#tantekthus if you wanted to embed some sort of markup for everyone else, you could either do so invisible (with empty abbrs, yuck), or experimentally with a nested HTML document in the 'data' attribute of an <object> tag.
#tantek(experimental, not part of how microformats2 are parsed right now, however could be if there enough demand / real world examples)
#tanteksnarfed, aaronpk the "Elsewhere online" is also a JS embed - I wasn't seeing it until you pointed it out and I allowed some scripts to execute
#danlykeI will keep my comments on the Zeldman piece to noting that I had to hit ctrl-- a few times to read it, and wish he didn't waste all that inter-line whitespace.
#KartikPrabhuimpossible to visual-design for everyone.
#tantekKartikPrabhu: especially for variances of visual impairment
#KartikPrabhuyup. but the new "em" units are very handy. and people can then zoom in/out or set browser text size as they want
#prtksxnaMost modern browsers are able to resize with px units too.
#KartikPrabhuprtksxna: yup. but it is good practice anyway
#danlykeSo has "em" in CSS been fixed? For me the way that's been completely broken has long been the most dramatic failure of CSS.
#tantekdanlyke what about it do you find "broken"? perhaps it's a documentation problem?
#KartikPrabhui don't find much broken about "em" in CSS. it behaves as it should I think (not tested in IE or Safari though)
#danlykeIt's never been either square or the width of an "M" (ie: the definition we used when I used to hand-set cold type), and on some of the stock fonts it ends up horrendously wrong. It's been years since I said "screw this, I'll never use it again" (and years since I touched CSS for more than minor tweaks), but it's never been usable as a base unit the way it should be.
#tantekI'm already itching for another post type. /quotation - when I want to like an article but also post a quote from it, without comment.
#tantekdanlyke: that sounds like you have a problem with particular font's metrics rather than CSS
#tantekem is defined very simply in CSS as whatever the font-size is set to
#tantekso when you specify other things in ems, it's just math
#tantekems are definitely square when used (vertical or horizontal equivalence)
#tantekthe width of an "M" thing is a font-metric problem, not a CSS problem
#tantekparticular fonts can be broken that way, but not CSS
#danlykeI remember thinking that's the way it should work (ie: set to the font size, it's square), but didn't. It could simply be that the core MS fonts were largely broken.
#danlykeMy only "wish I could easily override" is line spacing, since most "designers" slap way too much whitespace on to their pages for me.
#tantekI'm going to try to like an article with an h-cite and see what happens
#KartikPrabhudanlyke: maybe the line-height thing could be fixed by some browser setting. that would be nifty... I would use it to increase line space on most sites though ;)
#danlykeKartikPrabhu see, if I had my druthers I'd still be reading sites in Lynx. JavaScript broke the web.
#tantekdanlyke - some of us are still pushing for plain text first design ;)
#KartikPrabhudanlyke: blame the handyman not the tools :) JS is way over done
#danlykeKartikPrabhu it's kinda like the cyclist's view of automobiles: Sure, we can blame individual drivers, but in the end it's the entire culture of careless endangerment that's the problem. So it is with web design.
#KartikPrabhudanlyke: yes. the "culture" is the problem. but it seems to be going back at least a bit now. Most web design experts seem to be pushing heavily for progressive enhancement
#tantekhowever most web devs appear to be pushing for single page apps and JS all the thigns.
#danlykeYeah. And, let's face it, Facebook and Twitter are indeed beating the blogs. Just because I hate humanity's preferences doesn't mean it isn't the direction of the masses. Sigh.
#tantekdanlyke: blogs fell behind (functionality) FB & Twitter, so that's no surprise. that's what we're fixing. :)
#danlyketantek yeah, i just need the occasional reminder :-)
#danlykeG'night, all, I'm gonna shut this down and head home.
#KartikPrabhuCSS does define units very thoroughly and they are different from what print people would expect. E.g. a pixel is not a physical pixel. But the CSS definition is pretty sound
#tantekwell until I can post a /quotation I'll just post a /like for now and deal with the fact that a like of a non-tweet does not get POSSEd to Twitter (because that would get annoying fast I suspect for Twitter followers)
#kylewmif I asked what's actually wrong with js frameworks and client-side rendering, is that like asking if vi is better than emacs?
#KartikPrabhukylewm: also JS breaks much more easily being a programming language.
#tantekkylewm: the canonical example is games - where without dynamic interaction it doesn't make sense. however, for many games (e.g. turn based ) - there's no need for JS to make it "work"
#KartikPrabhuif you forget a semi-colon, boom no content
#KartikPrabhusnarfed: true. but that was only illustrative. many other things could go wrong. Some browser does not support the latest ECMA script functions and again no content
#tantekKartikPrabhu: last time adactio was in SF I posited a defn for web app that made him go hmm (didn't agree nor disagree, just made him think)
#KartikPrabhutantek: oh! that would be good to know/post
#snarfedKartikPrabhu: sure, agreed, browser compat is no fun. but the reasonable fallback is progressive enhancement, not not content. no sense in arguing against the dumbest version :P
#KartikPrabhusnarfed: yes. JS with prog.enhance is good definitely
#kylewmi just read another essay about how Python 3 is XHTML
#KartikPrabhui use it for marginalia.js so I can't say no JS ever :P
#KartikPrabhuand I also use it to load images on my site so...
#kylewm"Because as it stands, Python 3 is the XHTML of the programming language world. It's incompatible to what it tries to replace but does not offer much besides being more “correct”."
#KartikPrabhukylewm: from your own code, I have seen that it is posible to write pretty robust code to support both pythons
#tantekkylewm: sigh. they mean it's the XHTML2 - not XHTML
#KartikPrabhusimilarly JS should be written as an enhancement to HTML+CSS. I see no good argument against that.
#tantekhey at least you *have* a theme. I don't even have a theme (abstraction), much less even templates (except for my home page, which is built from a template).
#prtksxnaAlso, I am not doing status messages right. I am using the wordpress title instead of the content. That is stupid because it doesn't let me add links and stuff…
#tantekaaronpk - indeed. It's very tempting to re-use/repurpose already implemented post types for additional post types, and they sorta can work that way too.
eburcat joined the channel
#tantekhowever there are the dual problems of capturing author intent - i.e. what's the *focus* of the author's post, and explicitly allowing for / hinting at / encouraging post kind specific display
#tantekin those ways I think a /quotation post type deserves explicitly different treatment than a /bookmark
#tantekthat is - the *quote* being the focus, the primary content the author wishes to convey, and the URL just being a citation / info-about the quote.
#tantekrather than the *bookmark* or URL being the focus, and a description / quote is meant purely as sort of an optional mini-preview / snippet for the reader's convenience.
#tantekthe intent behind those two are very different, and should at least *allow for* different presentation (whether or not and how they are presented differently is a different matter, so to speak)
#tantekTumblr's quotation post type is a good example of how the focus on the quote can be better conveyed in a design.
#tantekback to clustered like presentation iterations - specifically the thorn that's sticking is the way time is conveyed
#tantekI realized that a parenthetical expression of time after a piece of media/link already has some shared meaning, that it's a duration indicator/hint about the media/link being shared (expectedly audio/video)
#tantekwhich is obviously not what my usage is intending, thus I need to alter the presentation to not misconvey that somewhat established convention
#tantekanyway that's going to take some rethinking
#tantekthe other clustered like case I'm looking into improving is when I like/favorite a lot of things from the same person / source
#tantekpart of my reading pattern is reading individual people's profiles / sites serially post by post, and often liking a bunch of them in a row
#tantek(I find reading people like that much more satisfying / calm than any kind of random aggregate "news feed" of updates from all the random sources)
#neuro`tantek: can you have some /people/someone pages gathering those individuals posts?
#neuro`It would mean being able to extract the origin of the posts you liked, but this must be in the various APIs metadata
#tantekneuro`: perhaps. For now I'm looking at the simpler question of can clusters of likes in a home page stream be even further tightened up
#tantekno APIs nor metadata needed. origin of posts - what we call *author* of posts (h-entry) is right there in the visible markup
#tantekor even more minimally, leave out explicit authorship in such posts - just use a permalink/URL and presume that anyone wanting detailed info about a post you like, can simply retrieve that post and follow the /authorship algorithm on it
#voxpelli!tell KevinMarks I pushed a quick fix now for the duplicated text in the mentions on your site – it displays a sensible default now rather than the imported like/repost text
#tantekthat quotation with a photo to me seems more like the primary content / intent is to show the *photo* and then the quotation is intended secondarily as the caption of the photo.
#tantekthat is the quotation is about the photo, rather than the photo being about the quotation
#aaronpkoh funny, I thought that was a tweet reply
#tantekdo you have the entire original? did you do the ellipsing or was it in the original?
#tantek… inside a quotation is supposed to indicate that the ellipsis was in the original quotation, whereas […], like any [words inserted] insidde a quotation implies insertion (deletion) by the quoter.
#tantekwaits for a "View Cache" button on comments on aaronpk's posts ;)
#aaronpklooks like most of the page included the ellipsis
#aaronpkoh that'd be neat. not sure how i'd handle CSS though
#aaronpkthe version in the fb og tag was "fwd @timberners_lee RT @aaronpk "Because already got webmentions on my site, Brid.gy worked straight out the gate... j.mp/U5UH3O"
#aaronpkand in the entry-content element there was "fwd <a href="http://twitter.com/timberners_lee">@timberners_lee</a> RT <a href="http://twitter.com/aaronpk">@aaronpk</a> “Because already got webmentions on my site, Brid.gy worked straight out the gate… <a href="http://j.mp/U5UH3O">j.mp/U5UH3O</a>"
#tantekright, no summary means you have to pick between name to keep the comment short for presentation or content for more but which risks making the comment much bigger (off topic?) than your post.
#tantekhence name before content without knowing anything more about either's specific value
#tantekwhen displaying a name or summary for a comment, and the comment has content, AND the name/summary being used is a strict whitespace normalized text abbreviation for the content, show a "More…" link which dynamically shows the entire content in place instead of the name/summary
#Loqitantek meant to say: when displaying a name or summary for a comment, and the comment has content, AND the name/summary being used is a strict whitespace normalized text prefix abbreviation for the content, show a "More…" link which dynamically shows the entire content in place instead of the name/summary
#tantekthe algorithm is, text only, collapse all sequences of any whitespace to single whitespace characters
#tantekdo that to both name/summary and content and then you can do a simple string prefix compare
#tantekthen if you're worried about ellipsis, think of it as a *choice* for you the publisher
#tantekthat is, you could implement a "More…" button *ONLY* when the name/summary ends with an ellipsis
#tantek(in addition to the previous prefix requirement)
#tanteksince the comment author's use of ellipsis in the name/summary indicates that they are explicitly communicating in the name/summary that there IS more content
#tantekthus you as the comment displayer could choose to listen to that explicit communication, and show a "more…" link accordingly.
#tantekas cumbersome as it was, OpenID 1.0 didn't require special sauce to consume or produce.
#aaronpkI've been working on a blog post about OAuth 2.0 implementations. It's a checklist that points out all the things where OAuth 2.0 implementations can differ
#aaronpkthis is more pointing out all the holes in the spec
#aaronpkalso is kind of a guide for people building oauth servers
#tantekthat makes sense, even such a "all the holes" guide on its own is useful
#tantekthen you can separately post real world examples of the holes in practice
#aaronpkyeah. I'm toying with the idea of making this a live form, so people can fill it out based on their server implementation. then I could collect results that way
#aaronpki'm not sure it would actually be super useful and is going to add a ton of work to the post, so haven't decided yet
#tanteknot sure people could be trusted to answer accurately (not just because of malintent)
#tantek.comedited /quotation (+2337) "why section, primary / portion / emphasis exception, and when to instead use a repost, bookmark, or photo" (view diff)
#tantekand I think the primary/secondary reasoning still applies
#aaronpkthat example seems more like clarifying the context rather than adding commentary
#aaronpkso I agree that the reasoning for using a quotation post applies
#tantekif your post is primarily about the quotation, and only secondarily about any summary / comment / description, then it should still be a quotation
#tantekif however, your comment *is* your primary contribution, then are you are writing a reply, and the quotation is actually acting as just a reply-context. Post a reply instead, in-reply-to a fragmention link to the start of the quoted content.
#tantekaaronpk - hah - that's a hashtag - as commentary
#tantek.comedited /quotation (+2569) "subheads for why, secondary brief commentary, when to use a reply instead, with examples, note connection to marginalia" (view diff)
#tantekWhat I find particularly interesting is that I needed the conceptual building blocks of "reply-context" "fragment" and "marginalia" to actually answer the seemingly "simpler" question about quotation vs. reply.
#Loqitantek meant to say: What I find particularly interesting is that I needed the conceptual building blocks of "reply-context" "fragmention" and "marginalia" to actually answer the seemingly "simpler" question about quotation vs. reply.
#aaronpkalso it was the act of scanning through past notes on my site looking for these examples that made me realize I already have a lot of the reader work done
#aaronpkin that case the quotation is there so that I remember why I bookmarked it
#tantekhowever, upon reading the the linked article, I *think* the bookmark is more primary, and the quotation is just a secondary "reminder" or "summary" of what kind of thing is in the bookmark, or *why* you bookmarked it.
#aaronpki suppose accepting contributions is easier for CC0 source code since you're asking very clearly for the contributor to give up their copyrights rather than assigning them or licensing them
#tantekbecause I started CASSIS as CC-BY-SA deliberately to slow down / constrain some of its propagation until I felt like there (enough) test cases, review of the code etc.
#tantekaaronpk, and yes, for contributions, CC0 is a no brainer
#tantekbecause it doesn't make them *give* anything to *you* in particular
#tantekthat's the problem with all the copyright assignment/license crap
#aaronpkinteresting, hadn't really considered it before
#tantekit limits your opportunity to contribute to multiple places that each ask for that
#tantekwhat CC-BY-SA gives you above CC0 for such "uncertain" code is that you're communicating an expectation that derivative works MUST also be openly published, which means if you find a bug in your original, you have *a chance* of finding (or other finding) downstream copies and fixing
#tantekwhereas with CC0, you basically absolve yourself of any such responsibilities
#tantekanother reason is if you're doing a project essentially by yourself and you have some fear someone may try to hijack it e.g. incorporate it in its entirety into some larger project and then claim its theirs, take it over, etc.
#tantekusing CC-BY-SA minimizes that to only other larger projects that use CC-BY-SA - which are not typically not the type of people/communities that hijack things
#tantekbtw by take it over - I also mean put a more restrictive license on it that causes you trouble with working on your own code - since fixes may become "obvious" and they may make fixes before you do.
#tantekanyway that's borderline off topic - or perhaps it's an FAQ for indieweb publishing of code
#tantekkylewm - re: u-syndication links for POSSEs of likes - I will point out that FB *does* have permalinks for likes
#tantekthus it *does* make sense to link from an indie like post of a FB object, to the FB permalink of the POSSE like of that object.
#tantekhmm I think I just made myself more work - since I don't think Bridgy Publish returns the FB permalink of the POSSE like of a FB object, does it?
#snarfedalso those permalinks redirect to the post url
#tantekkylewm: not for me. however I assume that Bridgy sees the URL permalink of a like as some sort of return value to its FB API call to like something
#snarfedactually, nm, they don't redirect but they do just show it
#kylewmtantek: can you share one of the like permalinks you are looking at?
#snarfedkylewm: ok, i understand now. i'd love to use those urls! i'm not actually sure we get them, but i'd happily use them if possible
#GWGThe issue being that Indieweb conventions see no difference between the two and WordPress does. Trying to decide how to address in a presentation manner
#aaronpkkylewm: what happened to your mf2? your body tag has a class of bh-entry
#tantekGWG, simple answer, WP categories are an old obsolete methodology and architecture. ignore them completely. pretend they don't exist. Use only tags.
#tantekAll I know is that we're documenting more real world post types with actual citations of examples in the wild than the ActivityStreams crowd ever did :P
#tantekI think proposing "This" and "That" posts as additions to W3C Activity Streams would be good April 1 submission.
#voxpelliworks well with kylewm's "like" endpoint, although perhaps one would like a bit quicker publish flow there – like a 1-click confirm rather than a form?
#Loqireal-time comments are the display of comments on a view of a post (typically on its permalink) as they are received by the post (presumably via webmention), without needing any explicit user reloading, refreshing, or any other tapping/clicking etc https://indiewebcamp.com/real-time_comments
#tantekclick empty heart, send web+action URL handler, your server posts like to your own site, sends webmention to the post you're looking it, post you're looking at updates its appearance to show the heart is now full
j12t joined the channel
#tantekah, in that case you should at least show up in the likes facepile
#kylewmi'm with you until the last step -- how could it figure out to fill in your heart in addition to add you to the facepile
#tommorrisat some point, we should probably get the VisualEditor onto IWC (and maybe microformats) wiki. VisualEditor promises less painful diffs and happier non-technical users. :)
#tantek.comedited /POSSE (+739) "add subheads to other approaches, add pesetas, incorporate notes about Tumblr to Twitter in pesetas section" (view diff)
#tantekthat page needed to have an explicit reference to PESETAS anyway under Other Approaches
#tommorrisalso, Tumblr does allow you to map your own domain to it. as does Squarespace.
#tommorrisI should braindump some notes about Squarespace. I used it for a family website recently.