tantekaaronpk, I agree, this is fascinating, from a data ownership perspective: " the client is the canonical message store, and the server is a router with temporary storage."
tantekI'm looking at two data points: $3B offer for Snapchat (refused). and $16B offer for WhatsApp. I bet if you did $/phone# math you might find the same ratio.
tantekwhich therefore also implies that FB *will* make an offer on Secret for the same thing. userid/phone# association, along with implied network of which phone#s communicate with which other phone#s.
tantekthat black & white clean design look of threema's home page appeals to geeks and design nerds but bores pretty much everyone else. also, looks unprofessional to the average person.
tantekwhereas the slick graphics and textures (and hey, screenshot right there!) of whatsapp is MUCH more professional looking (= trustworthy) and appealing
tantekJeena, so far my family and friends do not use Whatsapp. I have gotten ZERO requests from friends for me to install it so they can use it to communicate with me.
bretMy cousin uses it. I asked her 'whats up with whatsapp?' says she uses it to sms friends in other countries, which I guess she has a few freinds abroad
tantek!tell nitot, how about setting up standblog.org as your identity instead of standblog.org/blog - I mean, do you really need to say "blog" twice? ;)
JeenaSadly the devices are kind of crapy and the os still is very buggy and lacks important applications like WhatsApp, a Podcast app or simple things like address book sync (at least via CardDav or what was it called)
breti wonder if detrich would be interested in doing a firefoxOS hack session in portland, where people could install FFOS on random android hardware together
Jeenathe by far most annoying bug is that if someone calls you and you exidentally press the home button the "unlock phone" UI pops up and you can't take the call anymore because it is in front of this UI
JeenaI use Firefox OS 1.1, which introduced Push Notifications, since september last year, but I haven't used any application which would use them yet.
Loqiacegiak: tantek left you a message 8 minutes ago: do you have any documentation of how you're doing POSSE to Tumblr including native retumbls on Tumblr?
tantek.comedited /Tumblr (+1402) "/* How to */ POSSE details for higher fidelity on Tumblr (for a better UX for your Tumblr friends that read/see your posts/activity there)" (view diff)
acegiakI'm trying to work out if there's an easy way to add a rel="syndication" link because there's no information returned by the call to post the reblog just 201 success
snarfedah. yeah, one of my favorite parts about python is, hard things can take a while to figure out, but once you do, they're usually very little code
snarfedbtw, it'll be at least a few days before i get back to it seriously, so if anyone's interested in exercising mf2py more deeply, feel free to dive into https://github.com/snarfed/bridgy/issues/10
Loqitommorris: KartikPrabhu left you a message 2 hours, 54 minutes ago: working mf2py using BeautifulSoup. https://github.com/kartikprabhu/mf2py testing/comments appreciated. I am working on testing it on the test examples
@skoopskauft euch einen raspberry pi, mietet euch einen server, nehmt ein ausgemustertes notebook... installiert nen mailserver drauf. #ownyourdata (twitter.com/_/status/436389766083670016)
@skoopsinstalliert euch eine kleine owncloud, einen jabber server und schon habt ihr chat und dropbox und mails unter EURER kontrolle #ownyourdata (twitter.com/_/status/436390176575987712)
Loqinitot: tantek left you a message 9 hours, 14 minutes ago: about setting up standblog.org as your identity instead of standblog.org/blog - I mean, do you really need to say "blog" twice? ;)
barnabywalters, marcthie_, LauraJ, glennjones_, jjuran_, jcbsnd, bnvk and nitot joined the channel
jonnybarnesyeah, he worked at Google before joining the Tor project and hated it, I think he's much happier now he's doing something he actually cares about
bnvkbarnabywalters: if you're able to sneak away today or tomorrow you should come to the Tor hackathon and see if you doing some IndieWeb stuffs with .onion addresses
jonnybarnesbarnabywalters just checking, but I think I know why php-mf2 isnt working, if the $html string passed to it also includes the HTTP headers before <!DOCTYPE html> then it wont work?
bnvkbarnabywalters: in the right setup you wouldn't need to type that .onion address, you click on a persons name and links back to posts on their onion node
barnabywaltersjonnybarnes: I have a little function which takes a guzzle response, figures out how best to parse it (i.e. using php-mf2 or mf2-shim) and returns the microformats, so I don’t have to get the body and URL out manually
bnvkif I replied to something you posted at waterpigs.co.uk, but I replied from my indieweb node that happened to exist at an .onion route, you'd be able to navigate to my node by clicking on my name, avatar, reply link
bnvkI don't know if i'd ever make a .onion addresses be my PRIMARY address, but I can definitely see a world where people post and interact from multiple different addresses
barnabywaltersin practise I think the barrier of using and understanding Tor, how to set up a hidden service, what .onion addresses are, then actually remembering one for yourself, then persuading anyone who wants to interact with you to install Tor is greater than the current barrier to hosting a website
barnabywaltersand whilst there isn’t much of a barrier to *using* Tor, there are larger barriers to setting up a hidden service, then the *huge* barriers of remembering and using ugly .onion addresses and persuading everyone else to use Tor so the .onion addresses actually work
barnabywaltersbnvk: so one thing which might allow this to work and be a seamless transition from conventional domain names, whether it’s ugly .onion or nice namecoins…
barnabywaltersthen UAs (browsers, indieweb nodes, etc) which supported alternate DNS systems would detect and use them, and everyone else would fall back to using the original
barnabywalterse.g. <a rel="in-reply-to" href="http://brennannovak.com/notes/100">in reply to a post on brennannovak.com</a> <link rel="in-reply-to" href="https://skbgdsrgbekjshrgbks.onion/notes/100" />
barnabywaltersout of interest what actually is the HTML content you’re trying to parse? have you passed it through waterpigs.co.uk/php-mf2 to make sure it parses correctly?
JeenaI assume some of you implemented POSSE with Facebook, how does it work? I created a facebook app and I got the app access token, but I can not find out how it relates to the user access token
barnabywaltersjonnybarnes: mainly because headers can be added easily in code by middlewares without having to change templates or manually mess with response bodies
Jeenahm damn, now when pfefferle is gone I got a problem: "OAuthException :: Error validating access token: Session has expired on den 20 februari 2014 07:00. The current time is den 20 februari 2014 07:08."
snarfedbtw, i'm guessing you're already far enough in, but let me know if you want to work together on posse as a service instead of built into your site. i'm starting on that right now, and i have many of the hard parts done (and tested) from bridgy. https://github.com/snarfed/bridgy/issues/10
jonnybarneshow consistent are microformats? i.e. when trying to grab the content of a reply from the output of php-mf2, this is currently the location: `$content['items'][0]['properties']['content'][0]['value']`.
jonnybarnesnot fully, im still working out how to store webmentions, currently i'm serialising the array produced by php-mf2 and saving that in the database, but I dont think thats a good way of doing it
aaronpkmy recommendation is to split up as much of the logic as possible. so for example when you receive a webmention, just store the raw HTML somewhere, and parse it later
aaronpkthe idea being if I update my microformats parser, I can delete the parsed representation of the pages and everything gets re-parsed automatically
aaronpkon disk I have a file with the HTML of each URL that mentions me, and a corresponding file that ends in ".json" with the parsed version. so I can always rm -rf *.json and everything gets re-parsed
caseorganicsnarfed: tantek: remember when i mentioned visiting silicon valley next weekend? i had to move the trip by two weeks. we'll have an indiewebcamp dinner then! new dates 2014-03-15 to 2014-03-17
snarfedit's fri and sat, right? wish i could volunteer my office, but it'll be full of us doing normal work on fri, and weekends are tough for our bldg in general
caseorganici love the idea of doing it at long now foundation. i have a friend who works there. it's very small though. likely not a lot of break out space.
KevinMarks_"I’m interested in the power relationships around identity and privacy, and how to use market forces and regulations in the interest of the people who use the Internet."
tantekKevinMarks, if you have his ear, then start suggesting small immediate practical changes to his website along the lines of IndieMark and IndieWebify.me
tantekby getting to engage in our easy/simple building blocks one at a time, I think that will "sell" him on it more than any grand vision attempt could
KevinMarks_The exception to the above rule is that you are encouraged to use 3rd party services, APIs, open source projects, libraries, and frameworks.
caseorganicKevinMarks_: i was told i should head out there to support it, but i'm going to be in denver. sometimes i get really short notice on what esri's involved in.
aaronpk"Given URL is not allowed by the Application configuration.: One or more of the given URLs is not allowed by the App's settings. It must match the Website URL or Canvas URL, or the domain must be a subdomain of one of the App's domains."
Jeenaok it only took half a day but I think I am able to post to facebook now, even though it wants a image which it can show for the link and I don't want it to.