#GWGtantek: That you would be showing contacts based on logged in status. It was based on your work. But, being as login is often automatic, how do you know what a person wants you to not share?
#tantekGWG - where/how were those UI thoughts captured on the page?
#GWGThe Status format of my site, the one with the picture of me when I post, links to the author url. But that page does not have an h-card.
#GWGIt should link to the site or not link at all.
#kylewm.comcreated /session_facilitating (+1859) "Created page with "<dfn>Session facilitating</dfn> refers to facilitating a brainstorming session at an IndieWebCamp event. Session facilitators are responsible for setting the discussion topic/goa..."" (view diff)
#tantekin general, AS is a good source for previous research, but should not be used as any kind of canonical list of types - because they're based in "wants" - not cited research of existing examples and implementations implementing publicly.
#kylewmrascul: I really like how minimalist your html is. almost no classes that aren't for mf2
#rasculkylewm the site generator uses mf2 internally also
#GWGtantek: Semantic Linkbacks generates the code for the different types of responses, but doesn't mark them up. That is what I want to add, but need a simple clean markup
#tommorrisalso, my Eye-Fi arrived today. I'm going to play around a bit this evening to see if there is an easy way to have my pictures go from my camera straight to somewhere I control on the web.
#tantektommorris: that WP page is good general wiki editing advice, which I tried to avoid with /wikifying
#tantekand rather focus on aspects of /wikifying that are more in the spirit of what indiewebcamp is about
#tommorristantek: yeah, just pointing out the potential namespace conflict, to ensure we are clear that if they aren't the same, we should document that.
#aaronpkprobably supports sftp, I kind of assume that when I say ftp now
#aaronpkpoint is it's not like it goes directly from the card to your server
#tommorrisaaronpk: got the Pro one. was more thinking of having some scripts on my desktop, or some processes on my phone, to do the auto-upload.
#aaronpkah yeah, if you are in range of a computer and your computer is connected to the card's wifi then it goes direct to your computer
#tommorrisI mostly bought it so I can quick-post stuff from my phone while out-and-about. seems silly to be taking both a good photo and a phone snap. :)
#aaronpkbut that ties up the wifi connection of your computer so you wouldn't get an internet connection that way
#tantekthanks tommorris - added the ref to WP wikify
#tommorrisEyeFi + iPhone + posting to my own site in Safari is a good first start.
#aaronpkeyefi + mifi would mean the camera can upload via the cell connection any time which is cool, but still relies on eyefi.com
#aaronpkjust depends on how much you don't want to use their servers
#aaronpkI really don't mind because it's not like they store the photos or anything
#tommorrismy use case is this - back on 2014-03-29, I posted this https://tommorris.org/posts/8872 - I would have liked to have taken it with my decent camera (X-Pro1), then posted it to my own site from my phone, with the photo hosted on my own (sub)domain.
Sebastien-L, alanpearce, alanpearce_ and pauloppenheim joined the channel
#tommorrisooh. https://itunes.apple.com/gb/app/camerasync/id406645727?mt=8 - 1.99 GBP app that promises background image upload to S3. S3 lets you set up an SQS queue to notify you of every upload. this means you could actually have your website have access to all your photos and post them relatively easily.
#tommorrisI'll test it out. My plan is dump everything into an S3 inbox. then have some code that handles the file from S3 (makes sized images, notifies web server). ideally, I'd take a photo, then go to my site and it'd already be there ready to post into my stream.
#tommorristhat said, customer reviews on that app look pretty crap
#tantekso it must be somehow doable - just need to care
#sparveriusi mean if you hook your phone to a mifi, the expected behavior is that the phone's going to act like it's on wifi
#sparveriusand id rather have a slightly worse but consistent behavior than really smart but occasionally messes up
#tantekexpected behavior - whatever is best for the user :)
#reedstrmthere's also diff. use cases: live post just for personal archive/control, and live post to avoid loss-of-picture if loss-of-camera occurs (equipment failure or 3rd party action)
#sparveriusfor the former, as long as it happens eventually (within a day), i think it's fine
#reedstrmAn option to ship a low res as 'better than nothing' seems like a good thing.
#sparveriusfor the latter, the compressed version is part of the reason for that
#sparveriusi think as devices become more location aware, you'll be able to fold things like "if !connectedToFitbit then ohShitIveBeenLost()"
#sparveriusbut anyways tldr the reason ive been thinking about it is that i set up a new phone, and google plus photo always gets my compressed images, but i dont really think of those as "backed up"
#reedstrmhehe: I want to see the code for that later method :-)
#sparveriusand im using that doubletwist photo sync which seems to work well, but i'd love for it to be over the internet vs requiring that i'm at home
#sparveriusincidentally, my phone has a 20MP camera and can shoot 4k video, so bandwidth and cpu/battery usage for compression are becoming real issues :/
#reedstrmsadly, the pixel count wars have subsided in real cameras, but continue in phone-attached-cameras.
#Loqireedstrm meant to say: sadly, the pixel count wars have subsided in single purpose cameras, but continue in phone-attached-cameras.
#sparveriusreedstrm: it's a really good sensor, pixel count aside. they're using it to compensate for lack of physical zoom... for better or for worse :/
#sparveriusbut im seriously running into issues with self hosting regarding bandwidth usage
#reedstrmright - it's all about the glass (lens systems) theses days. That's an interesting trade, though. Esp. if they squeeze in OIS
#sparveriussony seems to think super highly of their camera software, so much so that if you unlock the bootloader, you lose the drm keys for it
#reedstrmSony thinks super highly of anything they license to you: and even your own stuff. cf: sony root kit
#sparveriusactually in the case of their phones, they've been extremely open, even sending code back to AOSP before releasing the new OS
#sparveriusim okay not having proprietary stabilization if it means i can have cyanogenmod and appops
#reedstrmHuh, that's good to hear. Typical big-corp with many personalities, I guess.
tilgovi joined the channel
#sparveriusyeah it's why i bought a sony. it was basically that or OPO or nexus or motorola, and none of them made phones that i could use with one hand
#bretmko: sweet! i'm not sure the target of the WM yet... maybe if I have a permalink page for a repo/branches on my site (with syndication links to GH)
#tantekand I didn't see anything captured on the wiki
#tantekso it can't have been that clever/important
#tantekor perhaps we thought we could re-brainstorm whatever we came up with
#bretmko: you wouldn't have to. but then again this is all getting super complicated.... just making a post with a single diff of a specific repo would be enough.
#brettantek: there is some documentation about how to mark up post update edits.. you and barnaby have that iirc
#tantekan actual edit/diff *post* itself would (should need) not contain the entirety of the original - though I suppose that's an easy but dumb way to do updates/edits.
#tantekjust a complete replacement e-content for whatever it is in-reply-to
#tantekperhaps if the in-reply-to URL had a fragment referring to a specific paragraph id, then the edit post could contain "only" the replacement for that entire paragraph
j12t joined the channel
#tantekperhaps if the in-reply-to URL had a fragmention referring to a specific phrase, then the edit post could contain *only* the replacement for that entire specific phrase
#tantekperhaps if the in-reply-to URL included *two* fragmentions referring to a specific start and end phrase, then the edit post could contain only the replacement for that entire range of content from start to end phrase
#reedstrmtantek: have been thinking about adding that very bit to an editor API we're using - eventually to support sync with simultaneous/collaborative editing
j12t joined the channel
#reedstrmgotta be careful about the underlying structure, of you do two pointers ...
#tantekwhich implies that the e-content of the edit post should replace the e-content of the in-reply-to URL
#tantekor lacking an e-content at the in-reply-to URL, the entirety of what is at that URL
#tantekno need for u-target - there's already u-in-reply-to
#tantekalso - why prematurely optimize the markup for just code? I'd think little prose wiki edits would be more common
#tantekhence why I documented those use cases on /edit
#tanteknote that by "simply" (re-)using u-in-reply-to, you can reply to /edit posts as well with another /edit post - to patch a patch as it were
#tantekand for any client that doesn't understand edit/update semantics, they'll still show it sensibly as a thread
brianloveswords joined the channel
#reedstrmtantek: I'm not seeing this as the serialization format for an edit-mechanism. Coders can talk in patches, much like maths type talk in mathematical statements and proofs. A patch can be a statement of it's own.
brianloveswords joined the channel
#reedstrmOr am I a rep. of a fading dinosaur class, for reading patches? :-)
#reedstrmright -perfectly readable for me, even though I don't know the codebase. I'd +1 that or sign off, or extend the discussion w/ a short form post rebmentioning back to it, etc. W/o having to apply it to a repo.
#tantekaaronpk, I'm not sure replicating a specific software's one-off textual diff format is a good longterm strategy
#tanteksaying, here's a diff text thing, good enough, is as bad as the pingback folks with there "… nonsense …" text
#aaronpki dunno, i read the diff format all the time when i'm coding, it's not that bad. certainly github's red/green coloring is better but...
#tantekexactly, and some people read MIME headers, or .vcf, or .ics or CSV, or some other one-off text format
#tantekand we don't pick those because the UX sucks
#Jeenatantek but doesn't github offter this nice frontend for diffs you're talking about when you say nice representation?
#aaronpkalso I suspect there are plenty of open source libraries for generating pretty html versions of diff listings
#tantekwe instead pick something where the UX is fairly *good* (competitive with the best in silos / software out there), and then make sure there's markup underneath to make it machine readable. not vice versa
#tantekJeena - then cite the new frontend views of diffs from github that you speak of
#aaronpkthat's basically waht i'm getting at with this whole exercise... started with github's diff view (link I gave previously)
#reedstrmtantek: you didn't read down: yes that's a man page format, but it's also IEEE Std 1003.1 2013, and give complete implementable details in the STDOUT section, complete w/ IEEE 'shall' language.
#tantekis that also a "changeset"? or is that something different?
#Jeenahm but it is not easy to find a description of the diff patch format
#tantekreedstrm - right, it was tl;dr, too many pages to scroll down to see anything relevant so it looked just like more software docs
#reedstrmJeena - I feel your pain. I think the links we both found are in fact it.
#aaronparecki.comedited /edit (+146) "/* Silo Examples */ add example of a github diff URL and clarify pull requests are a collection of commits" (view diff)
#bretthe the type of diff is really decided by the target... which knows what it represents to some degree... i.e. received a wm with a diff in it... I am a git repo.. therefore this can only be a git patch diff... or this is a diff.. I'm a note, so this can only be some kind of prose diff... apply if approved and valid format
#tantekbret - just look at all the design work that aaronpk has done to post nice human undertstandable aaronparecki.com/metrics - instead of just raw lists of lat/longs, or other CSV data dumps
#tanteksame reason you shouldn't be posting raw plain text diff files
#aaronpkyeah basically i'd add some tags and css to make it look like github. I think the mf2 parser would even result in the plaintext diff-compatible format as the "text" of the e-content
#Jeenahm my comment doesn't show up on aaron-gustafsons website, I wonder if this is my or his fault