#microformats 2015-11-08

2015-11-08 UTC
fuzzyhorns, KartikPrabhu and Calli joined the channel
tantek, KartikPrabhu and fuzzyhorns joined the channel
fuzzyhorns and nitot_ joined the channel
tantek, nitot, KartikPrabhu, ivc, fuzzyhorns, elf-pavlik and nitot_ joined the channel
nitot, fuzzyhorns, Phae, benward, tommorris, iSRAELi, JonathanNeal, bret, twisted`, tantek, Garbee, Yilmaz_Erkan, Erkan_Yilmaz and Calli joined the channel
nitot, gRegorLove, Calli and fuzzyhorns joined the channel
#
Calli
Anyone feel like educating me? I'm kind of curious about MF2 vs HTML microdata (HTMLMD) (By HTMLMD, I mean both itemscope/itemprop and rel w/ article-scoping for author, bookmark)
#
Calli
In sites I look at I see HTMLMD in use, but not MF2 (I know that people here will have MF2 data in your pages, but more generally I don't see too much)
#
Calli
I understand that MF2 vs MF1 was to make parsing rules more general...
#
Calli
...but HTMLMD has general parsing rules
#
Calli
So is there some advantage to MF2 parsing rules over HTMLMD?
#
Calli
Or is there some advantage to MF2's use of class instead of itemscope/itemprop?
#
Calli
Assuming that the MF vocabulary offers things not in other vocabularies, I'm wondering why MF2 isn't a vocabulary that uses itemscope/itemprop?
#
Calli
Thanks for any info you can offer.
Calli joined the channel
#
kevinmarks
Calli: mf2 is easier to publish and has more parsers
#
Calli
Hi Kevin. Thanks for replying. I'll look at the link.
#
kevinmarks
(also the Schema folks seem to have deprecated microdata for JOSN-LD islands now)
#
Calli
This is interesting. I'm really trying to get at the difference between the MF2 and HTML parsing/serialization (rather than the MF2 vs schema vocabulary differences)
#
Calli
What makes MF2 easier to publish than HTMLMD?
#
kevinmarks
look at how less verbose it is in that example
#
Calli
Something to do with CMS limitations?
#
KartikPrabhu
i don't see the need to introduce new HTML attributes for everything
#
Calli
Oh. So implied properties and nesting is a big part of the diff?
#
KartikPrabhu
HTML gives heirarchy and classes tell you the "type" of element so why invent itemprop/itemscope
#
Calli
Are implied properties dependent on vocab or part of general parsing rules?
#
KartikPrabhu
implied props greatly simplify writing mf2
#
KartikPrabhu
general parsing rules
#
Calli
KartikPrabhu: Thx. Will look at the parsing rules again. In terms of "new attributes", as far as I am concerned HTML5 is spec'd, released and the attributes exist, so they don't feel new. ;-)
#
KartikPrabhu
e.g. <article class="h-entry">Title</article> gives same mf2 as <article class="h-entry><span class="p-name">Title</span></article>
#
KartikPrabhu
by "new" I mean for the pubisher to learn
#
KartikPrabhu
I really don't need to learn how itemprops/itemscope work to have "data" in my HTML
#
KartikPrabhu
so mf2 is easier to learn and write
#
kevinmarks
well, microdata is out of html5
#
KartikPrabhu
also that ^
#
Calli
oh yeah - ok - I've been fooled by old specs
#
KartikPrabhu
to me it seems people have given up on microdata for RDFa and now for JSONLD
#
Calli
But: I see content that uses itemprop/itemscope and scoped-rels - and it matches the HTML drafts or whatever
#
KartikPrabhu
people don't really update their sites to match latest HTML5 spec do they?
#
aaronpk
I wonder what is going to come after JSON-LD islands
#
KartikPrabhu
Revenge of the XML
#
KartikPrabhu
errr sorry WebComponents
#
Calli
So comparing microdata vs MF2 (serialization), the main advantage is implied props? Anything else I've missed?
#
KartikPrabhu
Calli: IMO the main difference in ease of publishing rather than what happens in the parsing/serialisation
#
KartikPrabhu
and implied props is a huge part of that
#
aaronpk
also it's just less verbose
#
KartikPrabhu
mainly due to implied props right?
#
Calli
Yeah - I'm calling out "serialization" to distinguish it from "vocabulary".
#
Calli
Don't mean to imply that the main concern is parser developers rather than content creators
#
aaronpk
the other main advantage is mf2 has an active community vs one that has moved on to other methods of embedding data :)
#
Calli
Yeah - the community seems really great!
#
Calli
Still I feel compelled to give you my perspective as an outsider/newcomer: if I look at pages I'm interested in extracting data from, I see microdata some MF1, no MF2, and no RDFa. I see a draft HTML spec that describes parsing rules that match data that I see in pages I'm interested in. I also see JSON+LD.
#
KartikPrabhu
If your goal is to parse specific pages then you'd have to parse whatever they provide
#
Calli
Yeah. Which means I need microdata parsing and scoped-rel parsing.
#
Calli
Be great if I only needed that right?
#
Calli
So I'm trying to understand what are the differences and why they exist?
#
KartikPrabhu
I have never looked at microdata parsing so can't tell if there is more to it than that
#
Calli
Oh. I had another question about article vs h-entry
#
Calli
Ever considered just replacing h-entry with <article>?
#
aaronpk
most of the new usage of microformats2 has been in the social web space
#
Calli
aaronpk: yeah - that stuff is new to me and looks very cool
#
KartikPrabhu
Calli: mf2 does not assume that a given HTML tag corresponds to some microformat
#
aaronpk
Calli: one of the goals of mf2 is to not dictate which html elements people use, so the ability to use h-entry on things other than <article> is good, and also means smaller spec with fewer edge cases
#
Calli
makes sense
#
Calli
And how stable is the MF2 parsing spec?
#
aaronpk
the guts of it hasn't changed in a long time. there are a couple things in the works right now but it's things like how to add things like language attributes
#
Calli
thanks