#dev 2022-01-11

2022-01-11 UTC
gnoth[d] joined the channel
#
@grafovolaverunt New post on my site! This time about how to add webmentions to your webpage https://grafovolaverunt.xyz/blog/webmentions/ (twitter.com/_/status/1480701955148185600)
darkkirb, KartikPrabhu, Seirdy, [fluffy], jessealama and MarkJR84[d] joined the channel
#
Ruxton [tantek]: re accessing netflix watch history, it's buried in the app, traktflix grabs it somehow clientside - https://github.com/tegon/traktflix
#
Loqi [tegon] traktflix: Trakt.tv + Netflix = :heart:
#
Ruxton A good rundown on the whole faker.js thing.. Marak pretending this is like Aaron Swartz is an insult to Swartz' morals.
#
Ruxton Ruby community handled it much better when why the lucky stiff disbanded and deleted everything, NPM community has failed hard both times this has happened.
jessealama, [Zegnat] and angelo joined the channel
#
[James_Van_Dyne] capjamesg: Can you add assign me to issue #11 on indieweb-utils. I can’t seem to assign it to myself (maybe don’t have permissions?)
jessealama and tetov-irc joined the channel
#
GWG I know very little about SEO... anyone have an intelligent comment about https://github.com/dshanske/syndication-links/issues/149
#
Loqi [kuz-z-zma] #149 Hidden links on Front page/Archive/Search results are bad for seo
#
capjamesg[d] Maybe a good middle ground would be to use <link> tags if you can?
#
capjamesg[d] I don't think someone's going to see serious SEO issues because of those links being hidden. But, I can see quite a few people (like the author) that would prefer not to have hidden links.
#
[KevinMarks] I'd ask Danny Sullivan for advice directly
#
capjamesg[d] (Note: this is my opinion, not necessarily the truth)
#
GWG Link tags don't have a class. These links are there for microformats parsing
#
capjamesg[d] Ah, of course 🤦
#
GWG Probably the best compromise is to let them be turned off entirely. But I'm not sure if there is a middle ground to satisfy SEO people and still get the benefits of being able to parse an h-feed even if you don't want to see them.
#
GWG So, search engines don't like hidden links...but giant blobs of hidden data in JSON format are fine. Perfect sense.
[grantcodes] joined the channel
#
[grantcodes] Hidden links were historically used to try and trick the system. Could you use a `data` element instead?
jessealama joined the channel
#
GWG [grantcodes]: How would you mark up a u-syndication property with a data element?
#
GWG I probably could, just wondering if that is valid HTML
#
capjamesg[d] GWG https://developers.google.com/search/docs/advanced/guidelines/hidden-text-links
#
capjamesg[d] I don't know how to interpret Google's guidelines for a syndication link based on the information above.
#
capjamesg[d] The GitHub issue mentions that hidden links in "any way" are frowned upon but that is not true.
#
capjamesg[d] For instance, most skip links are hidden and they are considered a best practice for accessibility (and Google acknowledges accessibility in hidden links above).
#
Zegnat GWG: if you never want to display them, use <data> elements instead of <a> elements
#
GWG Zegnat: Is that valid HTML, if so that's a solution
#
Zegnat It is valid HTML, and part of the mf2 parsing spec that I believe every parser has implemented
#
Zegnat <data value="https://example.com/" class="u-url"></data>
#
Zegnat Instead of <a href="https://example.com/" class="u-url"></a>
#
Zegnat Should parse the exact same, but will not trip up accessibility tooling, and I would bet search engines ignore links in value attributes.
#
GWG I'll experiment
#
[KevinMarks] if they're there for the rel=me, then you can use <link>
#
GWG They aren't as this is the h-feed
#
[KevinMarks] ah, sorry, missed the context.
#
@capjamesg Request for tool: A #Indie poll tool that lets you aggregate Yes/No or multi-variant responses to a question via webmention. An API then returns the percentage of votes for each option so it can be embedded / rendered on a site. (https://jamesg.blog/notes/2022-01-11-380/) (twitter.com/_/status/1480897346791784448)
sayanarijit[d] and jessealama joined the channel
#
[snarfed] ^ capjamesg do we have any indieweb polls like that in the wild?
#
[snarfed] manual until it hurts, I'd be curious to see one done manually first
#
Zegnat I guess you could request /reacji ... or go the old school Twitter way: reblogs vs likes
jessealama joined the channel
#
[KevinMarks] The vote links way was to provide different urls for the different choices and have people link those. I suppose that you could do the same page with a fragment too.
jessealama joined the channel
#
capjamesg[d] [James_Van_Dyne] I assigned you to the ticket and made you a collaborator on the repo.
#
capjamesg[d] Zegnat that's true re: reacjis.
#
capjamesg[d] Good idea!
zack1 joined the channel
#
capjamesg[d] It's not an urgent need for me right now but if I could do it I would.
#
[James_Van_Dyne] Thanks :)
KartikPrabhu, aspenmayer[d], angelo, [manton] and Seirdy joined the channel
#
capjamesg[d] Has anyone read this?
#
capjamesg[d] https://datatracker.ietf.org/doc/draft-nottingham-avoiding-internet-centralization/
#
capjamesg[d] “ Standards efforts should focus on creating new specifications for
#
capjamesg[d] functions that are currently only satisfied by proprietary,
#
capjamesg[d] centralized protocols. For example, if social networking is thought
#
capjamesg[d] to be a centralized function, this might mean creating specifications
#
capjamesg[d] that enable decentralized social networking, perhaps using some or
#
capjamesg[d] all of the techniques described in Section 5.”
#
capjamesg[d] Sorry about the bad spacing. I’m on mobile and the copy didn’t come out well there.
jessealama and tetov-irc joined the channel