j12t, reedstrm, dariusdunlap, edsu, daf and lukebrooker joined the channel
# 00:18 dariusdunlap I had thought I would come up to Homebrew Website Club tonight, but I think I will go home and _work_ on my website instead.
j12t_ and mlncn-agaric joined the channel
lukebrooker, gRegor` and KartikPrabhu joined the channel
# 00:48 gRegor` Hello from Chicago HWC
# 00:48 gRegor` Ho-ho-hombrew Website Club!
# 00:48 Loqi gRegor` meant to say: Ho-ho-homebrew Website Club!
# 00:51 snarfed so aaronpk, kylewm, i'm thinking about invitee markup in bridgy
# 00:51 snarfed fb's api doesn't expose who invited people, just that they've been invited
# 00:51 snarfed i could just say the event owner invited everyone…or i could leave the inviter blank
# 00:59 kylewm snarfed: this might be controversial, what would you think about dropping invites altogether?
# 01:01 kylewm well I don't know, there's no real harm in having bridgy send a bunch of notifications and letting hte receiver decide whether to ignore them
# 01:01 snarfed i'm actually not sure that 321 is such a big hurdle
# 01:03 snarfed just compare a few response fields and resend if they've changed
# 01:04 snarfed dropping invites would definitely fix it for free though
# 01:07 kylewm I thought maybe it was an issue where you tell facebook "give me updates more recent than ID #####" like the twitter api
# 01:08 kylewm to actually answer your question, I'd say leave the inviter blank... better to be honest that we don't know than to say something that might be wrong
# 01:09 snarfed kylewm: yup, we do do that "more recent than" w/fb. i just don't trust that it uses updated time as opposed to created
# 01:09 snarfed that's just fud though, we can obviously find out
# 01:10 kylewm ok, so yeah if FB already tells bridgy that the item has been updated, and bridgy ignores it because it's already seen it, that would be much easier
# 01:11 kylewm than re-polling old items to see if they've changed
prtksxna joined the channel
KartikPrabhu and tantek joined the channel
demis joined the channel
amblin joined the channel
carlo_au joined the channel
# 02:10 pdurbin unfortunately, the week that I was in SF there was no HWC :(
# 02:11 tantek pdurbin - wait what? bummer. which week was that?
mdik_ joined the channel
# 02:31 tantek and looks like we'll be coming back to Cambridge soon!
# 02:32 tantek and if the dates sound good to you add yourself
# 02:32 tantek alright, we're going to get started here in SF
# 02:33 pdurbin huh. thursday and friday. not the weekend like last time in cambridge
jjuran joined the channel
# 02:34 pdurbin whelp. i put in in my calendar at least. "tentative" for now. thanks for the heads up, tantek
# 02:35 aaronpk !tell snarfed what would happen if I were to tell webmention.io to re-parse all the old webmentions so I can fill in the mf2 data? would I cause you to hit any rate limits?
# 02:36 Loqi Ok, I'll tell them that when I see them next
# 02:39 tantek tantek: personal itch, have been running NoScript, turns out Twitter's web UI doesn't work to favorite tweets without JS
# 02:40 tantek … so my 2015-01-01 launch goal is to implement favoriting from my site, which then would POSSE the faves to Twitter
# 02:41 tantek pius: I'm going to put some VR content on my personal site
# 02:42 tantek … I just want to put something up and start there
# 02:42 Loqi tantek meant to say: … maybe a 3d scan maybe something else
# 02:44 jjuran My sites are jjuran.org, metamage.com, and others
# 02:44 tantek jjuran: working on a portable distributed programming system
# 02:44 tantek … for which leading competitor is the web platform
# 02:45 tantek … one of the core tech is network file sharing protocol
# 02:45 tantek … wants to build a communication system on top of it
# 02:45 tantek … what would an alternative to the web look like?
# 02:45 tantek … file protocol is designed to work as a virtual file system, like Plan9
# 02:46 tantek … one of the key pieces for that is implementing TLS first
# 02:47 tantek … it's a binary protocol and gives you filing semantics, list files, open a file
# 02:49 tantek Brett Slatkin: I'd like to get https on my site
# 02:49 tantek … you can now get certs without having to fax a driver's license
# 02:52 tantek … still make available over http as well, for older devices, perhaps with warning content
# 02:55 tantek … I don't checkin often, but I want to do more of those
# 02:56 tantek … I do them manually now, want to do them (semi-)automatically
# 02:56 tantek … for bigposts, a creation UI hurdle is not a big deal
# 02:56 tantek … but for little things, likes, favorites, checkins, any UI friction takes more work than worth it
# 02:57 tantek … owning my own likes, that UI friction where I can favorite or retweet something easily was key. like are down 4 taps and 10-15 seconds, on mobile
# 02:57 tantek … still not as good as one tap under a second which is what silos do
# 02:57 tantek … has a bookmark now for any venue / event where it prefills everything for him in a checkin post
# 02:57 tantek … POSSE to facebook is difficult - mapping to their venues
# 02:58 tantek … would like to figure out something to also do the appropriate checkin on FB
# 02:59 tantek … last time I talked about working on citations on the web
# 02:59 tantek … maybe I can show you all by 2015-01-01 a working citeproc that generates HTML citations with microformats
# 03:00 tantek … by 2015-01-01 at least one h-cite post on my site. but no new papers :)
# 03:01 tantek Art - domain is aribanatal . (didn't quite get it)
# 03:01 tantek … 1. taking projects I've done in the past, and moving them to my own website
# 03:02 tantek … 2. playing around with Federated Wiki, trying to figure out what version of that makes sense to use
# 03:02 tantek … getting some version of that running and then changing the frontend
# 03:02 tantek … just a styling thing I want to figure out how to fix
# 03:03 tantek … where I have some external data about the post, date posted, location, etc.
# 03:03 tantek … it's all really cluttered, click targets small
# 03:03 tantek … I want to restyle it so that it's easy to see and click
# 03:04 tantek Katie Johnson: would like to add a blog to katiejohnson.me
# 03:04 tantek Brett: I'm trying to subscribe to everyone's site right here
indie-visitor joined the channel
# 03:06 Loqi Welcome, indie-visitor! Set your nickname by typing /nick yourname
# 03:12 aaronpk emmakuo - notenoughneon.com has been reading exclusively from her own site
# 03:13 aaronpk currently she is only fetching peoples' home page feeds, and not fetching content from the invidual permalinks
# 03:13 aaronpk this means sometimes she only gets article summaries for example, which is not necessarily a bad thing
# 03:14 aaronpk other issues she has found: most people don't put full h-cards on each item on their home page
# 03:14 aaronpk (yes, but currently documenting issues with consuming :)
# 03:15 aaronpk currently she did imlement representative h-card algorithm, except for when it requires additional http requests.
# 03:17 aaronpk currently storing remote data as HTML files on disk after they have been parsed, one file per entry
jonpierce joined the channel
# 03:19 aaronpk yeah she is going as far as looking at everythign on the page without fetching additional pages
# 03:20 aaronpk some people's home page h-cards are not marked up in a way they can be found by the authorship algorithm
pius joined the channel
# 03:22 aaronpk hmm maybe not, kylewm I see your photo URL in your h-feed (which is the thing that's missing from her reader right now)
# 03:23 aaronpk pdx: now discussing sanitizing markup for readers and reply context
# 03:24 aaronpk aaronpk's implementation of reply context uses the plaintext versino of the post, then re-links things like URLs, @-names (based on his own /nicknames-cache) and hashtags
# 03:24 aaronpk but that means we lose things like <a> tags, or embedded images / videos
# 03:24 aaronpk which may not be bad for reply context, but can't use the same rules for readers
martinBrown, amblin and kajohnson joined the channel
# 03:26 aaronpk for readers we need more of a sanitizing version so we don't lose as much content in the reader view
pius and martinBrown joined the channel
# 03:30 aaronpk "if everything were as simple as webmention, this would be easy"
KartikPrabhu joined the channel
# 03:34 tantek aaronpk, it would be interesting to see specific "how do I do x?" questions where X is what is being assert as "not easy" etc.
# 03:35 Loqi tantek meant to say: aaronpk, it would be interesting to see specific "how do I do x?" questions where X is what is being asserted as "not easy" etc.
# 03:35 aaronpk one question was "how do I know two URLs are equivalent?" where some issues encountered were a permalink on https was saying that its url was http, and the combination of some sites redirecting http -> https or not
# 03:36 aaronpk ok the reason kylewm's photos are not showing up with the representative h-card algorithm is because:
# 03:37 aaronpk a) kylewm.com needs a rel=me link to kylewm.com, or,
# 03:37 aaronpk b) the primary h-card needs a uid property set to kylewm.com
# 03:37 aaronpk tantek.com is looking good in this reader because his h-card has both the url and uid property set.
kajohnson_ joined the channel
# 03:41 tantek we're going to 21st Amendment for continued discussions over drinks & eats
# 03:41 tantek plus apparently there's an XOXO meetup at Virgil's Sea Room til 22:0
# 03:42 Loqi tantek meant to say: plus apparently there's an XOXO meetup at Virgil's Sea Room til 22:00
# 03:42 aaronpk more feedback: emmakuo didn't even know that she *should* publish an h-card a certain way to be found by the authorship/representative algorithm because that information wasn't discoverable in the 'getting started' page
# 03:42 tantek but got removed in a "simplification" step I think :/
# 03:42 aaronpk another consideration is some readers may or may not actually follow post URLs, so authors should consider both possibilities when publishing a feed page
# 03:43 aaronpk consuming reposts: was having issues getting the context for getting the original post. similar for replies. (mostly due to problems with peopels' markup)
cmhobbs, martinBrown and prtksxna joined the channel
kerosene joined the channel
KartikPrabhu, mlncn-agaric, j12t, gRegor` and martinBrown joined the channel
lukebrooker, j12t, snarfed, fakehenson and jjuran joined the channel
j12t, tantek and jonpierce joined the channel
LauraJ, j12t, prtksxna, pfefferle, julian`` and eschnou joined the channel
LauraJ and wolftune joined the channel
j12t, petermolnar, loic_m, Sebastien-L, pfefferle and michielbdejong joined the channel
Sebastien-L, friedcell, eschnou, stream7, adactio, j12t and prtksxna joined the channel
Sebastien-L and pfefferle joined the channel
LauraJ and j12t joined the channel
marcthiele joined the channel
pfefferle and adactio joined the channel
wolftune and mlncn-agaric joined the channel
j12t joined the channel
friedcell, sanduhrs and eschnou joined the channel
danlyke joined the channel
j12t, LauraJ and reedstrm joined the channel
# 14:41 pfefferle GWG yes, sorry! my son thought he has to come a bit earlier than expected…
# 14:42 Loqi Ok, I'll tell them that when I see them next
# 14:42 GWG pfefferle: No sorry required. I was just pleased to see you around.
# 14:43 pfefferle GWG I will be a bit more present the next year… hopefully :)
# 14:45 GWG pfefferle: I had to rebuild my server, which has been a delay. I took a side detour into an improved cache invalidation plugin for the method I use.
# 14:47 GWG But some minor improvements. I've been adding functions using php-mf2 to the hcard-tools plugin to get the representative h-card and thus the profile picture based on a URL.
verdi_ joined the channel
# 14:54 pfefferle GWG nice! have to update and test your progress, if I find some spare time between the years…
# 14:55 GWG pfefferle: Never enough time for me either.
# 14:56 pfefferle GWG yes… we need a employer that pays us for developing indieweb plugins ;)
# 14:57 GWG I'd settle for an employer that actually told me what they were doing before they did it.
# 14:58 GWG petermolnar: They exist in our minds.
# 14:59 GWG In the meantime, I've been driving to/from work wondering if I should have stuck with my original idea and done custom post types in addition to a custom taxonomy.
jonnybarnes joined the channel
# 15:07 GWG I originally decided against it because it moves farther away from the WordPress experience, but there are reasons why that might be good
# 15:07 GWG I'm not doing it right now though. I'm going to try to solve my issues within the post type
nloadholtes, j12t, snarfed, gRegor`, mlncn-agaric, chalettu, LauraJ, danlyke and fiatjaf joined the channel
prtksxna and tantek joined the channel
yakker, mlncn-agaric and snarfed joined the channel
# 17:27 tantek basic summary: nearly everyone made a verbal commitment for what they're going to ship on their own site on 2015-01-01 - noted in the logs
# 17:28 GWG ben_thatmustbeme: I am still searching for a HWC person to join me in NYC
# 17:28 GWG ben_thatmustbeme: I put a note on the wiki
# 17:29 tantek ben_thatmustbeme, GWG: what feature/functionality are you going to ship on you own site on 2015-01-01?
# 17:29 tantek new year, new feature, new declaration of independence
# 17:29 GWG Oh, good. Because I'm working 8:45-5:15 that day
# 17:32 GWG I think I just had an idea to solve the last problem on my list
# 17:32 GWG I may need to run this by a WordPress person
# 17:33 ben_thatmustbeme hmm, not sure I'll have much free time before then, so I'll set a reasonable goal of getting checkins working finally
# 17:33 tantek GWG, I picked 2015-01-01 as a proposed ship commitment date since I shipped "notes" on my own site on 2010-01-01
# 17:34 tantek ben_thatmustbeme: getting checkins working is a GREAT commitment for 2015-01-01
# 17:34 GWG tantek: I have to see what I can commit to.
# 17:34 tantek we had a bunch of checkins related discussions last night
# 17:34 tantek GWG - note that it's ok to pick a small thing too
# 17:34 tantek e.g. I committed to shipping "likes/favorites" on my own site, so I can post them myself on tantek.com, and POSSE out to Twitter, instead of using Twitter's UI to favorite tweets.
# 17:35 tantek ben_thatmustbeme: you'll see it in the logs, but heads-up, snarfed also committed to making his checkin UI better (more automatic) so that he actually uses it more often (right now he has to do a lot manually).
# 17:36 tantek and I've been tracking various /checkin efforts too - we reviewed existing IndieWeb Examples of checkins
# 17:36 tantek e.g. both mko and kylewm have really beautiful presentation of their checkins.
# 17:37 tantek part of the (implied) goal of getting a post type working like that (likes or checkins etc.) is that as of 2015-01-01 you do it *exclusively* from your site, and only on silos via POSSE
# 17:37 tantek to be clear, that's what I'm talking about when I say getting likes/favorites posts working on my site
# 17:38 tantek ben_thatmustbeme: so I want to encourage you to be clear about that too (if you use Foursquare/Swarm or Facebook or some other checkin silo)
# 17:38 tantek in snarfed's case, he only uses Facebook for checkins now, so his challenge is to figure out a way to make POSSE checkins to Facebook work at least semi-automatically (he does it fully manually now)
# 17:39 tantek off to go to a meeting - hope to see y'all post your commitments here in #indiewebcamp, and even better, blog your 2015-01-01 ship commitment on your own site!
# 17:39 ben_thatmustbeme I never checkin with anything anymore. I still use FB for replies. I just usually find it too much trouble to grab the link to a post from my phone to try and reply that way
# 17:40 GWG I'm thinking of renaming Indieweb Taxonomy.
# 17:42 GWG While names aren't the most important thing, it seems like that doesn't really describe it.
# 17:44 GWG I'm thinking of changing it to Indieweb Post Formats.
# 17:44 GWG Because it mirrors the Post Format feature in WordPress, which is a built-in taxonomy
# 17:51 aaronpk tantek: I like that theme for the last HWC of the year!
# 17:53 tantek aaronpk - not too late! you can make a 2015-01-01 ship commitment here and blog it - and encourage bret and the other HWC PDX attendees to do so too!
# 18:02 aaronpk maybe mine should be get my site moved off my crappy old server
# 18:02 aaronpk the visible portion of that will be I'll get fewer SSL errors connecting to other peoples' sites when fetching reply contexts
# 18:03 aaronpk the more ambitious but fun thing would be checkins. although that's kind of unrealistic at this point.
# 18:03 aaronpk OR... change my permalink structure to drop the post type from them
wolftune joined the channel
colintedford, j12t, lupinedev, crossdiver and mlncn-agaric joined the channel
# 18:32 kylewm I was challenged last night to finally get rid of the mixed content warning on my site, that is moving up my priority list
j12t joined the channel
scor joined the channel
benjamin-agaric joined the channel
agaric joined the channel
# 19:11 ben_thatmustbeme although I suppose I could drop the trailing / on my posts instead of just leaving them, its actually nicer this way i think..... only I just realized... crap
# 19:12 ben_thatmustbeme The way i designed it, you can use just about anything for the slug and it also returns the same content... that doesn't redirect
# 19:13 aaronpk my URLs work where whatever you put in the slug will redirect to the canonical slug. a nice side effect is that I can change the slug and everything Just Works™
# 19:13 kylewm aaronpk: do you think it's important for the image proxy to have some sort of hash, to prevent unauthorized people from using it?
# 19:14 kylewm will has it configured to only mirror images that originate on his own site, much less possibility for malicious use
# 19:14 aaronpk kylewm: one thought I had was running this proxy as a private service, mostly using it to just download the image. my website would use it, downlaod the image to disk somewhere, then serve those URLs in the HTML.
# 19:14 aaronpk but that's a lot more work than just writing the img tag
# 19:16 aaronpk another option would be adding a sort of "signature" verification
# 19:16 aaronpk it would require hacking into the imageproxy source
# 19:16 aaronpk if my server and imageproxy shared a secret, my server would have to send the signature along with the request. that way only my server would be able to generate the URLs that imageproxy would process
# 19:17 aaronpk I kind of like that option now that I think about it
# 19:18 kylewm ^ that's basically what you are thinking, right?
# 19:22 aaronpk eh, i wouldn't bother. how would you end up with an all-caps URL somewhere?
# 19:24 aaronpk kylewm: another thing I was considering is adding a timestamp to the URL
# 19:24 aaronpk because I want the avatar of someone who commented in 2013 to always look like it did at that point, regardless of what happens to the actual avatar URL later (disappears, changes, etc)
# 19:24 Loqi gives aaronpk the avatar of someone who commented in
# 19:25 kylewm oh interesting, I like that idea, but would you end up with a different copy of their avi for every comment?
# 19:25 aaronpk possibly. it depends on how granular your timestamps are
kajohnson_ joined the channel
# 19:25 aaronpk also it's like a 40kb file, so not really the end of the world even if you have a bunch of duplicates
# 19:26 aaronpk i'd say de-duping the storage of the files is an optimization that can be done later. storage is cheap.
# 19:27 aaronpk that could be handled by the image proxy pretty well too
# 19:29 aaronpk so I might have a URL like aaronpk.com/img/
# 19:29 aaronpk the proxy would go check if there's an updated version of the image since the last time it stored one
# 19:29 aaronpk if it's the same, then it could just return the previous version
# 19:30 aaronpk so no matter how many timestamps you give it, it would jhust store the one latest image
# 19:31 aaronpk does that makes sense? I may not be describing this well...
michielbdejong joined the channel
# 19:32 aaronpk it does require the proxy storing state, or at least symlinks...
# 19:32 kylewm hmm, trying to figure out what would happen if you asked for 10:00, then 12:00, then 11:00
# 19:33 kylewm i guess the assumption is that you always either request a previously requestsed timestamp, or a timestamp later than every other one
# 19:34 aaronpk and the failure case is that you get a newer image for an older timestamp, right?
# 19:34 aaronpk I wonder if I can learn enough Go to add that to will's thing
# 19:38 aaronpk oh actually, in the 10:00 -> 12:00 -> 11:00 case what would happen is the 11:00 URL would show the picture from 10:00
# 19:39 aaronpk and assuming the image from 12:00 is different, there isn't actually a way to know when it changed, so it may or may not be correct
# 19:39 aaronpk but there's no well-defined behavior of which image to show at the 11:00 URL
tantek and KartikPrabhu joined the channel
# 19:50 ben_thatmustbeme okay, i think i have it worked out so that only the exact canonical URL will return 200, the others will redirect to it
# 19:51 bret marc andreessen needs to learn how to POSSE a blog post to twitter
wolftune and jjuran joined the channel
# 20:32 tantek bret - I think there's a Tumblr that PESETAS marc's chunks of tweets into blog posts
# 20:33 tantek ben_thatmustbeme: looks like there is no "why" section for /sitemap - thus indicating no currently documented reasons to build one.
# 20:33 aaronpk my main motivation for building a sitemap was so that google could index the site better
# 20:34 tantek aaronpk - since Google has demonstrated worse indexing this year, I don't see any reason to try to do anything special for them.
# 20:35 tantek aaronpk - there are many support google answers that *prescribe* what to do - there is little evidence that Google provides that *show* that what you do makes any difference.
# 20:35 ben_thatmustbeme actually the one issue of my old content not being surfaced would be a problem, i currently don't have any way to really go back into the archives
# 20:36 aaronpk at least with the twitter cards, they provide a validator that actually shows what the result will be, and then you can immediately see the effect later
# 20:36 tantek that is - the prescriptions provided by those support google answers are purely *hypothetical* until/unless someone provides concrete before/after data/results
# 20:38 tantek aaronpk - the twitter card validator is *better*, but like the Google Structured Data Testing Tool, is still somewhat theoretical, in that you don't actually get to see results right away on Twitter.com
# 20:39 aaronpk not right away, but as soon as you get the confirmation email after they manually approve you, the results are apparenty
# 20:39 tantek the showing of Twitter Cards on twitter.com tweets requires that they explicitly put you / your site on a whitelist, which we don't know how long it will take
# 20:39 tantek they send you an explicit confirmation email? well that's better than Google at least which does not
# 20:39 ben_thatmustbeme hmm, i should try a specific search against my site to find something that has fallen off the main page. then retry a week after creating a sitemap
# 20:40 aaronpk i've never seen any evidence one way or the other that a sitemap has any effect on google indexing my content
# 20:40 tantek contrast FB /OGP - which works *immediately* on FB.com as soon as you add the meta tags to a page, and then post a link to that page on FB
# 20:40 tantek not theoretical at all - immediately concrete
# 20:40 ben_thatmustbeme aaronpk, i think the only argument is just that, when you don't have links to easily get through all your pages
# 20:41 aaronpk ben_thatmustbeme: yeah but even if *you* don't have links to all your pages, someone else may have linked to a page, and google could have found it that way
# 20:42 aaronpk because you submit it to google and you get absolutely no feedback out
# 20:42 ben_thatmustbeme if there is a page on my site not linked anywhere, on my site or others, then it would not be able to be found
# 20:43 aaronpk I am more curious to see whether indie indexers would find a sitemap useful
# 20:43 tantek the classic (pre-google) definition of sitemap is as listed on /sitemap - that is, a *visual* list of links to your pages
# 20:43 aaronpk or whether they would just expect to find all the posts by crawling home page feeds and searching for rel=prev and rel=next links
# 20:43 tantek which follows the invisible sidefile antipattern - a specific form of DRY violation
# 20:43 ben_thatmustbeme i think the lesson here is to just make it so you can go back in your feed history fully
# 20:44 tantek and if each archive year/month has a rel=prev link to the previous one then you can crawl the entire thing
# 20:45 ben_thatmustbeme though i need a way to go through those as i believe they are limited to only so many
# 20:48 tantek "… the email credentials of several staff members were compromised …"
# 20:48 Loqi tantek meant to say: TFA for all email accounts people. Seriously.
# 20:48 aaronpk "SMTP email is a failed experiment causing untold damage to millions of users around the world."
# 20:49 ben_thatmustbeme tantek, i think i only display like the last 25 or so, so if i start getting too many posts it wouldn't show some
# 20:50 tantek aaronpk - quite successful for a failed experiment
# 20:57 tantek aaronpk, indie indexers should be able to process a *visible* sitemap page if they have such needs, instead of some decrepit XML format in a likely out-of-date sidefile
# 20:58 aaronpk oh yeah I wasn't even considering the xml sidefile
# 20:59 aaronpk I was just considering the value of a sitemap (single page listing all URLs on the site) vs being able to crawl /archives or just paging through feeds
# 20:59 tantek and feel free to document the links to the XML sidefile flavor(s) too
# 20:59 tantek aaronpk - perhaps start with real world examples of good sitemaps on existing sites
nloadholtes joined the channel
# 21:04 aaronpk way faster to find product pages using that than browsing the js and image-heavy apple.com
# 21:06 aaronpk lol oops my sitemap.xml file is empty and i don't know why. probably time to delete it
# 21:08 aaronpk is there a requirement or implication that a sitemap is a single URL? e.g. is not a paged list?
# 21:08 tantek and you just provided a good real world example of why invisible XML sidefiles are an antipattern
# 21:08 tantek aaronpk - currently documented sitemap example(s) show a single URL
# 21:09 tantek you're free to research other examples to see if you find a pattern or exceptions
# 21:11 aaronpk searching the text of my most recent post shows it on several of my pages on google. this is despite my sitemap being empty.
petermolnar joined the channel
snarfed joined the channel
lukebrooker joined the channel
KartikPrabhu joined the channel
# 21:47 Loqi The sidefile-antipattern is a violation of the DRY principle by the use of secondary files (typically in some one-off XML format) to provide information that is a duplicate of information available in primary files on a website (in HTML), and is an antipattern due to typical DRY violation problems such as out-of-date, corrupted, or outright false data https://indiewebcamp.com/sidefile
gRegor` joined the channel
minsky joined the channel
# 22:11 tantek pops the stack back to 2015-01-01 personal site launch commitments
# 22:12 tantek checks aaronpk and ben_thatmustbeme and GWG sites for 2015-01-01 posts
snarfed joined the channel
# 22:12 tantek welcome KartikPrabhu - did you see the logs from yesterday?
# 22:14 tantek ben_thatmustbeme: except in this case what you're calling the inverse is a behavior that preceded it
# 22:17 tantek you're right, inverse is imprecise but I decided not to argue that ;)
# 22:19 tantek KartikPrabhu: Google itself is too dependent on Blogger - all their corporate blogs (search, APIs, Chrome, etc.) are on it
# 22:19 tantek KartikPrabhu: that is funny - and no one pointed it out at the time
# 22:20 aaronpk ben_thatmustbeme: uhoh I suspect you're going to beat me to checkins
# 22:20 tantek KartikPrabhu: but nevermind worrying about what Google may or may not kill in 2015 - let's focus on what YOU will ship on your site on 2015-01-01
# 22:21 KartikPrabhu can't make indieweb newyears resolution. have to make a friends portfolio site by then
# 22:21 tantek it's not a new year's resolution - which implies something you will do *in* the new year.
# 22:21 tantek rather we're talking about what you're commiting to ship *by* the new year
# 22:22 KartikPrabhu tantek: yes which is why I cant commit. Have to roll out friends site by Jan 01 :P
# 22:22 ben_thatmustbeme I've also started saying the hell with it and syndicating far more of my posts/notes to FB
# 22:23 tantek KartikPrabhu: still - that's a pretty awesome ship goal!
# 22:23 tantek helping someone else ship an entire personal site (and presumably start using it as of the 1st) is quite admirable/honorable
# 22:25 tantek KartikPrabhu: it's ok either way - definitely post that as your 2015-01-01 indieweb ship commitment
# 22:30 kylewm ben_thatmustbeme++ legitimately laughed out loud at that
tilgovi joined the channel
pwcc, verdi__, PMurphs, lupinedev1 and KartikPrabhu1 joined the channel
prtksxna joined the channel
mdik joined the channel
sammachin joined the channel
# 23:15 tantek somewhere I remember docs about a minimum width of 200px or 300px
# 23:17 aaronpk actually proper thing to do would be run all the IRC images through that proxy which resizes them so there aren't ay 400px images as avatars
# 23:18 tantek true! https proxy and image size reduction at the same time - would help with reducing page load size
j12t joined the channel
KartikPrabhu joined the channel
# 23:56 GWG pwcc: Did you look at home page webmentions?
# 23:58 pwcc GWG: I noticed they were mentioned in the readme using filters & a default post ID. Thoguth is to make that native.