

15:39:48 [bblfish]

> Dave tags the photo with Tantek

15:39:59 [sandro]

bblfish, what URL does Dave post to?

15:40:08 [sandro]

and how does he authenticate that?

15:40:12 [betehess]

I remember the use-case was going way beyond that (notifications, back-links, etc.)

15:40:27 [betehess]

it's not a simple system to describe

15:40:28 [bblfish]

betehess: I just wrote up the first two lines

15:40:31 [betehess]

ok

15:40:40 [sandro]

1. With his phone, Dave takes a photo of Tantek and uploads it using a service

15:40:40 [sandro]

2. Dave tags the photo with Tantek

15:40:40 [sandro]

3. Tantek gets a notification on another service that he's been tagged in a photo

15:40:40 [sandro]

4. Evan, who is subscribed to Dave, sees the photo on yet another service

15:40:40 [sandro]

5. Evan comments on the photo

15:40:40 [sandro]

6. David and Tantek receive notifications that Evan has commented on the photo

15:40:47 [bblfish]

thanks :-)

15:40:49 [betehess]

I haven't looked at swat0 for a long time (more than a year)

15:41:42 [bblfish]

Authorization will be taken care of by Web Access Control, which can be made to use any number of authentication protocols

15:41:56 [bblfish]

and I think LDP is interested in LDP 2 to look at that

15:42:13 [bblfish]

so we can delegate that to LDP2 and use what we have for the moment

15:42:42 [bblfish]

"what URL does Dave post to?" that's a good question. Somehoe the client has to find a link to the container. So we need to work out which link this would be

15:42:59 [bblfish]

> 3. Tantek gets a notification on another service that he's been tagged in a photo

15:43:10 [sandro]

I think a credible SWAT0-sketch needs to cover all that, bblfish. One thing this illustrates is where LDP 1.0 falls short.

15:43:20 [betehess]

I think it's a bit too easy to say "oh yeah, just use webacl". If you can't convince anybody that webacl is the solution, then your whole solution for swat0 might be useless

15:43:30 [betehess]

+1 to what sandro said

15:43:59 [bblfish]

Alex are you trying to help here or are you just trying to undermine the discussion?

15:44:18 [sandro]

Like, when Dave is POSTing the picture, is it public? If so, how does he say so during the POST?

15:44:19 [bblfish]

I am just trying to point to things we have here

15:44:50 [bblfish]

we can edit ACLs just like we edit other LDP resources with PATCH

15:45:25 [betehess]

no, I am just saying that you need to include all the technologies, and LDP is far from being enough. That means you need to convince people to adopt a lot of technoologies they don't know or don't like, or this will fail

15:45:45 [bblfish]

I know LDP is just a really good building block

15:46:22 [bblfish]

true. We may not be able to convince everyone, but I was of the understanding that this group in particular has a lot in common

15:46:26 [sandro]

Moving on to: 2. Dave tags the photo with Tantek

15:46:35 [bblfish]

also LDP 2 wants to do Web Access Control

15:47:20 [sandro]

do we do that via a POST of a new resource which says <photo> foaf:depicts <Tantek>, or with a PUT/PATCH of the metadata of <photo> ?

15:47:21 [bblfish]

ok Arnaud Le Hors mentioned that the Annotations working group was interested in using LDP

15:47:22 [betehess]

anyway, have fun! (I am not even sure why I started commenting here :-)

15:47:39 [bblfish]

thanks :)

15:47:56 [bblfish]

so for annotations we should probably see what they would do.

15:48:30 [bblfish]

so for notification we can develop something more semantic along the lines of http://bblfish.net/tmp/2011/05/09/

15:48:43 [bblfish]

it's reasonably easy to see how to do that.

15:49:08 [bblfish]

you have a relation from a profile to an LDPC notification container

15:49:28 [sandro]

or just use webmention

15:49:47 [sandro]

which is arguable the same thing, if you squint, and make that relation "webmention"

15:50:28 [bblfish]

ok, so we can just see if we can LDPCize webmention

15:50:38 [bblfish]

do you have a link for WebMention?

15:52:24 [sandro]

http://indiewebcamp.com/Webmention

15:52:35 [sandro]

I

15:53:04 [sandro]

I'm sketching out swat0 in ldp, with lots of issues. I guess I'll put it on the socweb wiki?

15:53:21 [bblfish]

ok, sounds good

15:53:24 [bblfish]

or github

15:53:42 [bblfish]

wiki is easier for the moment

15:54:26 [sandro]

any obvious URL to use?

15:54:46 [bblfish]

SwatLDP ?

15:57:04 [bblfish]

ah here is the access control document from LDP https://dvcs.w3.org/hg/ldpwg/raw-file/default/ldp-acr.html

15:57:49 [bblfish]

I wonder if that has been put forward to the W3C yet

15:58:09 [sandro]

What does that mean?

15:59:29 [bblfish]

well that is meant to be a document from the LDP WG for the next LDP. I suppose it has to be used at some point. Perhaps for LDP next?

15:59:45 [bblfish]

and I suppose we have not yet reached that point yet

16:00:17 [sandro]

page started https://www.w3.org/wiki/Socialwg/LDP-SWAT0

16:00:59 [sandro]

http://www.w3.org/TR/ldp-acr/

16:01:07 [bblfish]

ok, nice.

16:01:26 [deiu]

deiu has quit (Quit: deiu)

16:01:30 [bblfish]

( for both links )

16:01:59 [bblfish]

ok, so how would I fill that wiki out sandro?

16:02:16 [bblfish]

I suppose I need space for "potential answer"

16:03:01 [sandro]

Maybe: "bblfish: <your proposed resolution>"

16:03:03 [sandro]

?

16:03:06 [bblfish]

ok.

16:03:45 [sandro]

are you going to do that right now?

16:03:54 [sandro]

or should I fill it out some more first?

16:04:27 [bblfish]

I suppose the tricky bit is that wiki's are not very good for simultaneous editing

16:04:46 [sandro]

right. maybe give me 30 more minutes, then I'll go do something else for a bit?

16:04:54 [bblfish]

ok

16:05:10 [bblfish]

In 1. I suppose we can agree that the URL is an ldpc

16:05:47 [bblfish]

in fact a basic container would do

16:06:07 [bblfish]

then the issue is: how does Dave's phone know where to post it?

16:06:55 [bblfish]

I suppose the simple answer is that the phone knows Dave's WebID and there is a link there to a number of photo containers that Dave Uses. Dave either chooses one of them at each upload or has a default

16:07:33 [bblfish]

s/there is a link there/there is a link from his webid profile /

16:08:36 [bblfish]

Issue 2. How to set access: one could imagine that the access is already set for that container in advance - and all the containees of the container have the same access restrictions

16:18:48 [bblfish]

I suppose we could then also show LDPSwat to the LDP group to get their feedback, and also to get them interested in what the Social Web WG is doing

16:18:48 [bblfish]

/me Going to check on baby Anaïs quickly, will be back

16:19:02 [sandro]

say hi for me :)

16:21:28 [deiu]

deiu (~andrei@unaffiliated/deiu) has joined #dig

16:22:34 [sandro]

I think https://www.w3.org/wiki/Socialwg/LDP-SWAT0 is stable enough that others could add lines now without an edit conflict

16:24:09 [deiu]

deiu has left #dig

16:24:13 [deiu]

deiu (~andrei@unaffiliated/deiu) has joined #dig

16:32:21 [bblfish]

thanks a lot sandro :-)

16:36:44 [sandro]

I'm mentioning it in a reply to your email

16:44:27 [bblfish]

I gave a few answers to the first part https://www.w3.org/wiki/Socialwg/LDP-SWAT0#To_what_URL.3F

16:44:40 [bblfish]

Perhaps we should make the questions stand out a bit more

16:47:12 [sandro]

can i post a reformatting of your answer?

16:47:31 [bblfish]

yes certainly

16:47:37 [bblfish]

I'll fill in the next bit

16:55:00 [sandro]

i rearranged "To what URL" -- I think it's somewhat clearer, and keeps the meaning, I hope.

16:56:01 [bblfish]

"app just posts to the user's URL" that's not very clear. Perhaps "to one of the users image containers" ?

16:59:55 [sandro]

fixing

17:01:48 [sandro]

fixed

17:01:52 [sandro]

I hope that's clear now

17:06:51 [bblfish]

mhh. That's weird. You want to post a picture to the users profile page?

17:07:18 [bblfish]

quoting: "sandro: app just posts to the URL which the user used in authentication ( eg Dave's http://dave.example/ or https://dave.service.example/ or http://service.example/dave )"

17:07:37 [melvster]

I think kingsley has already implemented swat0+ldp but didnt have anyone else to test with

17:08:12 [bblfish]

yeah, we were too early at the time. We needed a bunch of people to be able to play with, and a structure like a WG to be able to help build consensus

17:08:40 [melvster]

i could probably implement it from the command line, if that helps at all

17:09:26 [melvster]

no having to worry about cross origin stuff, proxies, persistence etc.

17:09:31 [sandro]

yeah, right now, we need to surface HOW those implementations work

17:09:45 [bblfish]

yep

17:10:31 [bblfish]

btw. am I correct in understanding that Activity Streams 2.0 is just an outgrowth of Atom Syndication format and the other RSS formats?

17:10:42 [bblfish]

json-ld version of those?

17:11:09 [sandro]

That wasn't my impression, but I really don't know

17:13:24 [bblfish]

http://www.w3.org/TR/activitystreams-core/

17:13:33 [melvster]

btw working from command line 3 other advantages 1. it knows my webid (saved) 2. it implicitly has authenticated the user 3. I can get realtime notifications of file changes etc.

17:14:47 [melvster]

and I can easily do this too: 'This can have a relation ( to be defined ) to one or more LDPCs to which the client can post a picture'

17:14:58 [melvster]

I just need to know the predicate

17:15:14 [bblfish]

melvster: yes curl is a good way to do initial testing of the service see https://github.com/read-write-web/rww-play/wiki/Curl-Interactions

17:16:26 [melvster]

bblfish: thanks, that's what I do now, tho wrapping curl in shell scripts for convenience, end goal is to run in the browser also

17:17:29 [melvster]

useful, I should wrap all these examples

17:19:16 [bblfish]

mhh mistakes in the Turtle <http://example.org/blog/> a <"urn:example:types:blog">

17:19:24 [melvster]

bblfish, sandro: re the swat0 wiki page ... if using tantek is the normative example, that's kind of hard, when we dont know when we're talking about tantek the homepage, or tantek the person ...

17:19:58 [melvster]

or should we use the tantek profile as a 'typical web 2.0 profile page' ?

17:20:01 [melvster]

then add #this ?

17:20:10 [melvster]

or #webid

17:20:31 [bblfish]

ok, I think activity streams is meant to be an evolution of Atom syntax

17:21:01 [bblfish]

in which case we could use that for the subscription format

17:23:21 [sandro]

melvster, personally, I'm done with #. I suggest keeping the wiki page neutral about that stuff.

17:23:37 [sandro]

but do what you have to, to get something that works, I suppose

17:24:32 [melvster]

sando: that's the thing ... it's ambiguous so without making a decision out of band, you cant have consistent identifiers, which may mean the tests failing

17:25:32 [sandro]

Yeah, so sketch out LDP-SWAT0-proposal1 with all the details you need?

17:27:05 [betehess]

sandro, I'll be at MIT tomorrow. do I have a chance to see you?

17:27:53 [sandro]

yes, I'm free at lunch, at least

17:28:02 [betehess]

well, I'll be with deiu :-)

17:28:09 [betehess]

so probably around the whole day

17:29:33 [melvster]

sandro: ok ... most interesting part to me is the notification ... id like to know how to do that for payments too

17:30:40 [sandro]

betehess, deiu might have some conflicts he might not know about yet :-]

17:31:01 [sandro]

betehess, (I only told him loosely we had time Thursday to talk to Tim)

17:31:01 [deiu]

oh? :)

17:31:04 [betehess]

oh I didn't plan to work with him all day :-)

17:31:10 [sandro]

oh, okay

17:31:15 [deiu]

we’re not working together

17:31:20 [sandro]

gotcha

17:31:20 [betehess]

I hope to have time for all of you guys!

17:31:21 [deiu]

not this time at least

17:31:33 [betehess]

well, we have to update LD Patch

17:31:50 [betehess]

we could plan 1 hour for that maybe

17:31:51 [betehess]

anyway

17:31:54 [betehess]

I have to leave now

17:32:09 [betehess]

the bus station will be crazy after this snowstorm

17:32:16 [deiu]

yeah

17:32:26 [deiu]

I’m going to be at MIT today until you get here

17:32:51 [betehess]

betehess has quit (Quit: Leaving)

17:32:56 [sandro]

melvster, I know of three general approaches. (1) folks publish the URL of where you can post notifcations for them, (2) webmention, and (3) people create standing queries for the stuff that interests them. I think (1) and (2) can probably be merged into on thing, with a little work.

17:33:14 [bblfish]

yes

17:34:03 [bblfish]

what is cimba? pointer?

17:34:17 [melvster]

http://cimba.co

17:37:01 [bblfish]

mhh is there something that explains how it works?

17:37:09 [melvster]

yes

17:37:16 [melvster]

one sec

17:37:27 [bblfish]

ah it uses WebID :-)

17:37:32 [bblfish]

let me see if mine is up to date

17:39:02 [bblfish]

cymba is taking a lot of time calculating

17:39:21 [melvster]

bblfish: https://fcns.eu/docs/crosscloud-cimba.pdf

17:41:41 [bblfish]

thanks

17:44:53 [bblfish]

ok, it's looking pretty good

17:50:27 [bblfish]

I'll try to think about how Activity Streams fits in.

17:50:40 [bblfish]

but this exercise was definitely worth the effort.

17:59:47 [bblfish]

bblfish has quit (Remote host closed the connection)

18:04:49 [bblfish]

bblfish (~bblfish@p200300624776CAA9505A05AF85C68C10.dip0.t-ipconnect.de) has joined #dig

18:05:41 [melvster]

i should start running GOLD locally via a freedombox

18:06:13 [melvster]

anyone know a good dyndns solution?

18:07:06 [bblfish]

bblfish has quit (Remote host closed the connection)

18:14:05 [scor]

scor (scor@drupal.org/user/52142/view) has joined #dig

18:52:01 [bblfish]

bblfish (~bblfish@p5DC79F88.dip0.t-ipconnect.de) has joined #dig

18:52:24 [bblfish]

DIGlogger: pointer?

18:52:24 [bblfish]

See http://dig.csail.mit.edu/irc/dig/2015-01-28#T18-52-24

18:56:12 [bblfish]

Ok, I wonder if we can send that page to the LDP WG.

19:02:30 [bblfish]

I'll wait a little

19:06:56 [bblfish]

bblfish has quit (Remote host closed the connection)

19:09:01 [deiu]

deiu has quit (Quit: deiu)

19:09:20 [deiu]

deiu (~andrei@unaffiliated/deiu) has joined #dig

19:11:14 [deiu]

deiu has quit (Read error: Connection reset by peer)

19:11:20 [deiu_]

deiu_ (~andrei@unaffiliated/deiu) has joined #dig

19:23:18 [scor]

scor has quit (Quit: scor)

20:11:21 [scor]

scor (~scor@c-98-216-67-164.hsd1.ma.comcast.net) has joined #dig

20:11:21 [scor]

scor has quit (Changing host)

20:11:21 [scor]

scor (~scor@drupal.org/user/52142/view) has joined #dig

20:35:01 [scor]

scor has quit (Quit: scor)

20:36:28 [melvster]

when you post to an LDPC you get back Location: Link: rel="acl" and Link: rel="meta"

20:36:44 [melvster]

what's the rel="meta" for? ... it's the same as Location it seems ...

20:37:07 [sandro]

I don't think rel="meta" is in the spec

20:37:32 [melvster]

sandro: i get it back from Cimba.co ... is it extraneous, do you know?

20:37:58 [sandro]

it might come in to play if the resource is an image or something that cant include RDF

20:38:10 [melvster]

ah ok, didnt think of that case, thanks

20:38:21 [sandro]

that's just an educated guess

20:39:05 [melvster]

sandro: interesting you should say that as uploading a photo is the first part of LDP-SWAT0

20:39:13 [sandro]

:)

20:39:49 [melvster]

btw it authenticated me automatically using the cookie (which I copied to curl)

20:40:06 [melvster]

normally these days you can right click from a browser and get the curl line

20:40:45 [sandro]

? I didn't think cimba used cookies. I thought it only used webid for auth.

20:41:19 [melvster]

sandro: I seem to have a cookie set, I know it uses WebID at least to sign in, but may use a cookie after that, which I think makes sense

20:41:31 [melvster]

I can test it

20:42:02 [melvster]

'sandro: app just posts to the URL which the user used in authentication ( eg Dave's http://dave.example/ or https://dave.service.example/ or http://service.example/dave )'

20:42:27 [melvster]

so I post to an LDPC ... with the photo, I presume the created resource will come back in Location:

20:42:51 [melvster]

at this point it seems just to be a generic photo

20:43:00 [melvster]

that we know nothing about

20:43:16 [sandro]

Right.

20:43:37 [melvster]

so cimba already handles the acl using rel='acl' and it does a PUT on that

20:43:43 [sandro]

I think some metadata, like the acl, and maybe the tagging, can/should be sent along with the photo, in Link headers.

20:43:47 [sandro]

That sounds about right.

20:44:01 [melvster]

im just looking at the tagging

20:44:31 [melvster]

"{ photo foaf:depicts: tantek } "

20:45:02 [sandro]

of course we run into httpRange-14 again on tantek. :-(

20:45:33 [sandro]

for now I guess use tantek.com/#this ?

20:45:46 [melvster]

sandro: funny you should bring that up ... I had a long conversation with sandeep who created "web mention" about that -- it's popular with the indie web folks

20:46:05 [melvster]

it was hard to explain HR14

20:46:26 [sandro]

yes, and working with the indieweb folks is one of the reasons I've given up on it.

20:46:36 [melvster]

but then I said to him ... "I may like ricky martin's home page, but may not like ricky martin" -- and he totally got it :)

20:46:44 [sandro]

but I haven't convinced timbl, etc, yet.

20:47:31 [sandro]

getting it in a conversation like that is entirely different, though, that getting it throughout the architecture, and having EVERYONE get it.

20:47:48 [melvster]

yes very misunderstood issue

20:48:05 [sandro]

so, IMHO we should just give up and stick with indirection identification.

20:48:38 [sandro]

it works perfectly well. it gets complicated only when you want to do complicated things, like Ricky Martin but not his profile page.

20:48:46 [melvster]

what do you mean by indirection indentification?

20:49:03 [melvster]

oh indirect identifiers

20:49:06 [melvster]

as per awww

20:49:26 [sandro]

just use http://www.w3.org/People/Berners-Lee/ as the URL (indirectly) for Tim.

20:49:43 [sandro]

I need to run. bbiab.

20:49:50 [melvster]

np cu! :)

20:50:45 [Sebastien-L]

Sebastien-L has quit (Ping timeout: 276 seconds)

21:08:19 [bblfish]

bblfish (~bblfish@p200300624776CAA9F5B9590BC4A14FAB.dip0.t-ipconnect.de) has joined #dig

21:13:41 [GitHub20]

[13rdflib.js] 15deiu pushed 3 new commits to 06master: 02http://git.io/FzHp

21:13:41 [GitHub20]

13rdflib.js/06master 14107113a 15deiu: Revert "Replaced deprecated nodeValue with value."...

21:13:41 [GitHub20]

13rdflib.js/06master 149c1dfab 15deiu: Revert "Added missing typeof check for tabulator."...

21:13:41 [GitHub20]

13rdflib.js/06master 14cef075f 15deiu: Merge branch 'master' of github.com:linkeddata/rdflib.js

21:16:50 [GitHub192]

[13rdflib.js] 15deiu pushed 1 new commit to 06master: 02http://git.io/Fz7R

21:16:50 [GitHub192]

13rdflib.js/06master 1439246fd 15deiu: Added missing typeof check for tabulator

21:28:18 [bblfish]

bblfish has quit (Remote host closed the connection)

21:50:12 [bblfish]

bblfish (~bblfish@p200300624776CAA9790C2B121594E97A.dip0.t-ipconnect.de) has joined #dig

22:13:00 [deiu_]

timbl: I hope you haven’t updated rdflib.js the past few days, there was a bad bug in it (all fixed now)

22:46:41 [klip]

bblfish, sandro: JFI there exists another approach to notifications: AKSW's incubator contains http://dssn.org/ where they are using SemanticPingback http://aksw.org/Projects/SemanticPingback.html There is a PHP impl. http://aksw.org/Projects/Xodx.html

22:47:45 [sandro]

klip, are you arguing that this is what should be in the standard for the SocialWeb WG? I know of dozens of protocols. Not terribly interested in more, frankly.

22:48:03 [sandro]

aka, what makes this one better than all the others?

22:48:37 [melvster]

sandro: I thought semantic pingback was the established technique ... what other ones are there?

22:49:26 [melvster]

unless you mean webmention which is neat, but not particularly webby

22:49:39 [sandro]

Not terribly webby?

22:49:48 [melvster]

semantic webby :)

22:50:01 [sandro]

You mean RDFy?

22:50:13 [melvster]

yes

22:50:16 [sandro]

I think RDFy is a liability here, not an asset.

22:50:27 [sandro]

Unless it makes really good use of RDF

22:51:06 [sandro]

Pingback is a forgotten failure, so how does RDF make it good again?

22:53:15 [melvster]

just uses URIs from two parties and a message, which is extensible ... then you can find an endpoint with pingback : to ... deiu_ implemented it in my profile and it worked pretty well, kingsley also implements it ... I have one in my profile ... not really sure if there are any alternatives out there of standards quality?

22:54:10 [deiu_]

melvster: honestly..it sucks :)

22:54:25 [melvster]

deiu_: well what's better?

22:54:35 [deiu_]

nothing, that’s the problem

22:55:31 [sandro]

arguably webmention is at least (1) simpler, and (2) makes some efforts to deal with spam, and (3) isn't past its expiration date

22:55:58 [sandro]

I don't think (2) is good enough yet, though.

22:56:53 [melvster]

spam is an interesting problem, proof of work used in bitcoin was originally used as an anti spam technique

22:58:15 [melvster]

im not sure how widely deployed webmention is, doesnt pingback have more traction?

22:59:20 [sandro]

I think pingback has been dead for many years now, killed by spam.

22:59:34 [deiu_]

wordpress still uses it

22:59:46 [sandro]

webmention hasn't caught on, and they're trying to make sure it can handle spam before it does.

23:00:10 [sandro]

Hm. I haven't seen a trackback/pingback on a post in years.

23:01:04 [melvster]

we get some occasionally on community group blog posts (I think wordpress)

23:01:34 [klip]

sandro: I'm just passing information related to notifications as you were talking about available solutions. That's why I wrote "JFI" at the beginning :)

23:02:20 [melvster]

semantic pingback is linked at the bottom of the webmention spec with pingback and trackback

23:03:19 [sandro]

re pingback and trackback in WP, here's a debate (in the comments) about whether to take them out of the code, since few people use them any more.... http://www.elegantthemes.com/blog/resources/wordpress-trackbacks-and-pingbacks

23:03:58 [sandro]

understood, klip, sorry I reacted too negatively. (it's just all a bit much sometimes :-)

23:05:46 [melvster]

O_o https://www.w3.org/wiki/Socialwg/API/Layers

23:09:25 [melvster]

"SWAT0 is a terrible use-case for this discussion" -- Evan

23:09:32 [melvster]

hehe what I said 3 years ago! :P

23:09:54 [melvster]

but I actually think it's not too hard to implement, interesting bit seems to be notifications

23:12:08 [melvster]

http://www.w3.org/wiki/Socialwg/Social_API/OpenSocial_Activity_Streams_API

23:12:46 [melvster]

Read

23:12:47 [melvster]

GET "/activitystreams/" User-Id "/" Group-Id [ "/" App-Id [ "/" (Activity-Id / Array<Activity-Id>) ] ]

23:13:01 [melvster]

I guess opacity is not in vogue in 2015 :)

23:19:21 [melvster]

minutes are interesting

23:19:29 [melvster]

good points from bblfish

23:19:46 [melvster]

ldp does lack notifications

23:20:22 [melvster]

so they only just reviewed the facebook api last week ... I have to say I quite like the facebook social framework and api, they even support webid

23:20:50 [melvster]

if activity streams can be made to do what facebook can do that'd be grat

23:20:54 [melvster]

s/grat/great

23:21:16 [melvster]

at one point facebook were saying that they would adopt open social if it got mature ... but that never really happened

23:25:43 [melvster]

i might actually write a browser as a front end to my command line stuff

23:26:06 [melvster]

turns out it's super easy to add a webkit style browser to an ubuntu app

23:28:33 [melvster]

sandro: re your facebook usecase: excellent!

23:30:47 [sandro]

thanks

23:33:33 [klip]

np, sandro :)

23:36:22 [bblfish]

bblfish has quit (Remote host closed the connection)