#mkoKevinMarks: Regarding /link-preview -- have you thought about just implementing a simple OEmbed provider that returns an iFrame or modified view of the page requested? I just finished implementing an OEmbed and OpenGraph Embed consumer for my site ( see https://mowens.com/notes/2014/08/20/1/ as an example ).
#mkoNext, I'm going to make an OEmbed Provider for my site that provides a Rich HTML response similar to that seen below that post.
#mkoThose were the pages I was using to test various implementations. :-)
#kylewmmko: wow! do you have to do some special coding for each provider?
#mkosnarfed: that was actually why I mentioned it to him, as being an OEmbed provider seems more useful, as it seems like a well-known and well-structured standard.
#mkoOn oembed.com you can find most, if not all, of the known OEmbed providers.
#mkoI did have to write custom code to convert OpenGraph data into a rich-type OEmbed, but that was pretty easy, too.
#mkoIt's still pretty rudimentary, but for now it meets or surpasses most of Twitter.com / Facebook.com link-preview functionality.
#mkoThe process of testing and validating providers is pretty fast. I figure if I wanted to add all of the known providers it would be a half day's work since some of them actually don't follow the rules set forth by OEmbed to the letter.
#snarfedkylewm: not afaik; O is just a common prefix for "open" standards
#snarfedmko: iirc people here have mixed feelings about oembed, esp the mf2 advocates, since they think you should use mf2 for everything
#kylewmirc it that way too, was looking back at logs for confirmation
#kylewm"tantek: oembed seems like extra work to support (as a publisher) when you can simply publish rich h-entry markup instead"
#kylewmsnarfed: what's the thing you say, "be lenient on what you'll accept and strict on what you publish"?
#snarfedsure! definitely not me though. forgot the name, but it's an old school tech "law"
#snarfedi interpret it as making sure you emit valid, well-formatted data, not necessarily that you harshly minimize the number of formats, but that's just me
#acegiak_is it ok to have your h-cite inside your e-content?
#kylewmacegiak: totally up to you I think. if the thing you're citing is part of the post content, I think it's appropriate in the e-content. if it's more like a reply-context, that might be better outside
#jonnybarneswith that error ive screwed up bridgy :(
#jonnybarnesit normally resolves my jmb.so links correctly to jonnybarnes.uk links
#jonnybarnesi.e. when it makes a POST request the target is jonnybarnes.uk/notes/id
#jonnybarnesbut because it was getting 500 errors when resolving the inital jmb.so link from barnabywalters tweet it then tried sending POST with target jmb.so/t/id
#jonnybarneswhich my end isnt configured to deal with :(
#jonnybarnesbasically, for some reason, php-mf2 doesnt get an actual html string from guzzle4
#barnabywalters$resp->getBody() rather than $resp->getBody(true)?
#jonnybarnesnot sure why, forcing the issue by tyecasting works, i,e, $html = (string) $response->getBody();
#barnabywaltersat least in guzzle 3, $response->getBody() returned a resource by default
#barnabywaltersI added a string cast to php-mf2 to handle cases like that but I think they only cause ->toString() to be called if they’re syntatically inline, not in a different function
ngoldman_ and tommorris joined the channel
#jonnybarnesso I think I have this working now barnabywalters, only issue being I have to look not into $rels, but into $rels[0]
#jonnybarnesbarnabywalters: can I not worry about wether or not they've used link's or a's, as long as either have rel="webmention" stuck in there php-mf2 should expode it in the rels array?
tecgirl, snarfed and crossdiver joined the channel
#GarbeeGWG: I'd be more concerned about spam-bots reading phone numbers and then having automated systems call you non-stop. Or them taking it and adding it to a list to sell to telemarketers.
#GarbeeI don't think too many people (or even marketers directly) go around the web just looking for phone numbers to call.
#bearI had to take mine down - got added to robocall lists
#Jeenahas someone here got the new Firefox Accounts server up and running? It seems very complicated to get it running with 5 or six deamons which have to run at the same time at different ports, etc.
snarfed joined the channel
#GWGGarbee: I was told Robert Scoble has his number up online. I could call him and ask
#ben_thatmustbemeto clarify, I have just started to clean up the microformats on my site. Currently it has h-card for each post which just has my name and photo
#ben_thatmustbemehmm, i have an author link in there, perhaps i should link that to my new h-card page
alanpearce joined the channel
#GWGben_thatmustbeme: I am working on a sidebar for it
#ben_thatmustbemeKevinMarks, I publish my number only to people who i white list to see it.
#ben_thatmustbemeonce I have the admin pages for it done i plan to have it so I can group contacts and then select what contacts can see what info