2016-08-20 UTC
tantek, doesntgolf, KevinMarks_ and mblaney joined the channel
# 01:35 mblaney AngeloGladding: I just had an idea related to your project.
# 01:37 mblaney instead of (or as well as) looking for explicit connections via h-cards, you could parse blogs for links.
# 01:37 mblaney ie if you follow the rel=follow in my h-card, you will find my blog.
# 01:39 mblaney if you parse my blog for links you should find links that give you more h-cards or match ones you've already found.
# 01:39 mblaney now you've got linkage between community members who mention each other.
# 01:42 mblaney also I think most people have their blogs on the same page as their representative h-card, so the rel=feed thing is less common.
# 03:09 AngeloGladding mblaney definitely going to follow all relevant rels (ignoring css, ..) and gather all items and dump into the graph
# 03:10 AngeloGladding just got the idea to push it to a public git repo while i write a graph parsing utility
# 03:11 AngeloGladding simple graph analysis for everyone!
# 03:11 AngeloGladding historical as well.. diffing the pretty-printed json appears to work quite well
# 03:11 AngeloGladding we'll see how quickly i use my github storage allotment :)
AngeloGladding, KevinMarks, bnvk_, kline and KevinMarks_ joined the channel
cmal joined the channel
# 09:40 AngeloGladding tomorrow will a) attempt to incorporate 100 more people using modified repr h-card algorithm and b) begin an analysis script
cmal joined the channel
cmal, neoPixel and ben_thatmust joined the channel
cmal and KevinMarks joined the channel
# 16:01 GWG Is there a scenario where an HTTP header and an HTML header are both present for a page but the HTTP cannot be read? Thinking about things like the webmention header.
# 16:02 aaronpk normally the opposite is the problem. for example static sites can't send HTTP headers, they can only set HTML tags
tantek and mindB joined the channel
# 18:23 aaronpk I like those suggestions but what a mess it would be to actually migrate to that
mindB, AngeloGladding, doesntgolf, KevinMarks and tantek joined the channel
# 19:42 KevinMarks Trailing commas would be good, that does make things easier in python
# 19:48 aaronpk whoa. i forgot to deploy webmention.io back in like march
# 19:49 aaronpk i did a git pull and a whole bunch of changes came through. not what i was expecting.
gRegorLove joined the channel
# 20:16 voxpelli aaronpk: mimicking my experience of IWC Nürnberg fairly well ;) "What is all of this new undeployed code? Why is everythng broken?"
# 20:17 voxpelli aaronpk: just noticed you're going to Brighton – yay!
# 20:22 aaronpk definitely thought i was going crazy for a minute
AngeloGladding, KevinMarks_ and cmal joined the channel
KevinMarks and KevinMarks__ joined the channel
tantek, cmal, bear, mindB and KevinMarks joined the channel
# 23:49 rascul quotes because it will never be done, but it's almost done enough for now
# 23:50 rascul i'm concerned that feed readers will get a weird order for my posts
# 23:51 rascul unless i make a rss version or something, the h-feed will be out of order
# 23:52 rascul the only way i think to fix that is with the javascripts but i don't know if i want to do that
# 23:52 aaronpk i think readers will end up sorting by the published date of the post so you should be fine
# 23:53 aaronpk i can't imagine a reader that doesn't do its own sorting
# 23:53 rascul it's just that because of my two columns, there's two lists, with every other post
# 23:54 rascul i could put them all in one list then use the javascripts to move every other post to the other list but i'm concerned that it could take too long to do all that when it loads
# 23:55 aaronpk the way my polling works is i look at every entry on the page and if the URL of it hasn't been seen before, it adds it to the list of posts
# 23:55 rascul i imagine it has to be a fairly dumb reader though to get my feed mixed up
# 23:56 rascul the not dumb ones probably do their own sorting and/or or would mark the ones they've already noticed