#microformats 2021-12-29

2021-12-29 UTC
gRegor, wackycity[d], jeremycherfas and [snarfed] joined the channel
# 04:20 
@SamRose ↩️ In the beginning of all of these centralized services: they offered RSS feeds, used microformats (so their pages could be crawled and data extracted), they used OpenID, they offered freely available API's. Until bigger capital got involved ... (twitter.com/_/status/1476044514255155203)
jessealama[d] joined the channel
# 05:00 
@alamajesse ↩️ Thanks! I signed up for Discord now and see the microformats channel. (twitter.com/_/status/1476055088347111428)
# 05:04 
jessealama[d] very impressive Loqi! I am indeed in 🤣
# 05:07 
jessealama[d] not sure if this is the right place to ask this kind of thing, but I'll fire away:
# 05:07 
jessealama[d] I've got a Schema.org puzzle I can't get my head around
# 05:09 
jessealama[d] I sell ebooks on my website (entirely handmade) and would like to represent the sales page as both a Book and as a Product. There's also an Offer (capitalized words refer to schema.org entities). The page is arguably *both* a Book and a Product. But it seems I need to pick one or the other. Anyone know how to cut this knot?
# 05:10 
aaronpk You have found the problem with schema.org 😉
# 05:10 
jessealama[d] heh I figured as much
# 05:11 
jessealama[d] are there any alternatives I should know about?
[aciccarello] joined the channel
# 05:11 
[aciccarello] Is that what the "mainEntityOnPage" is supposed to handle? I was looking at that this week and still don't entirely get it.
# 05:12 
[aciccarello] *mainEntityOfPage https://schema.org/mainEntityOfPage
# 05:12 
jessealama[d] hmm, I didn't know about that attribute, taking a look now
# 05:13 
jessealama[d] > Indicates a page (or other CreativeWork) for which this thing is the main entity being described.
# 05:14 
jessealama[d] I wonder if I should address this by just making one section the Book, and another section (with the "Buy" button) the Product
# 05:14 
jessealama[d] then the page would simply have two entities
# 05:14 
jessealama[d] neither contained in the other
# 05:14 
jessealama[d] and then I could use mainEntityOfPage to refer to the Book
# 05:15 
jessealama[d] (seems like it's an editorial choice which is the main entity)
# 05:17 
jessealama[d] thanks! this is already helpful
cygnoir joined the channel
# 05:38 
aaronpk Most of Schema.org is theoretical, and all that really matters is what you want someone (probably google) to get out of it, so your best bet is to look at what you want from the end result and read those docs
# 05:40 
jessealama[d] 👍
Cygnoir[d] joined the channel
# 05:54 
[tantek] There's a theory that Schema-org's complexity/theoretical nature is a giant nerdsnipe/honeypot. Like semantic HTML debates of old of em vs strong, or b vs i vs em vs strong, except 10x worse.
# 05:55 
jessealama[d] * looks up nerdsnipe
# 05:56 
[tantek] 10x is likely underestimating
# 05:58 
jessealama[d] sounds about right to me!
# 06:00 
jessealama[d] I guess the countertheory would be that adding good metadata helps make people more likely to find your stuff. Of course this comes with the caveat that if your site sucks or if you have very little content, it's a waste of time
# 06:02 
aaronpk It only helps people find the stuff is there is something helping them find the stuff
# 06:02 
jessealama[d] I ask myself whether search engines really *need* the extra help that schema.org allows you to give them. Whether natural language techniques combined with hierarchical analysis of HTML documents (and possibly knowledge of who links to whom) is enough
# 06:02 
aaronpk in other words, without a consumer of the data it's useless
# 06:02 
jessealama[d] the theory to refute would be: investing time into nailing your schema.org metadata is totally worthless (even for someone who genuinely has something to offer on their site)
# 06:03 
jessealama[d] yeah
# 06:04 
[tantek] other way around. burden of proof (of relevance/utility/benefit) is always on the proposal to do extra work (in this case to do the work of figuring out schema-org nerdsnipe questions, and then adding it to your site)
# 06:04 
jessealama[d] right
# 06:04 
[tantek] "adding good metadata helps make people more likely to find your stuff." <-- needs proof that there is/are implementations that actually "helps make people more likely to find your stuff" for any particular specific "good metadata".
# 06:05 
[tantek] lacking that proof/evidence, there is no basis for that theory
# 06:05 
[tantek] closed implementations (like big search engines) aren't consistently testable/verifiable (as scientific experiments would require)
# 06:05 
[tantek] thus making it even more difficult to substantiate any such theory
# 06:06 
jessealama[d] agreed. It makes the whole investigation really unsatisfactory
[fluffy] and ur5us_ joined the channel
# 08:42 
PK[d] Schema is gaining popularity in SEO circle, but it’s difficult to create it plus implement it for a non tech person. It needs a team in an organization usually. I find different datapoints it has for any entity like restaurant, business helpful to create content around.
# 08:44 
PK[d] Ive been thinking about a browser plug-in that would highlight visible micro data on a web page. Like the plug-in will just highlight the h card entries on the page.
# 09:45 
jessealama[d] interesting idea!
[tonz], juanchi, KartikPrabhu, [jacky], MarkJR84[d], [jeremycherfas], angelo and ur5us_ joined the channel