snarfed, lukebrooker_, pfenwick1, pasevin, fmarier, KartikPrabhu and j12t joined the channel
#KartikPrabhutommorris: sent pull request to mf2py. nothing major just some extra comments and restructuring that I thought useful. On to BeautifulSoup now
#waglethe cloud might let you snap images of the virtual machine or something.. of course, thats digging into the expenses
#aaronpkbut holy crap that's gonna be a project to switch everything over
#aaronpkyeah I can do snapshots of the DB server on linode, but mysql doesn't really play well with live snapshots. it wants the DB to be shut down during the snapshot
#aaronpkJeena: would love to get your input on this translation! http://indiewebcamp.com/principles-de I think people are translating it inconsistently between formal and informal "you" so that probably needs to be cleaned up. which would make more sense?
#neuro`!tell tantek tomorrow xtof and I have a meeting at the French Mozilla Foundation headquerter to start a Paris chapter of the Homebrew Website Club
#aaronpki'm curious to see how this translation feature works on duolingo. I've been using it to brush up on my German the last couple weeks, but thought this would be an interesting experiment.
LauraJ joined the channel
#pfefferlecweiske I would use the english words for all indieweb related stuff... "indieweb" instead of "unabhängigen Netz"... what do you think
#jonnybarnesdid you see my blog post by the way? about micropub and oauth tokens.
#aaronpkoh yes haven't yet read it completely. was looking forward to doing so!
anselm and benprew joined the channel
#jonnybarnesis this a stupid table structure to store my webmentions in?
#jonnybarnesalso, do you process webmentions before storing them, or store the raw content and filter it before outputting to your site?
#jonnybarnesI'd lean towards storing the raw content myself
#aaronpkI store the full HTML of each page plus the parsed JSON. I can go back and re-parse the JSON if needed that way, like if I update my mf2 parser
#aaronpkthat may not be sustainable, cause my html folder is already 170 megs
eschnou and LauraJ joined the channel
#jonnybarnesaaronpk so when I visit a page with a webmention-ed comment on, is that coment being generated from the JSON?
caseorganic, hober, _6a68, paulcp and tantek joined the channel
#Loqitantek: aaronpk left you a message 1 day, 11 hours ago: I'm having a tough time turning this algorithm into code: http://indiewebcamp.com/comments-presentation#How_to_display It seems to contain conditionals that overlap/wrap in odd places. Is there any chance you could rewrite to clear that up?
#Loqitantek: barnabywalters left you a message 1 day, 2 hours ago: any objections to merging https://github.com/tantek/cassis/pull/11? it would make using cassis much easier in PHP+composer projects
#Loqitantek: neuro` left you a message 2 hours, 58 minutes ago: tomorrow xtof and I have a meeting at the French Mozilla Foundation headquerter to start a Paris chapter of the Homebrew Website Club
#snarfedyeah, aaronpk suggested a similar thing. definitely a good idea, but based on my testing, it wasn't the number of users alone that triggered the 401s. details in the GH issue
#KevinMarksThough that "email is the future" conclusion is sad
#pfefferleaaronpk oh, it seems to be some kind of a snippet... how can I change that? and if I do, is it also used on the french, ... site?
#aaronpkoh yeah it's a template. there should be a link to the tmplate at the bottom of the page when you edit principles-de
#JeenaI tried to install the software so I could become my own browser id provider and failed miserably
#tantekKevinMarks, it's an incremental conclusion based on the fact that every non-vendor-specific-sign-up in apps uses email.
#aaronpkit's not on the french pages since xtofr has been doing a great job translating
#tantekjernst - I have to disagree with this: " all that energy wasted, with an entirely predictable outcome". It was not energy wasted, Persona was a new enough incremental approach that is was very much worth pursuing. And it was not entirely predictable, that's the point. It helped reveal real vs. theoretical issues.
#JeenaI kind of would have written: "Diese Seite wird mit Hilfe von Duolingo übersetzt."
#tantekKartikPrabhu - I disagree - on all three points.
#tantekand no, they are not lessons you can draw from any of the data
#Jeenait is a bit odd to work with Duolingo, if someone already translated a paragraph then you only can vote it down but not add a alternative version, at least I wasn't able to find out how.
#aaronpkhm really? I was able to type over someone
#aaronpkyou just click on the sentence and can change it
#KartikPrabhutantek: point 3 us shaky. But the first one for instance. If you have a decentralized identity then wouldn't that hurt the business model "let's find out most things about you"?
grantmacken joined the channel
#tantekKartikPrabhu - email as decentralized identity as worked for far longer than any of those web business models have existed.
#Jeenatheir UI is pretty cool for this kind of work. But the first translations were kind of horrible ^^
#aaronpkyep! I wanted to start with a page that doesn't change very often. I figure it's probably not necessary to translate the more technical pages, but this one seemed like it'd be useful
#KartikPrabhuthat is true. I feel like that is the premise of the article. That identities are used as products today, and hence... other conclusions. If identity is not a product then yes decentralized ones are very good
#KevinMarks_I was wondering how close a static external site could get
#KevinMarks_this is the extension of tantek's file-oriented view
#KevinMarks_my sense is that your site is a bit like that already, aaronpk ? With the dynamic overlay?
#aaronpkyou know like how your SMTP server could be offline for most of the time, and when you'd dial up then servers would re-try and you'd get your email
#snarfedyup. basically a static site generator plus cdn
#aaronpkKevinMarks_: yeah my site is generated from flat files. it only changes when I post something or when I get a webmention
#snarfedi have to admit, i appreciate the old school comment form. to support that, i'd need the webmention service to also support sending the comment content directly
#aaronpkso if I could put a CDN/cache layer in front of it, my server wouldn't even need to be online very often
#snarfedbut other than that, it might work for me too
#KevinMarks_I think camlistore is groping towards this, but still assumes you're hosted on a public server
#snarfedyup. it's also kind of a side effect for them. camlistore's primary use case is inward-looking, more like personal cloud, as opposed to outward-looking indieweb style publishing
#aaronpksnarfed: over the years I've had comment forms on my sites, and inevitably I turn them off and remove them because I just don't get good quality comments that way
#KevinMarks_well, it has config hooks for s3. maybe I need to look again.
#bearaaronpk - my site is also flat files and generated behind my home firewall - so i'm wrestling with the same things
#bearmy thought was to have a service on the outside that forwards to xmpp (which for me is more central than smtp for my personal stuff)
#snarfedyeah. it's subjective. people feel both strongly and differently about comment style
#aaronpkmy last attempt was using facebook comments on my site, which meant I no longer got spam comments, but *still* the quality of comments was extremely low
#KevinMarks_I like the idea of an essentially static site, that can be archived by archive.org etc safely
#KevinMarks_it assumes you can parse the site back into the canonical form, which mf2 makes potentially possible
#bearfor me all of the source markdown files are available from the static site