#snarfedlet us know when it's out of stealth mode! :P
#kylewmhappy to talk about it now, it'll keep me from abandoning it :)
#kylewmi went through archive.org the other day looking at my friends pages from 2002-03 era, and noticed they were mostly static html + some external comment service that I can't remember the name of
#kylewmGWG: you indieauth into the service, and choose a username. it gives you a snippet of code to stick at the bottom of your pages, and that's it. i was planning on having like a dashboard where you can see and possibly moderate incoming webmentions.
#aaronpkkylewm: "choose a username" <-- why not just use their domain name as the usernaem?
#kylewmaaronpk: so I guess I do want to work on this for my own use (I basically just pulled out the wm-receiving part of redwind), but if I wanted to open it up as a "service" does that feel too much like ripping off webmention.io?
#kylewmI was thinking of it more as "diversity" rather than "competition", but now I'm not sure if it would be perceived that way :(
#snarfedkylewm: i expect most people would see it as diversity, not competition
#snarfedwe do already have wm.io, wm.herokuapp, and wm.js :P
#snarfednot to mention the majority of people roll their own CMS!
#snarfedi think our (hopefully mild) push back was more about using effort efficiently, esp from someone like you who's obviously good
#snarfedbut motivation is huge, so if you want to do it, go for it!
#snarfed(to be clear, by "we do already have…" i meant that there's already duplication of effort, and we're generally still positive about those projects)
#acegiakGWG: I'm redoing my posse in my own plugin because I accedentally updated the posse plugin I was using which got rid of my modifications to the tumblr code
danlyke_ joined the channel
#GWGacegiak: I want to POSSE in my own plugin because SNAP is a bloated mess and its still the best choice so far.
#snarfedi'd happily accept a PR! bridgy does already oauth to tumblr, fwiw
#acegiaksnarfed: wait, will bridgy send me webmentions from tumblr reblogs etc?
#snarfedacegiak: sorry no. bridgy currently accepts and renders incoming webmentions on tumblr, and sends outgoing webmentions for links, but doesn't yet handle tumblr-specific post types like reblog
#GWGI'm trying to build toward author data in Indieweb Taxonomy, which has led to plans to design in an option to change where the author data is stored for future.
tgbrun joined the channel
#GWGWhich led to thoughts on linking that to inbound mentions with Semantic Linkbacks in the form of a nickname cache, which led to comment moderation
#acegiakalso: with your new way of storing metadata in the taxonomy plugin, doesnt that make checking for and displaying the metadata more complicated for theme authors?
#GWGacegiak: I keep trying to avoid what I did with Indieweb Taxonomy...changing the data structure and breaking things...
#GWGacegiak: I'll be writing in a few functions to help out.
#GWGHelper functions...also known as wrappers, are common in the Codex.
#GWGthe_title is a wrapper around get_the_title, for example.
#acegiakcause i feel like with the old way you could have multiple plugins that do thr metadata and multiple themes that read it and they would be interchangable
#GWGadd_options_page is a simple wrapper around add_submenu_page that adds an option page, as opposed to add_plugins_page, etc.
#GWGThe authors you are likely to respond to are the same who are likely to respond to you
#GWGSo, I want to craft it so, as a future add-on without changing anything in the structure, the authors could be stored separately from the posts and then retrieved.
#GWGacegiak: Local caching of profile pictures, mostly.
#GWGI may never do it, but I want to write it so I could. So I wanted to look at the functions WordPress uses for gravatars and write my functions similarly.
#GWGacegiak: I had a loop going to migrate the data. I didn't unset the variables, so if the response didn't have a quote, it used the one from the last loop iteration
#aaronpkI think they have a way to delegate access to other google accounts
#GWGaaronpk: According to Google, if you have it linked to a Google+ page, you can assign other Google Plus individuals as page managers, thus giving access to the Youtube page
#GWG8AM Pacific is 11AM Eastern is 4PM UK. That's a hard division, to use an example
#ben_thatmustyeah, if we have people in europe interested in participating / co-organizing I would say we should start in the morning EU time, then break things in to smaller sectiongs
#KartikPrabhui suggest you guys wait for a week to see participating time zones then make the schedule
#ben_thatmustso say, 3 hours sessions, 3 hours talk, bu then it would be late enough that when we start the second sessions set we are able to get US in to the groups
#GWGI'll get up at 4AM, but I'd be a bit of a zombie.
#ben_thatmusthaha, i will get up for 6 am no problem
#ben_thatmustbut any earlier and i won't be getting much done
#GWGWell, 6AM Eastern is only 11AM UK Time. That isn't so bad.
#ben_thatmustif there is no one outside of US timezone i'd say we start somewhat like 8AM(p)/11AM(e) or 7AM/10AM
#ben_thatmusti think we just have to look at where participants would be joining from, we can't get all groups in on it unless we plan to do a lot of sessions like that
#Loqibarnabywalters: jonnybarnes left you a message 5 days, 3 hours ago: have you upgraded to Yosemite yet? Messing around with getting guzzle working with sites like alpower.com (missing intermediate certs) and Yosemite's cURL now repliactes browser behaviour and just works™
#barnabywaltersjonnybarnes: that’s good to hear! I’m not on Yosemite yet, but I’m glad that Guzzle is defaulting to using the system chain. At some point I need to go over all my projects and update to the latest version of Guzzle and get this behaviour — hopefully then I can stop shipping hard-coded cert bundles in projects
brianloveswords, chalettu, ShaneHudson, shiflett and reidab joined the channel
#barnabywaltersAMD (of which require.js is an implementation) allows you to define your code as a function of it’s dependencies, which makes a lot of sense on the web where dependencies are loaded async
#barnabywaltersyup, that’s the library I’m using for event listening
#barnabywaltersat this point I could probably get away with ignoring IE and using the built in DOM APIs, or writing my own tiny wrapper for them
#bretahh i see. yeah that might be a little bit cleaner in the long run. sweet, this all looks doable
#Loqitantek: bear left you a message 1 day ago: thanks - I was hoping my post made sense from the IndieWeb perspective, I have more thoughts but those were the ones that I dashed off before I forgot them
#bretbarnabywalters: do you have any experience with embedded content loaders? (i.e. what boingboing does to their videos? http://boingboing.net/page/2 )
#barnabywaltersI’ve never built anything (or used anything other than cassis auto_link) for embedding stuff, but am going to have to for Shrewdness
#bretone thing i noticed when developing on ancient hardware, is that having more than like 1 youtube ebed kills performance and loadtimes
#barnabywaltersat the very least, to cache 3rd party images, and more than likely to wrap video and audio embeds with a poster image which loads the content when clicked
#barnabywaltersbret: yup, even on my macbook, early version of shrewdness with 5 columns full of 100 posts each hung ff very quickly
#barnabywaltershence the number being reduced to 10 loaded by default from each column
#barnabywaltersit was mainly the 3rd party profile photos, but video embeds are even worse
#mkoPfft. Just a little slip-up. I will say this, though: since these are public logs, you should probably change passwords anywhere that you've used that one.
tantek, davidpeach, Pea1, benjamin-agaric, gr0k, mko, ooland, Sebastien-L, snarfed, mlncn-agaric, brianloveswords, alexhartley, michielbdejong, tilgovi and j12t joined the channel
#kylewmoh interesting, they recommend using www. (or another subdomain) because you can use a CNAME record rather than hard-coding an IP address https://status.heroku.com/incidents/156
#aaronpkit doesn't actually make much sense to use refresh tokens because there is no client secret involved either
pfefferle joined the channel
#glennjonesSorry ro explain, I have mixed my indieauth support into my preexisting token system which is also used for oauth2 - so my bearer token timeout afterr x amount of time. Do need to make them last forever
#aaronpkso if you wanted to use ownyourgram then you'd need to give it a token that lasts forever, or you'd need to log in periodically to give it new tokens
#aaronpkyou could for example only issue permanent access tokens to that client ID tho
#glennjonesOK for indieauth issued tokens I will extend the timeout well into the future, so I dont have to keep logging back into clients - Just want to check I had not missed a refresh feature
#glennjonesHave ownyourgram.com and https://quill.p3k.io/ up and working with my transmat.io app. All the devloper related information in the flow was very useful
#bretis it pretty integrated into transmit? or could I run it on its own?
#glennjonesThis use is a bit different in that the links are in my blog http://glennjones.net but all the token and micropub inteaction is handel by https://transmat.io which supplies the content to the blog through an API
lukebrooker joined the channel
#bretps, the font on transmit.io looks a little funky in safari
#glennjonesI am just about to refactor it so it can be installed and run on its own should take around 4 weeks (hopefullly) to rebuild it, so soon