18:01:56 RRSAgent has joined #social 18:01:56 logging to http://www.w3.org/2014/12/02-social-irc 18:01:58 RRSAgent, make logs public 18:01:58 Zakim has joined #social 18:02:00 Zakim, this will be SOCL 18:02:00 ok, trackbot; I see T&S_SOCWG()1:00PM scheduled to start 2 minutes ago 18:02:01 Meeting: Social Web Working Group Teleconference 18:02:01 Date: 02 December 2014 18:02:03 mechanic has joined #social 18:02:08 zakim, who's on the phone? 18:02:08 T&S_SOCWG()1:00PM has not yet started, Arnaud 18:02:10 On IRC I see mechanic, Zakim, RRSAgent, nicolagreco, jasnell, tiborKatelbach, AnnBassetti, evanpro, Arnaud, tilgovi, bblfish_, Morbus, shepazu, Loqi, nickstenn, oshepherd, 18:02:10 ... cwebber2, rektide_, dwhly, kylewm, pdurbin, rhiaro, jaakko, wilkie, Tsyesika, mattl, bret, bigbluehat, JakeHart, jessica_lily, hadleybeeman, aaronpk, trackbot, sandro, wseltzer 18:02:21 zakim, who's on the phone? 18:02:21 T&S_SOCWG()1:00PM has not yet started, Arnaud 18:02:23 On IRC I see mechanic, Zakim, RRSAgent, nicolagreco, jasnell, tiborKatelbach, AnnBassetti, evanpro, Arnaud, tilgovi, bblfish_, Morbus, shepazu, Loqi, nickstenn, oshepherd, 18:02:23 ... cwebber2, rektide_, dwhly, kylewm, pdurbin, rhiaro, jaakko, wilkie, Tsyesika, mattl, bret, bigbluehat, JakeHart, jessica_lily, hadleybeeman, aaronpk, trackbot, sandro, wseltzer 18:02:29 zakim, this is socl 18:02:29 ok, Arnaud; that matches T&S_SOCWG()1:00PM 18:02:45 zakim, who's on the phone? 18:02:45 On the phone I see ??P0, tiborKatelbach, jasnell, Arnaud, evanpro 18:03:09 oh I'm P0, one sec 18:03:13 ??P0 is me 18:03:23 hm, how do I do that again :) 18:03:28 Zakim: ??P0 is me 18:03:32 Zakim, P0 is cwebber2 18:03:32 sorry, jessica_lily, I do not recognize a party named 'P0' 18:03:32 Zakim, ??P0 is cwebber2 18:03:33 +cwebber2; got it 18:03:35 oh 18:03:37 + +1.408.455.aaaa 18:03:40 hm :P 18:03:41 +[IPcaller] 18:03:42 heh, thanks :) 18:03:51 Zakim, IPcaller is me 18:03:51 +wilkie; got it 18:03:52 Zakim, botsnack 18:03:53 I don't understand 'botsnack', cwebber2 18:03:55 +??P8 18:03:56 whaat 18:03:57 Zakim, who's on the call? 18:03:57 On the phone I see cwebber2, tiborKatelbach, jasnell, Arnaud, evanpro, +1.408.455.aaaa, wilkie, ??P8 18:03:58 KevinMarks has joined #social 18:04:09 Zakim, ??P8 is me 18:04:09 +rhiaro; got it 18:04:16 I'm on the call but muted as I'm driving 18:04:20 I can scribe if my connection stays good 18:04:32 scribenick rhiaro 18:04:39 Can't promise, it's been sketchy lately 18:04:56 +??P11 18:05:06 Zakim, ??P11 is me 18:05:06 +jessica_lily; got it 18:05:08 +bblfish 18:05:09 Zakim, mute me 18:05:09 jessica_lily should now be muted 18:05:26 https://www.w3.org/wiki/Socialwg/2014-11-25-minutes 18:05:36 evanpro: Review minutes from last week 18:05:54 ... Objections or corrections? 18:06:02 +Ann 18:06:16 ... Minutes approved 18:06:19 RESOLVED: approved minutes of 25 November 18:06:48 TOPIC: Actions and issues tracker 18:07:01 http://www.w3.org/Social/track/issues/open 18:07:09 evanpro: Checking open issues for any that need to be addressed 18:07:24 ... No names on issues, checking actions 18:07:35 +Doug_Schepers 18:07:36 ... A few for harry, mostly about pubsubhubbub and licensing 18:07:45 Zakim, who's on the call? 18:07:46 On the phone I see cwebber2, tiborKatelbach, jasnell, Arnaud, evanpro, +1.408.455.aaaa, wilkie, rhiaro (muted), jessica_lily (muted), bblfish, Ann, Doug_Schepers 18:08:17 ... one for me, to update review of Foursquare API. Plan to do for next week. 18:08:21 http://www.w3.org/Social/track/issues/3 is covered by the extended vocabulary proposal ... if we adopt that then this issue will be closed 18:08:28 ... A couple for jasnell 18:08:39 ... actions 8 and 9, one about social APIs due today 18:08:49 jasnell: Will have to be pushed back 18:08:56 ... Part in the wiki already, but lots of work 18:09:01 ... Another week or two 18:09:09 evanpro: Not just opensocial? Link to spec? 18:09:24 jasnell: Connections has its own atom based API, more complex than open social 18:09:32 ... In the wiki, linked to developer docs, but needs summarising 18:09:40 evanpro: Presentation about this next week? 18:09:46 jasnell: Tentative 18:09:52 +Sandro 18:09:58 ... Could give very high level, would take lots of time for detail 18:10:15 evanpro: jasnell to review next week 18:10:26 ... That's it for open actions, still some work to do. 18:10:31 http://www.w3.org/Social/track/actions/9 is covered by the extended vocabulary proposal 18:10:36 ... Any other comment on actions and issues in tracker? 18:10:54 TOPIC: Face-to-face meeting 18:11:03 evanpro: Doodle has been open for a while 18:11:16 ... Were close to consensus, but messed it up 18:11:27 ... Can't do beginning of Feb, no chairs 18:11:37 ... Looking like 10-19 March 18:11:37 jaywink has joined #social 18:11:49 ... But not everyone has filled out the doodle yet 18:11:59 ... sandro, mark, owen, a few others 18:12:05 Zakim, who's here? 18:12:05 On the phone I see cwebber2, tiborKatelbach, jasnell, Arnaud, evanpro, +1.408.455.aaaa, wilkie, rhiaro (muted), jessica_lily (muted), bblfish, Ann, Doug_Schepers, Sandro 18:12:09 On IRC I see jaywink, KevinMarks, mechanic, Zakim, RRSAgent, nicolagreco, jasnell, tiborKatelbach, AnnBassetti, evanpro, Arnaud, tilgovi, bblfish, Morbus, shepazu, Loqi, nickstenn, 18:12:09 ... oshepherd, cwebber2, rektide_, dwhly, kylewm, pdurbin, rhiaro, jaakko, wilkie, Tsyesika, mattl, bret, bigbluehat, JakeHart, jessica_lily, hadleybeeman, aaronpk, trackbot, 18:12:09 ... sandro, wseltzer 18:12:11 ... Don't know if we should pick a time without tantek on the call 18:12:24 sorry, those times are all fine with me 18:12:33 MarkCrawford has joined #social 18:12:33 ... Deferring to Arnaud. Should we pick a time today? 18:12:48 ... with the risk of losing tantek? Or defer until next week? 18:12:56 Arnaud: We can wait a bit more 18:12:59 AdamB has joined #social 18:13:05 ... We need a good chance of having as many people as possible 18:13:06 +aaronpk 18:13:19 ... Could people take the time to update the poll? 18:13:29 ... But no reason to rush a decision right now 18:13:50 +AdamB 18:13:52 http://doodle.com/3cfvrfqr6n9cgdzs btw 18:13:54 ... We need to decide 8 weeks ahead of time 18:14:02 + +1.703.670.aabb 18:14:04 - +1.703.670.aabb 18:14:07 ... Shouldn't lose track, need to keep the pressure on 18:14:17 evanpro: Slipping past the 8 week for the end of Feb anyway 18:14:28 + +1.703.670.aacc 18:14:30 ... Mid March looks best, we can decide by january 18:14:45 that week would be best for me for sure 18:14:59 (March 17-19) 18:15:04 ... postpone the decision for another week. People, please update the doodle! Or risk being unable to come to the event. 18:15:27 was location discussed yet? 18:15:37 TOPIC: Social API review: OpenSocial Activity Streams API 18:15:59 thanks! 18:16:15 evanpro: We're looking at different APIs for social networks, and doing in-depth analyses with the intention of bringing out requirements 18:16:26 ... We did twitter a couple of weeks ago, foursquare and facebook last week 18:16:51 ... Today, OpenSocial AS API is one of the inputs for the working group, and because it's very different from the others, will go over it today 18:16:51 https://www.w3.org/wiki/Socialwg/Social_API/OpenSocial_Activity_Streams_API 18:17:03 ... Spend 5 minutes going through it now 18:17:16 ... The AS API for OpenSocial is one of the services that's part of the OpenSocial suite of APIs 18:17:49 ... Core activities, use other APIs 18:18:09 ... Provides a CRUD interface for AS events 18:18:33 ... Supports XML representation that's a direct mapping of JSON from AS 1.0 to XML elements 18:18:41 ... Not sure about level of support, but interesting 18:18:47 - +1.408.455.aaaa 18:19:02 ... OpenSocial uses OAuth 1.0; there are other mechanisms for auth too 18:19:15 ... complicated, but decoupled auth from API endpoints 18:19:25 ... Entities within OS API map directly to AS 1.0 18:19:34 + +1.408.455.aadd 18:19:41 ... An activity which is an event, has subject-verb-object 18:19:52 ... Activities can reference objects, which are nouns. ie people, images, places, groups 18:19:54 (CRUD: Create Read Update Delete) 18:19:58 ... Or extended, self-defined objects 18:20:09 ... API has four endpoints, not all required, two required (create and read) 18:20:26 ... Interesting in that they're similar to how atompub manages creation of atom entries 18:20:39 ... Create endpoint, that you post a new activity to 18:20:49 ... Read endpoint, can be the same as post endpoint, or different 18:20:54 ... Can return a stream or a single activity 18:21:09 ... Complicated because you can change which activities are returned based on url parameters 18:21:23 ... can get back a user's activities or a group's activities, or activities for a particular application (eg a game) 18:21:34 ... Can also filter down to an activity ID 18:21:43 ... Update endpoint, using PUT 18:21:49 ... should update an activity. optional. 18:21:52 ... Also Delete 18:22:05 ... Introspection endpoint, which returns fields for AS container 18:22:15 ... Overall, very different from the other ones we've talked about 18:22:31 ... Not a lot of access to other parts of social network like friends or followers list, or getting profile information about a user 18:22:36 ... Really about activity lifecycle 18:22:44 ... Direct and simple 18:22:59 ... Doesn't give a lot of information for reading 18:23:09 ?q 18:23:11 ... No concept of inbox, to get activities by people you follow or groups 18:23:17 ... Really about CRUD lifecycle 18:23:21 ack cwebber2 18:23:22 q? 18:23:26 q+ 18:23:41 cwebber2: Curious about whether supports addressing; privacy aspect of who something is addressed to; public or directed 18:23:44 there is some addressing support included but it's not based on the to/cc/bcc stuff 18:23:53 dromasca has joined #social 18:23:59 evanpro: Has extension to AS 1.0 for addressing 18:24:12 ... A way within the activity object to say it's to a particular audience 18:24:27 ACTION: Link to OpenSocial extensions to Activity Streams 1.0 18:24:27 Error finding 'Link'. You can review and register nicknames at . 18:24:32 ... Will link to Open Social extensions to AS 1.0 18:24:43 ACTION eprodrom Link to OpenSocial extensions to Activity Streams 1.0 18:24:43 Created ACTION-17 - Link to opensocial extensions to activity streams 1.0 [on Evan Prodromou - due 2014-12-09]. 18:24:52 ... Any more questions? 18:24:55 q? 18:24:58 q+ 18:25:06 yup, thumbs up from here 18:25:08 ack bblfish 18:25:11 +dromasca 18:25:14 http://opensocial.github.io/spec/2.5.1/Social-Data.xml#ActivityStreamOpenSocialExtensions 18:25:53 bblfish: A lot of APIs, seems like the URLs are hardcoded? Does AS always have to be /activitystreams? 18:26:18 ... When we use linked data, we have relations in the documents that point us to different endpoints, and it doesn't really matter what they're called. So we could have Tor URLs for example, that are opaque 18:26:27 tantek has joined #social 18:26:28 http://opensocial.github.io/spec/2.5.1/Core-API-Server.xml#REST-Request 18:26:41 ... How does one find out where an activitystream is, where the user id is? 18:26:49 evanpro: There is a discovery process for OpenSocial 18:26:49 http://opensocial.github.io/spec/2.5.1/Core-API-Server.xml#REST-Request-URI 18:26:54 ... but not 100% sure how it takes place 18:27:04 ... but I do think that OpenSocial clients expect these particular formats 18:27:10 s/formats/url formats 18:27:11 hei tantek 18:27:14 belated regrets for this telcon - out at a meeting today 18:27:25 catches up on logs: http://socialwg.indiewebcamp.com/irc/social/2014-12-02 18:27:33 ... So you can say my activitystreams server is this and I support the AS API, but you can't say where your get endpoint, post endpoint etc, are. Doesn't go fine-grained. 18:27:42 ... Don't believe there's a follow your nose through the activity itself 18:27:48 http://opensocial.github.io/spec/2.5.1/Core-API-Server.xml#Discovery 18:28:03 bblfish: Are the put, post and get all on the same url or different? 18:28:25 evanpro: Kind of. Read endpoint can have different structures, and accept POST and GET. Not entirely sure. 18:28:39 ack shepazu 18:28:45 http://opensocial.github.io/spec/2.5.1/Social-API-Server.xml#ActivityStreams-Service 18:29:04 KevinMarks has joined #social 18:29:19 shepazu: Is it the intent of this group... some limitations, read and discovery isn't great on OpenSocial. Is this group going to extend OpenSocial to improve? 18:29:35 evanpro: Right now we're trying to get an idea of what a social API should look like 18:29:45 ... going through the patterns of different proprietary APIs 18:30:01 ... OpenSocial is special because it's the only standardised one we've got 18:30:12 ... If we choose to standardise, there would probably be some changes 18:30:36 ... I would be comfortable with making things as easy as possible for ourselves, and adopting something this simple with potential extensions, but may be some questions around that and fine-tuning 18:30:45 ... Don't know if we'd do something that's 100% OpenSocial 2.5 compatible 18:30:47 q? 18:30:54 q+ 18:30:59 ack AnnBassetti 18:31:29 is there any actual active development, implementation, deployment of anything OpenSocial on the open web today? my impression is that it was dying/dead 18:31:32 AnnBassetti: Question about APIs in general. In previous descriptions of twitter and facebook, where does the advertising or the promoted tweet/post come into play? 18:31:42 great question AnnBassetti 18:31:50 brilliant question Ann :-) 18:31:55 ... I could see in the enterprise scenario where we would use such an element to advertise, to make a corporate announcement 18:32:00 AnnBassetti++ 18:32:03 AnnBassetti has 4 karma 18:32:29 evanpro: Twitter inserts promoted tweets into the inbox endpoint, they're not different from the people you follow tweets 18:32:31 they are marked as "promoted" but other than that appear as normal tweets 18:32:39 ... in the twitter ToS, clients aren't allowed to differentiate 18:32:49 ... i think that facebook's AS API is similar, but not sure 18:32:59 ... Those are the two that have advertisements in streams 18:33:06 ... Don't know about FourSquare. Don't think ads are in stream 18:33:12 ... Ads separate process, separate endpoint 18:33:20 AnnBassetti: Is that an aspect we should include in our descriptions? 18:33:21 q+ 18:33:32 ACTION eprodrom note about advertising in Twitter and Facebook APIs 18:33:32 Created ACTION-18 - Note about advertising in twitter and facebook apis [on Evan Prodromou - due 2014-12-09]. 18:33:33 evanpro: Interesting part of it. Going to add an action to note about advertising in twitter and facebook APIs 18:33:42 ... Great question! 18:34:06 q? 18:34:09 ack shepazu 18:34:34 shepazu: If we're going to talk about that, we should talk about the general case, not advertising specifically but service notifications, any other kind of update that is sent out that is not part of a normal stream 18:34:45 ... not generated by people that the user follows. Should be some general case. 18:34:58 AnnBassetti: Something you get that you're not asking for 18:35:09 shepazu: Could be something from emergency services that's required by law 18:35:13 and didn't come from your 'friends' 18:35:15 ... Could be priority or out-of-band mechanism 18:35:21 ... A 'class' of notifications 18:35:39 I'd push back on "general case" - please document on the #SocialIG wiki page the specific use-cases that are driving "the general case, not advertising specifically but service notifications, any other kind of update that is sent out that is not part of a normal stream" 18:35:40 evanpro: Maybe jasnell could address this: there were some extensions from AS 1.0 to specify priorities 18:35:49 ... Did the priority process get into 2.0? 18:35:54 jasnell: Yes 18:36:06 ... it's generic, range of 0-1, abstracted 18:36:10 ... No clear semantics on top of it 18:36:14 ... 1 being highest, 0 being lowest 18:36:18 ... Basic form. 18:36:22 (not sure that "priority" is the right mechanism, just threw that out there) 18:36:22 q? 18:36:38 evanpro: Now cwebber2 can overview flickr API 18:36:51 https://www.w3.org/wiki/Socialwg/Social_API/Flickr_API 18:36:59 nicolagreco_ has joined #social 18:37:00 cwebber2: Brief overview of Flickr, not finished yet 18:37:09 https://www.flickr.com/services/api/ 18:37:21 ... General idea with this API is pretty well documented, and in many ways pretty simple 18:37:24 just rechecked the Doodle for next f2f - my info is up to date 18:37:26 ... Uses OAuth 1.0 for auth 18:37:36 ... Used to have their own, but deprecated their own. Limited permissions system in place. 18:38:02 ... A lot of the APIs we've been reviewing have a lot of different endpoints for every single type of activity 18:38:21 ... Which is related to django and rails way of doing urls 18:38:23 https://www.flickr.com/services/api/ 18:38:29 ... But flickr just has a few 18:38:39 ... You can see there are request formats and response formats 18:38:41 https://api.flickr.com/services/rest/?method=flickr.test.echo&name=value 18:38:50 ... On the request formats page, a few endpoints, eg. rest 18:39:01 https://www.flickr.com/services/api/request.xmlrpc.html 18:39:20 ... Also XML RPC endpoint where you can do the same type of thing but different serialisation 18:39:43 ... Lots of different methods for these different endpoints 18:39:57 harry has joined #social 18:40:08 it's understandable, but lightly choppy 18:40:09 yes 18:40:11 better 18:40:13 Yes 18:40:15 sounds good 18:40:19 Little choppy but OK 18:40:47 ... Each endpoint requires different params, depending on serialisation format 18:41:06 Zakim, what's the code? 18:41:06 the conference code is 7625 (tel:+1.617.761.6200 sip:zakim@voip.w3.org), harry 18:41:07 ... Can request different ways of getting information back. So if you wanted to submit via xmlrpc API but get json back, you can 18:41:32 ... Although there are different endpoints for manipulating data and getting information back, the payload for these are pretty small 18:41:41 https://www.flickr.com/services/api/upload.api.html 18:41:44 +[IPcaller] 18:41:45 ... Except photo uploading. Entirely different endpoint for that 18:41:48 Zakim, IPcaller is hhalpin 18:41:48 +hhalpin; got it 18:42:04 ... And if you post to it and it has different params. Well documented. 18:42:22 ... Params include what the safety level is (eg family-friendly), title, description, tags 18:42:32 ... Does have restrictions around recipients: public or friends/family only 18:42:39 ... But that's as fine grained as it gets 18:42:46 ... Also has params to say whether it's hidden or global 18:43:00 ... If yo uwant to do anything on top of that you use the method API endpoint (?) 18:43:15 ... No serialisation formats options for photo POST API 18:43:29 ... interesting, because I'm curious about how we're going to handle media submission alongside activity 18:43:41 nicolagreco has joined #social 18:44:00 ... Can get Evan's comments? When Jessica was implementing pumpio API support for MediaGoblin, different APIs for uploading a photo, how to give other parameters 18:44:39 evanpro: That is a pattern. OpenSocial defers media upload to separate part of OpenSocial API. We'd need to take a look at that. 18:44:47 sorry about it cutting in and out 18:44:50 ... It seems to be a pattern that we see in different API, where media upload is a separate process 18:44:52 I wish I had a better connection :\ 18:44:54 ... Not sure if this is optimal 18:45:02 rhiaro: yes great job! 18:45:09 nicolagreco has joined #social 18:45:10 ... Questions about flickr API 18:45:26 ... Size of the API? Did a quick count - something like 200+ services 18:45:33 cwebber2: Didn't count 18:45:38 nicolagreco_ has joined #social 18:45:39 ... Didn't get to thta 18:45:45 s/thta/that 18:45:51 evanpro: It is a huge API 18:46:09 ... Another question: Don't see anything about adding/removing members of social network 18:46:30 cwebber2: Didn't get to that yet 18:46:49 evanpro: Flickr may be one of the oldest social APIs out there. Flickr is a decade old. 18:46:49 q? 18:46:58 q+ 18:47:01 ... Any questions for Chris about the Flickr API? 18:47:04 ack bblfish 18:47:24 bblfish: Didn't understand whether there were a lot of endpoints, or a lot of messages to one endpoint? Seem to be a lot of different ways of accessing the data. How RESTful is this? 18:47:36 ... Can you use the DELETE verb on the URL so it works with caching? 18:47:36 https://www.flickr.com/services/api/request.rest.html 18:47:45 ... One could ask this for every API. 18:47:55 ... To facilitate linking between resources and caching of them. 18:48:11 https://api.flickr.com/services/rest/ 18:48:14 cwebber2: In terms of RESTfulness of it, and number of endpoints, just a couple of different endpoints 18:48:17 so this is more SOAPY 18:48:20 ... Literally a single endpoint called /rest/ 18:48:25 ... simple HTTP GET and POST 18:48:33 ... more SOAPy, but with a different variety of layers 18:48:47 ... Multiple RPC serialisations, which all look very similar, but just adapt to a particular method 18:48:53 q? 18:48:54 Now that's pretty funny. A single endpoint called /rest/ - that's ironically not even remotely close to a "RESTful" API. 18:49:03 ... If you want to use REST or XMLRPC or SOAP, mostly the same but with different serialisation formats. Three separate URLs. 18:49:03 nicolagreco has joined #social 18:49:16 ah ok. Just 3 different end points, with many different messages to it. 18:49:19 ... Just specify method=flickr.blogs.getlist (eg) 18:49:25 thanks 18:49:29 yup 18:49:36 evanpro: Any other questions for Chris? 18:49:38 q? 18:49:54 nicolagreco_ has joined #social 18:49:56 Do any existing *real-world* (i.e. at least one company supporting it) social web APIs do ANYTHING with any HTTP verbs other than GET, POST? 18:50:09 I.e. I would claim that NONE of them (i.e. nobody) use(s) DELETE. 18:50:17 Here for AnnBassetti, Roy's thesis http://www.ics.uci.edu/~fielding/pubs/dissertation/top.htm :-) 18:50:18 I could possibly organize it, I'm not sure how you would want it organized 18:50:19 ... Thanks Chris! A thing that might be useful to cluster methods to inform the rest of our process 18:50:31 https://www.w3.org/wiki/Socialwg/Social_API <-- roadmap 18:50:32 ... Set aside last few mintues to discuss where we are with our process on social APIs 18:50:33 I could try to follow the flickr/twitter api patterns 18:50:35 ... See roadmap link 18:50:37 And thus would pushback on RESTians that keep bringing up using PUT, DELETE, etc. because in practice, they are dead ends. 18:50:58 :-) 18:51:01 ... 1) identify APIs, 2) identify functionalities, 3) assemble requirements 18:51:04 https://www.w3.org/wiki/Socialwg/API_requirements 18:51:10 ... We have a beginning of that (see link) 18:51:23 ... Would like to know from the group if it makes sense to continue with more reviews? 18:51:37 q+ 18:51:39 ... More idea about micropub? Indieweb API 18:51:50 I'd like to talk about LDP based API 18:51:51 ... Would like to talk about pump.io API, similar to OpenSocial AS 18:52:02 ... Main question is, do we continue with these reviews for another week or so? 18:52:05 the_frey has joined #social 18:52:09 q? 18:52:11 ... Or start to collect requirements? 18:52:16 ack shepazu 18:52:51 personally, I'd like to see a minimal baseline proposal soon. even if just a strawman to get started 18:53:12 shepazu: Useful to hear about different features etc of each. I would like to hear about more. But moving on to other things might be useful. But main thing I think would be useful is to have a matrix; a list of features and APIs 18:53:16 ... And break it down that way 18:53:26 ... Qualified checkmark sort of thing with an explanation 18:53:35 ... So we know what we're comparing in terms of features, and what gaps might be 18:53:48 even if we had a basic API started, we could still review other APIs and use it to critique our design 18:53:54 ... Useful when addressing use cases and requirements, because you can say which APIs support required features 18:54:03 the_frey has joined #social 18:54:20 ... If that's useful, who is willing to compile such a feature list? 18:54:32 ... And the people who presented the review on an API can fill it in on the matrix 18:54:42 I like the Matrix 18:54:44 ... Just a suggestion, but seems useful 18:54:51 evanpro: Makes a lot of sense 18:55:10 ... Main downside is tables on mediawiki is an exercise in patience... 18:55:22 Tables are terrible on media wiki - you could just email Doug and he could make the table :) 18:55:30 shepazu: Willing to do the table wrangling, then all pepole need to do is fill in checkmarks 18:55:41 evanpro: Great, happy to do that 18:55:56 q? 18:56:09 Sandro: Can just do html 18:56:09 ack sandro 18:56:46 sandro: Another thing useful is a side-by-side comparison, ie what does SWAT0 look like in each of these APIs (or some other simple test case) 18:57:15 evanpro: SWAT0 is specifically done for federation efforts, but there are things that could be captured 18:57:25 sandro: SWAT0 is kind of overkill, but the individual pieces 18:57:43 evanpro: Main thing to avoid becoming the API documentation reading book club 18:57:56 ... Can't read documentation forever, we need to use it 18:58:01 ... Like Doug's idea of matrix 18:58:15 ... Could aaronpk or tantek do a presentation about micropub next week? 18:58:21 sure, happy to talk about micropub next week 18:58:24 ... And any others? 18:58:25 q+ 18:58:27