#dev 2018-12-31

2018-12-31 UTC
#
[tantek] [kevinmarks] that's not really the focus of h-feed though, so it's a weird way to define it. It's more a place to put information about the set of entries, which only indirectly might have info that applies to all of the entries.
#
[kevinmarks] True.
[schmarty] joined the channel
#
[schmarty] i'm not super pleased with h-feed as a catch-all for collections of h-entry - unless they are meant to be followed like a feed.
#
[tantek] schmarty I think they make sense as a return format for paginating through a feed
#
[tantek] e.g. you follow someone's h-feed (typically on their home page), and want to "view previous posts" (in your reader)
#
[tantek] your reader should be smart enough to look for a rel=prev link to the previous "page" of posts, which would also be an h-feed
#
[tantek] but it might be something like "last week's posts"
#
[schmarty] tantek: agreed! and that is one way i publish h-feeds. i also use h-feed on tag pages.
#
[tantek] and then that page itself may have a rel=previous link, etc.
#
[tantek] I don't know any readers that do this, but they could do rel=previous discovery and provide this feature
#
[schmarty] i am less into the idea of using h-feed to describe an archive collection that isn't expected to update over time.
#
[schmarty] but i am probably splitting hairs
#
[tantek] I understand, "feed" implies some sort of update semantic
#
[tantek] we could use a new "collection" object instead that looked just like an h-feed but without any such update semantic, but would it really be worth dealing with a separate container object just because of that distinction?
#
[tantek] that's the tradeoff
#
[schmarty] tantek: yep. it's quite a tradeoff! makes me realize that if we changed the name from "feed" to "collection" in all cases, it would still make sense...
#
[schmarty] a reader could periodically check a collection for new entries, and look for rel=next/prev links for adjacent pages of the collection.
#
[tantek] except for the very common case of wanting to subscribe / follow
#
[tantek] people follow feeds, not collections
#
[tantek] abstracting the name into "collection" then makes the common case less obvious, which is also bad
#
[schmarty] i don't think it's intuitive that people follow feeds and not collections.
#
[tantek] really? I'd say the UX of past 15 years of readers pretty much makes that case
#
[schmarty] in the sense that i think the word "feed" is jargon.
#
[tantek] it's borderline. I think it used to be jargon, but got so well known / written about (e.g. press articles using the term in their headlines) that it became more of a common noun
#
[schmarty] that's fair. "news feed" "twitter feed" and more are pretty widely used.
#
[tantek] lots of citations on https://indieweb.org/algorithmic_feed
#
[schmarty] what is a feed?
#
Loqi A feed is a dynamic set of posts, typically listed in reverse-chronological order, often only the most recent n (like 10) of them https://indieweb.org/feed
#
[schmarty] that def fits my thinking that a feed is a special case of collection :}
#
[tantek] agreed. I suppose archive pages could just have a bunch of h-entry elements and a rel=previous link without an explicit h-feed
#
[tantek] they don't need the h-feed because they're already implicitly contained in the h-feed that linked to them (transitively)
#
[tantek] i.e. if a consuming reader starting following rel=previous links, there is no reason for it to change the name / author etc. of the feed it is paging backwards in time through
#
[schmarty] oh heh, that is interesting.
#
[tantek] point being, there is no necessary consuming use-case for an explicit "h-feed" on archive pages
#
[tantek] so might as well leave it out
#
jacky interesting
#
[tantek] and it turns out (just checked) I don't have explicit "h-feed" in my archive pages
#
[schmarty] ahaaaaaa. tantek, i found the crawler i was thinking of that starts on your main feed and follows rel=next/prev: http://mirror.rhythmandpoetry.net/indieweb-bestnine/
#
GWG I need to work on the id problem this week
#
[tantek] GWG were you able to try out the phpmf2 with the ID support to see if it gave you the results you needed?
#
[tantek] schmarty++
#
Loqi schmarty has 13 karma in this channel over the last year (58 in all channels)
#
[tantek] what is rel-previous
#
Loqi It looks like we don't have a page for "rel-previous" yet. Would you like to create it? (Or just say "rel-previous is ____", a sentence describing the term)
#
GWG tantek, see last statement
#
[tantek] rel-previous is the use of the HTML rel attribute on a hyperlink to page previous to the current page with the value of "previous" to provide a way for consuming code to discover that link and navigate to it to show content previous to the current page; like an indieweb archive page that links to the previous archive page in a time series.
#
Loqi ok
snarfed joined the channel
#
[tantek] rel-previous << Consuming code implementation: http://mirror.rhythmandpoetry.net/indieweb-bestnine/ - Indieweb BestNine - a site for crawling your site and finding your best nine photos according to the number of likes they have received
#
Loqi ok, I added "Consuming code implementation: http://mirror.rhythmandpoetry.net/indieweb-bestnine/ - Indieweb BestNine - a site for crawling your site and finding your best nine photos according to the number of likes they have received" to a brand new "See Also" section of /rel-previous https://indieweb.org/wiki/index.php?diff=55126&oldid=55124
#
[tantek] rel-previous << archive navigation
#
Loqi ok, I added "[[archive navigation]]" to the "See Also" section of /rel-previous https://indieweb.org/wiki/index.php?diff=55127&oldid=55126
#
[tantek] archive navigation << rel-previous
#
Loqi ok, I added "[[rel-previous]]" to the "See Also" section of /archive_navigation https://indieweb.org/wiki/index.php?diff=55128&oldid=53469
#
[tantek] schmarty ^^^ hopefully it will be easier to find next time
#
[schmarty] tantek++ nice
#
Loqi tantek has 27 karma in this channel over the last year (83 in all channels)
snarfed and ben_thatmustbeme joined the channel
#
@tdmsilvino Atualizei meu blog https://silvino.net. Usando Hugo, Cloudflare, Netlify. Habilitado para a Indieweb. Agora só falta habilitar webmentions! (twitter.com/_/status/1079583948622184448)
snarfed and kisik21 joined the channel
#
GWG !tell tantek I was able to use the ID property to do feed discovery on your site. Going to take some tweaks to make it work properly though
#
Loqi Ok, I'll tell them that when I see them next
[tantek] joined the channel
#
[tantek] what kind of tweaks?
#
Loqi [tantek]: GWG left you a message 5 minutes ago: I was able to use the ID property to do feed discovery on your site. Going to take some tweaks to make it work properly though
#
[tantek] glad you got it to work!
#
[tantek] !tell GWG can you comment with at least that and perhaps details of which implementation (phpmf2) you used? https://github.com/microformats/microformats2-parsing/issues/44
#
Loqi Ok, I'll tell them that when I see them next
#
Loqi [sknebel] #44 parse HTML id= property
#
[tantek] Thanks GWG, mf2 spec now updated accordingly!
[nick] joined the channel
#
[nick] why aren't webmention receive endpoints REST?
#
[nick] I feel like the endpoint should hit with `Content-Type: application/json` and `{"source":"url","target":"aurl"}`
#
KartikPrabhu why JSON? a simple endpoint with source and target allows for simple forms to submit webmntions
#
[nick] i guess mostly for convience... for me 🙂
#
[nick] all my validation is done through json serializers...
#
[nick] but i come from a world where everything is REST api based
#
[nick] I don't see anything in the docs about what content type should be expected..
[tantek] joined the channel
#
[tantek] why use the extra layer of JSON when everything works just fine in the URL?
#
[tantek] FWIW pingback made the same "extra layer" design error, by wrapping everything up in XML-RPC
#
[nick] i never used pingback so I don't know any of it's mistakes or design decisions...
#
[tantek] webmention simplified that by eliminating the extra layer is the point
#
[nick] i see webmention being able to, update, create delete a web mention object
#
[nick] which to me says makes sense in that being a `POST`, `PUT` and `DELETE` http request
[asuh] joined the channel
#
[nick] which translates to REST
#
[tantek] except it's across ownership boundaries, so no, the HTTP verbs don't make sense
#
[nick] and the "standard" for REST is json..
#
[nick] can you elaborate?
#
[tantek] REST predated JSON, so no that's not true either
#
[nick] (sorry this is all new to me...)
#
[tantek] when send a webmention, you only control the source. what happens at the target is up to the owner of that site
#
[tantek] it goes across ownership boundaries
#
[tantek] from one site to another
metbril joined the channel
#
[nick] so if you send a webmention to delete a post
#
[nick] say a reply
#
[tantek] as the author of a comment, don't presume to have absolute CRUD control over how someone else's site handes it
#
[nick] it's up to me to respect your wish to delete that?
#
[tantek] that's way it doesn't map directly to HTTP verbs which assume the same "owner" in "full control"
#
metbril For the record. I created a basic IndieAuth plugin for Grav CMS. This will make it easier to use a Micropub server in the near future.
#
metbril More info is added to the Grav wiki
#
[tantek] webmention delete is a good example - the receiver of the delete could decide to just show the comment with strikethrough
#
metbril what is Grav
#
Loqi Grav is a flat-file CMS built on PHP, with Twig templating, and YAML + Markdown for storing articles (YAML for metadata, Markdown for the content) https://indieweb.org/Grav
#
[tantek] awesome metbril! can you provide a post about it?
#
metbril A post on my own site, that is?
#
[tantek] nick, think of all the social considerations about commenting, moderating, muting, blocking, reporting for spam etc.
#
metbril working on that
#
[nick] i guess i still see it through my "rose tinted" glasses.
#
[nick] i send you a `DELETE` request a reply
#
[nick] you handleit how ever you want.
#
[tantek] "handle it however you want" is not how HTTP DELETE is defined
#
[tantek] so no, you can't use it
#
[tantek] in practice most HTTP verbs are overly strictly defined (about what they do to the resource) to be useful to actually build user-centric (especially cross-site) protocols for
#
[tantek] REST in general assumes a very simple ownership model between the user agent issuing the REST requests and the server responding to them.
#
[tantek] metbril, https://indieweb.org/Grav#IndieAuth looks good!
#
[nick] [tantek] so if I'm going to request a delete how does that work in webmention land?
#
[tantek] just as the spec says https://www.w3.org/TR/webmention/#sending-webmentions-for-deleted-posts
#
[nick] oh cool I didn't see that site before
#
[nick] i've been digging around the wiki
#
[nick] for indieweb
#
[nick] it just mentions that you can delete... but I didn't see anything about specifics..
#
[tantek] oh yes definitely check out the spec!
#
[nick] the spec is super helpful
#
[tantek] metbril looks like you got it done before midnight PST!
#
[nick] it just feels weird to send `links` as opposed to sending "content"
#
[nick] so it's a new way of thinking for me..
#
[tantek] reduces the amount of upfront "trust" or authentication etc. that you need
#
[tantek] nick, people tried sending "content" with such a protocol at first - that was Trackback - and it instantly died under spam
#
[tantek] what was Trackback
#
Loqi TrackBack was a protocol for web sites to notify other web sites when they've posted a link to them (respectively) https://indieweb.org/Trackback
[metbril] joined the channel
#
[metbril] Tantek, my TZ is CET.
#
[nick] ok so with just sending content... there's no way to verify that I actually linked to your post...
#
[tantek] metbril, still, we're counting everyone PST for the indieweb challenge to keep the deadlines the same
#
[nick] so with webmention i pull the content and verify the link is there.. and I inspect to find certain elements in the html
#
[nick] and pull those out to create content on my page..
#
[tantek] nick, exactly. "no way to verify" was TrackBack (with a bunch of unnecessarily complex XML/RDF syntax tossed in because that was someone's flavor of the month)
#
[nick] but my flavor is the best flavor 😉
#
[tantek] metbril++ nice work
#
Loqi metbril has 1 karma in this channel over the last year (4 in all channels)
leg joined the channel
#
[nick] soo what are the fears of webmention...
#
[nick] (that might be in the spec haven't gotten that far)
[pfefferle] and barpthewire joined the channel
#
[nick] [tantek] so as far as an "sending mechanism" on my page. If I'm going to write a post and lets say I have a bunch of links in there... Should my publishing mechanism filter through all the links that are in the post and blindly send webmentions?
#
[nick] in the hopes that the linked domains have a webmention receiver?
#
[nick] oh I see I look for a webmention link on the domain first..
#
[tantek] in order to send webmentions you have to do discovery to find their endpoints
#
[tantek] their endpoint is the webmention receiver. there's no "blindly send"
#
[tantek] it's why it's designed like that
#
[nick] yeah i just read that the receiver could be on a different domain...
#
[nick] so I'd iterate through my links. review the source of those links. look for a `webmention` link if found submit a request to that link...
#
[nick] I see why async is so important for this...
#
[nick] ok time for bed. thanks for your time
#
[nick] appriciate it
#
[tantek] no problem! more spec reading in the morning 🙂
#
[nick] wife kid and i are leaving for CA tomorrow... so I'll be reading on the plane
#
[nick] also trying to figure out a way to model all the differnet post types... so i can handle those... well make something thats expandable
swentel, jgmac1106, [jgmac1106], [xavierroy] and [Vincent] joined the channel
#
[Vincent] Morning IndieWeb
[voss] and jjuran joined the channel
#
jgmac1106 morning
#
Loqi *yawn* ... /stretch/
#
jgmac1106 I had no idea that Medium had native microformats,
barpthewire and jgmac1106 joined the channel
#
@fuzzylogicx ↩️ I started this earlier this year on my Perch-based site and got as far as implementing indieauth and webmention in both directions. Still got a list of to-dos with micropub, bridgy, POSSE and PESOS. If you ever want to chat to someone going through same process, just shout! (twitter.com/_/status/1079721427283202050)
#
Zegnat petermolnar++ for true OPML on personal website example
#
Loqi petermolnar has 4 karma in this channel over the last year (16 in all channels)
[davidmead] and [jgmac1106] joined the channel
#
GWG I have been trying to use the parsefromID function in php-mf2, but it seems to only parse the children. How do I include the element the id is on as well?
#
Loqi GWG: [tantek] left you a message 7 hours, 25 minutes ago: can you comment with at least that and perhaps details of which implementation (phpmf2) you used? https://github.com/microformats/microformats2-parsing/issues/44
#
GWG Handled that already
#
Zegnat parseFromID is actually a pretty old method, not sure what it does or how it is supposed to be used
#
GWG Also trying to figure out if I should use it if the URL has a fragment
#
GWG Should I assume a fragment is an attempt to identify part of a page?
#
Zegnat I think XRay does that, but there are some possible problems with it.
#
Zegnat If you only parse a subset of the page, you may not be able to do certain things like the author algorithm. You no longer know if there is an h-card elsewhere on the page you could use. Or if the h-entry was nested in an h-feed with author information.
#
GWG I think I need an algorithm for this
#
GWG Or at least a test suite.
#
GWG Where is Microformats.rocks when you need it?
#
GWG I need a bunch of different scenarios to throw at my post processor
#
Zegnat This seems to be like a pretty new use-case, so it wouldn’t have helped, haha
#
Zegnat Here you go, one test case of fragment permalinks: https://grapefruit.zegnat.net/2018/04.html#dt201804141608Z
#
Loqi Sebastiaan and I were on a philosophy bender, apparently. One of the concepts to come out of it is the head cache. When you have filed something in the back of your mind for future use. The problem, just like with a computer cache, is that ...
#
GWG I will give it a shot later. I have work most of the day
#
Zegnat No hurries. But that is one where the authorship algorithm should have you check the parent h-feed. Not sure how many other live examples of that use are out there
#
GWG tantek doesn't have a parent h-feed. His site is always a challenge.
#
Zegnat Not sure where tantek’s posts fall on the algorithm
#
sknebel petermolnar: did you get your OPML imported into a microsub server with ek? if yes, can you make a note somewhere about it?
#
petermolnar sknebel: yes, I did; it worked fine, however, I'm yet to actually figure out how to use microsub as my rss reader replacement, because despite having all the sources, nothing is publising in the topics.
#
petermolnar `ek connect [ your website url ]` (follow link it displays); `ek import opml [opml file path]` - that's it
#
petermolnar to clean it up in bash `for c in $(ek channels | awk '{print $1}'); do for url in $(ek follow $c); do ek unfollow $c $url; done; ek channels -delete $c; done`
[jgmac1106], [smerrill] and [kevinmarks] joined the channel
#
pstuifzand I wrote about how you can use `ek` to import and export OPML: https://p83.nl/posts/957
#
Loqi pstuifzand: petermolnar left you a message 2 days, 22 hours ago: I got ek working with aperture, it's completely happy now; didn't have time to poke at eksterd yet
#
Loqi [Peter Stuifzand] Import OPML files into Microsub servers
#
@mjmoriarity I think I've got basic support for receiving Webmentions working on http://mattmoriarity.com. At the very least, replies from http://Micro.blog should show up like comments on posts. This was a lot more work than I expected it to be. (twitter.com/_/status/1079763136251596800)
#
sknebel aaronpk: btw, did you have new(-er) thoughts on the entire plaintext thing while working on Aperture/monocle more?
#
sknebel (whitespace, image handling, ...)
#
aaronpk Oh huh, I haven't noticed, so I guess that means whatever it's currently doing is good
#
@mattl ↩️ Yep. http://IndieWeb.org. I have a blog that’s IndieWeb ready and my domain is linked from my Twitter bio so I can use indieauth (twitter.com/_/status/1079772933189451776)
#
Zegnat I should do a php-mf2 branch with the new innertext works and see if those are any good. As it is closer to a describable algorithm
#
Loqi I agree
[tantek] joined the channel
#
[tantek] Zegnat, Petermolnar, if you want a guestbook on your site, create a page /guestbook that accepts webmentions and let people comment away. presto you have a guestbook
#
sknebel Zegnat: yes, I stumbled over my code for that too today too
[schmarty] joined the channel
#
[schmarty] pstuifzand++ for ek OPML import and export! i am delighted to be able to back up my aperture feeds so easily!
#
Loqi pstuifzand has 7 karma in this channel over the last year (9 in all channels)
#
[schmarty] !tell pstuifzand: hit an issue with ek OPML export from aperture if any channel is set to "Show Unread Indicator" rather than "Show Unread Count": 2018/12/31 11:32:28 main.go:395: An error occurred: json: cannot unmarshal bool into Go struct field Channel.unread of type int
#
Loqi Ok, I'll tell them that when I see them next
#
@jgmac1106 ↩️ Can we just use Aperture/Indigenous as the client or do we only need Indigenous? Haven't dug into code to see if microsub native. Wouldn't have skills for PR but interest is growing for a simpler #IndieWeb version of WordPress(ish) that is… https://quickthoughts.jgregorymcverry.com/2018/12/31/can-we-just-use-apertureindigenous-as-the (twitter.com/_/status/1079778723480723456)
#
pstuifzand [schmarty], thanks for the report. The microsub spec allows for both a boolean or a number for the unread count.
#
Loqi pstuifzand: [schmarty] left you a message 8 minutes ago: hit an issue with ek OPML export from aperture if any channel is set to "Show Unread Indicator" rather than "Show Unread Count": 2018/12/31 11:32:28 main.go:395: An error occurred: json: cannot unmarshal bool into Go struct field Channel.unread of type int
#
pstuifzand The Go json decoder, can't handle both an bool and int for the same struct member. It's something I need to look into.
#
[schmarty] pstuifzand: static type systems strike again! thanks for the excellent utility!
[kevinmarks] joined the channel
#
[tantek] GWG you're right, parsefromID should include the element it is parsing! definitely file an issue on that for phpmf2
#
[jgmac1106] @schmarty my reader grabbed your last chunk of posts with your update,fun history of Marty saying Goodbye to Baltimore
#
[schmarty] [jgmac1106] i kind of wish i had captured more, but it felt weird to make public posts about it.
#
[jgmac1106] Private posts.. Or post you don't syndicate POSSE are good
#
[jgmac1106] I thought you did nice job
#
aaronpk KartikPrabhu: what happened to the mf parser here? https://kartikprabhu.com/connection/mfparser (found a link in a github thread)
#
aaronpk oh weird it's there but it gets hidden until i resize the window
#
aaronpk and sometimes typing in the box hides it again
#
petermolnar [tantek]: re guestbook: not the same. Guestbook entries mainly came from lurkers, not content creators.
#
[jgmac1106] Can't you then have a guest book that takes webmentions and native comments
#
[tantek] petermolnar, accept tweet guestbook entries via Bridgy Backfeed 😛
#
petermolnar the only solution would be self-webmentions, which is, in essence, hosting native comments, which I don't want this way, because spam. As for twitter: same problem, not for the anon lurkers.
#
Zegnat Does bridgy send homepage mentions? Maybe I should enable it if it means all @martijnvdven tweets go to my webmention endpoint
#
[tantek] Twitter is one step better than native comments in terms of barrier to entry
#
[tantek] Zegnat yes Bridgy sends homepage mentions
#
[tantek] I have a lot that are pending processing 😂
#
aaronpk "a lot" <-- that's an understatement
#
Zegnat Guess that would be a reason to turn bridgy on then. I have never bothered with bridgy
#
KartikPrabhu aaronpk: which browser?
#
aaronpk CHrome mac
#
KartikPrabhu yeah I have that behaviour on Chrome Linux too. Chrome is bad at girds
#
KartikPrabhu good thing eveyone is switching to Blink right?!
#
KartikPrabhu earlier I had a fix for it since Chrome dd not support "display: contents" but now it does and the grid is still broken! I am really not inclined to fix things for Chrome anymore
#
petermolnar KartikPrabhu: tried it in Dillo, just for curiosity, "CSRF verification failed. Request aborted." after posting content
#
KartikPrabhu petermolnar: I think you have blocked the CSRF cookie then
#
petermolnar I don't think Dillo knows about CSRF
#
KartikPrabhu if it is a browser it would know about cookies
#
KartikPrabhu what's Dillo anyway?
[smerrill] joined the channel
#
[smerrill] https://www.dillo.org/
#
KartikPrabhu aah dillo blocks all cookies by defualt
#
[kevinmarks] trying to make mention.tech json compatible with webmention.io - is wm-property Post Type Discovery, or is it just the first property of a given type?
#
aaronpk It's meant to tell you the property name that has a link to your page
#
aaronpk Someone asked me to add that at some point
#
[kevinmarks] ah, OK. I was looking at https://beesbuzz.biz/blog/3743-More-fun-with-Webmentions and it is very dependedent on it
#
[kevinmarks] so I should look through the properties and put the first one that has a url of the target into wm-property?
#
[kevinmarks] http://mention.tech/listmentions?target=https%3A%2F%2Ftumblelog.xyz&json=on
eli_oat and [nick] joined the channel
#
[nick] ok I think I got my receiver working. Changed my view to accept form content instead of json. i've had to take inspriation from a couple different projects that kinda didn't do everything or everything the way I wanted...
#
[nick] now time for a few more tests before I deploy..
#
[nick] next to implement some discovery and sending
eli_oat and [smerrill] joined the channel
#
[smerrill] style question: do you folks generally prefer a permalink anchor to come before or after the content?
eli_oat joined the channel
#
[kevinmarks] webmentions now shown on https://tumblelog.xyz/ via mention.tech and a mutated version of https://beesbuzz.biz/blog/3743-More-fun-with-Webmentions
#
Loqi Tumble log like it’s 2005
gRegorLove joined the channel
#
Zegnat [smerrill], for long content I like to see the publishing date upfront. So I know how current the writing is. And I prefer the permalink to be on the timestamp.
#
Zegnat So it's not so much that I want the permalink up top, but it would still end up there
#
[smerrill] @zegnat thanks. that’s kind of what i was leaning toward. Trying to twiddle this list of tweets into something marginally legible: https://skippy.net/files/test.html
#
GWG Zegnat: Tried your URL. Not if I use parseFromID
#
Zegnat GWG, it is an example that I dont think anyone parses right.
#
GWG It is the same problem as before. parseFromID doesn't add the encompassing element.
#
Zegnat Also note that parseFromID is not an addition from this latest spec change, has been in the parser a while now
#
Zegnat And I am not sure what the original goal of that method was
#
GWG Yes, but now with the new change, it becomes a feature request.
#
Zegnat How do you think is is more useful than parsing the whole page and just look for the microformats object who's I'd property matches the fragment?
#
Zegnat By doing that you'll still have information like the parent microformat available, which you would otherwise lose
#
GWG Zegnat, isn't that a function that might be helpful inside the php-mf2 library? Or you think it is should be separate?
#
Zegnat I think it should be separate. But I may just not be seeing the use case for parsing from a specific ID attribute down.
#
GWG I guess I need to build a function to do fragments then.
#
@vincentlistens @adactio Apologies, a typo in my code caused Telegraph to loop and send 5 webmentions to your site, before I could stop it. (twitter.com/_/status/1079838093816070145)
#
Zegnat There are also two different types of "fromid" for a parser. Because there are the IDs in the DOM before parsing, and now also IDs for the microformats objects...
#
Zegnat So I guess we need to see the use cases and see which make sense for the parser to do or for consumers to do
[Vincent] joined the channel
#
[Vincent] 😬 That was hair raising. I had a typo in my code, causing my function to loop repeatedly and send tons of the same webmentions through to Telegraph. Poor Jeremy K got 5 webmentions in a few seconds, before I snagged it and stopped the code.
#
[Vincent] I might consider adding some sort of queue thing, before I post to telegraph in case it ever happens again 😬
snarfed joined the channel
#
Zegnat To be fair, good test for receivers
[schmarty], [kevinmarks] and snarfed joined the channel
#
@tdmsilvino https://silvino.net/2018/12/31/webmentions-e-netlify-cms/ (twitter.com/_/status/1079850831145521153)
#
@tdmsilvino Webmentions e Netlify CMS https://silvino.net/2018/12/31/webmentions-e-netlify-cms/ (twitter.com/_/status/1079851060225810432)
tbbrown, [schmarty] and jjuran joined the channel
#
[schmarty] !tell tantek: looks like your 🕸 links are out of date. possibly you added them very early before I had the URL normalization and hashing down pat? either way: your site is not currently in the 🕸 database. you can log in again to get fresh links.
#
Loqi Ok, I'll tell them that when I see them next