#microformats 2014-07-23

2014-07-23 UTC
lmjabreu_, tantek and KartikPrabhu joined the channel
#
@sbmaxx
Какая херня этот hcard
(twitter.com/_/status/491828926864699392)
pfefferle, tobyink, krendil, RCheesley, KartikPrabhu and barnabywalters joined the channel
#
Loqi
barnabywalters: tantek left you a message 10 hours, 55 minutes ago: I found a page with hProduct but phpmf2 doesn't seem to find it: http://pin13.net/mf2/?url=http://www.staples.com/Swingline-747-Rio-Red-Stapler-20-Sheet-Capacity/product_562485 - however mf2py picked it up: https://gist.github.com/tommorris/5d9111af10b606c3f3bd (via tommorris)
#
barnabywalters
good grief, that’s disgusting
#
tommorris
barnabywalters: eh, just Unicode encoding nonsense. :)
#
barnabywalters
not that, the XML inside HTML with duplicate elements for microformats and whatever the XML vocabulary is
#
barnabywalters
all of which is invisible data
#
tommorris
the v:url stuff is schema.org
#
tommorris
in RDFa
#
tommorris
xmlns:v="http://rdf.data-vocabulary.org/#"
#
tommorris
rel="v:url" property="v:title"
#
tommorris
typeof="v:Breadcrumb"
#
barnabywalters
no, I’m talking about the bit of markup which actually contains hproduct: https://gist.github.com/barnabywalters/66bc200190c4ae52844e
#
tommorris
also, schema.org is confusing - rdf.data-vocabulary.org lists Breadcrumb as a valid type. but schema.org doesn't.
#
tommorris
inconsistent documentation - from a major tech company? Say it ain't so.
#
barnabywalters
aggressively ignores schema.org
#
tommorris
I'm not sure what a Breadcrumb type is supposed to be for and why that kind of thing can't be inferred from something like <nav>
#
KartikPrabhu
schema is a good example in overengineering
#
KartikPrabhu
i actively deleted old schema markup from my site when I upgraded to mf2
#
tommorris
I have RDFa on my site mostly to prove a point. I'm rather proud of this...
#
tommorris
<footer class="vcard h-card" typeof="schema:Person foaf:Person geo:SpatialThing gnd:Person yago:LivingPeople yago:SoftwareDeveloper yago:LGBTPeopleFromEngland yago:Adult109605289 yago:AlumniOfTheUniversityOfLondon yago:PeopleFromSurrey" about="http://tommorris.org/" id="me">
#
barnabywalters
horrible markup aside, it’s my fault the backcompat isn’t parsing as I somehow forgot to add hproduct backcompat. fixing now
#
barnabywalters
tommorris: uh, aren’t some of those mixing properties and types a little?
#
barnabywalters
or is this hypothetical? :)
#
tommorris
barnabywalters: and, yes, there's a bit of a mixup there. but I found existing types that are already used and reused them.
#
barnabywalters
e.g. type: yago:SoftwareDeveloper, job-title: Head of Forestry
#
tommorris
a property is just a more flexible expression of a type. ;)
#
barnabywalters
obviously such contradictions are possible with saner types, but less likely
#
barnabywalters
I suppose that’s true
#
tommorris
Well, without some kind of validation layer, there's nothing to stop you expression logical contradictions - RDFa, microdata, microformats
#
tommorris
<a rel="me parent child" href="...">
#
tommorris
(to be fair, there are theological uses for such markup...)
pfefferle joined the channel
#
barnabywalters
!tell tantek thanks for the heads-up about hproduct, for some reason the backcompat mappings were missing. Pushed to master, will add to the next tagged release
#
Loqi
Ok, I'll tell him that when I see him next
#
barnabywalters
tommorris: thanks for the ready-built class mapping :) what’s the rationale behind adding e-description to the classes for review?
#
tommorris
I don't know, tbh. ;)
#
barnabywalters
I removed it for maximum spec-compatibility
#
barnabywalters
assuming that if you had a solid reason with real-world examples it would be documented on the wiki
#
tommorris
I may have misread the spec at that point.
#
ChiefRA
barnabywalters your parser highlights the errors too ?
#
ChiefRA
I'm interested in parsing a mf1 hlisting and to spot the implementation errors
#
barnabywalters
ChiefRA: what do you mean?
#
barnabywalters
php-mf2 doesn’t do any particular error reporting
#
barnabywalters
as it’s quite difficult to figure out what an “error” is
#
ChiefRA
:-? but it should be no?
#
barnabywalters
it sounds like the sort of errors your talking about are the job of a validator, not a parser, to detect
#
ChiefRA
in deed, validator :)
#
ChiefRA
do you know any?
#
barnabywalters
e.g. indiewebify.me/validate-h-entry/ for h-entry
#
barnabywalters
not for h-listing, unless google’s rich snippet testing tool handles them
#
barnabywalters
I don’t even know what hlisting is tbh :)
#
ChiefRA
yeah, the problem with Google is that it doesn't say anything about it, after parsing.
#
ChiefRA
It would be tremendously useful such a validator.
#
ChiefRA
it's for real estate
#
ChiefRA
listing all the properties one after another, like hproduct
#
ChiefRA
but specifically designer for real estate.
#
barnabywalters
apparently it wasn’t widely enough used to make the transition to microformats2 (yet, at least)
adactio joined the channel
#
ChiefRA
well, that's in part Schema's fault.
#
barnabywalters
it sound like there’s a lot of overlap with h-product
#
ChiefRA
yes, to some point, but being specifically designed for Real Estate, it should help.
#
ChiefRA
the thing is, I need a validator :) do we have a validator for h-product at least?
#
ChiefRA
I can use it like that by replacing the main name.
#
barnabywalters
nope — building useful validators is a lot of work and I personally only have time and expertise to maintain them for the most commonly used vocabularies
#
barnabywalters
as can be seen by the number of issues on indiewebify.me :) https://github.com/indieweb/indiewebify-me/issues
#
ChiefRA
ok, so do we have ANY kind of mf2 validator?
#
ChiefRA
do you*
#
barnabywalters
but feel free to adapt any code from indiewebify.me which is useful into a h-product/hlisting validator
#
tommorris
there's not really any validator yet?
#
barnabywalters
“mf2” validator — validators generally are per-vocabulary
#
barnabywalters
so indiewebify.me has validators for h-entry and h-card
#
ChiefRA
ok, it seems like we need to add the h-product as well.
#
tommorris
I mean, we could have something that says "oh, you've put in a u-name rather than a p-name" but pretty much all properties are optional
#
barnabywalters
http://shrewdness.waterpigs.co.uk/test/ isn’t exactly a validtor yet but a testing tool for h-feed
#
tommorris
all you can really do is chuck markup at a parser and see whether it parses right.
#
barnabywalters
tommorris: indiewebify.me doesn’t even do that (u- vs p-) level of validation yet as it just looks at the parsed output
#
ChiefRA
Yeah, I would love to have a validator which can point you in the right direction, like: hey, your u-photo is outside of the corresponding DIV or smth like this.
#
barnabywalters
“see whether it parses right” — is very easy for humans but a lot more difficult for computers :)
#
tommorris
so, the problem of validation is you've got two parts to it - does it syntactically make sense and does it make logical sense?
#
tommorris
microformats is very light on the latter, somewhat intentionally
#
ChiefRA
barnabywalters yes, totally agree :) that's why we have to do it if possible :)
#
ChiefRA
to keep up with the Schema spreading...
#
tommorris
and syntax isn't really a microformats concern - that's a concern for the HTML parser.
#
tommorris
is there a schema.org validator?
#
ChiefRA
Google's own tool
#
barnabywalters
a logical validator for each of the hundreds of types? or more of a syntatic validator?
#
ChiefRA
barnabywalters I would say a syntactic validator, to look for the classes and to tell you that you have missplaced the elements in-out of the divs etc.
#
tommorris
"Error: Missing required field "entry-title"."
#
ChiefRA
as ussually that's the problem.
#
tommorris
Oh, Google, so cute, you honestly think that entry-title is required.
#
barnabywalters
tommorris: lol
#
tommorris
I still enjoy the fact that a few years ago, everyone told me that my strange insistence that titles ought to be optional was completely mad. Now they all use Twitter.
#
ChiefRA
:)))))
#
ChiefRA
seems that title is still a ranking factor, although browsers desn't display it anymore since TABs are taking over
#
ChiefRA
tabbed browsing
#
tobyink
If somebody supplemented http://schema.rdfs.org/ with some useful disjunctions (you can't be a Dentist and an OceanLiner simultaneously) then any OWL processor fronted by an RDFa/Microdata parser could act as a very nice schema.org validator.
#
tommorris
tobyink: what relation does schema.rdfs.org have to rdf.data-vocabulary.org ?
#
tobyink
http://rdf.data-vocabulary.org/rdf.xml is old. schema.org replaces it. schema.org's schemas are only published in HTML, so schema.rdfs.org is a third-party RDF-ification of them.
#
barnabywalters
tommorris: what if I told you… that tweets were all title
#
tobyink
http://rdf.data-vocabulary.org/ChangeLog - no changes since 2009.
#
tommorris
barnabywalters: then you'd build a very odd-looking renderer indeed. ;)
#
tommorris
tobyink: only, I found 'Breadcrumb' in rdf.data-vocabulary.org/rdf.xml but not on schema.org
#
ChiefRA
there is breadcrumb on schema.org
#
tobyink
rdf.d-v.org was Google's solo effort at creating an RDF vocab, before they threw their lot in with MS and Y!
#
tommorris
ChiefRA: ah, okay. thanks.
#
ChiefRA
and it has even a well defined implementation syntax :)
#
ChiefRA
Schema.org is winning ground in front of mf...
#
ChiefRA
the thing is, I still work with A LOT of HTML4* websites,
#
ChiefRA
so I can';t implement schema....
#
tommorris
Why do you care about HTML4? The version number is an illusion.
#
tommorris
There is no HTML4 or HTML5, there is just HTML.
#
barnabywalters
unless of course you actually care about validation
#
ChiefRA
well, indeed, but if I change it, I get A LOT of w3c errors.
#
ChiefRA
Oh, I do care about it :)
#
tommorris
Who cares about validation? ;-)
#
ChiefRA
me me , pick me!!
#
ChiefRA
:))
#
barnabywalters
ChiefRA: out of interest, why?
#
tommorris
Rephrased: why care about validation?
#
ChiefRA
it's an important step in keeping your code clean and "understandable" by browsers etc.
#
ChiefRA
Google does not take into account - is not a ranking factor - validation.
#
ChiefRA
but can get you down as a side effect, if Googlebot can't fully understand your HTML code.
#
tommorris
so, I'm just looking at the validator.nu results. they're silly.
#
ChiefRA
if there is harsh competition on certain niches, Googlebot will pick your competition instead of your site if it can better comprehend their HTML code.
#
barnabywalters
it can be useful for spotting syntax errors, but google has it’s own tools (as you posted earlier) for making sure it’s crawlers can understand your code
salanto joined the channel
#
ChiefRA
John Mueller said is not mandatory to validate your website, but it certanly helps if you do. :)
#
barnabywalters
so if google’s tools tell you they can understand the rich snippets, why care whether or not the markup is offically allowed in whatever fake version number of HTML you’re forced to publish?
#
tommorris
It's presuming I want to use RDFa Lite 1.1 rather than RDFa Full. It's whining about me using rel values that aren't listed on microformats.org.
#
tommorris
It's telling me that my <article> elements need to have <h1> elements even though there's no need.
#
ChiefRA
tommorris I do agree that they (Google) are not perfect.
#
ChiefRA
they do have a lot of room for improvements.
#
ChiefRA
but that depends on us, to place bug reports etc.
#
ChiefRA
they can't be everywhere anytime to spot all the bugs unless our community helps them with it.
#
tommorris
So, tommorris.org has 83 validation errors according to validator.nu. Most of them aren't errors.
#
ChiefRA
I always submit bug reports and alike when I find them, and not necessarely HTML related.
#
ChiefRA
I can give you the most recent example: because I submitted an Android bug related to how Android is displaying numbers when you save them in the phone book and how it's displaying them when you dial, now they fixed the bug and my contry, Romania, has the numbers sorted out the right way, as we are used to see them on all other devices.
#
ChiefRA
and that's worldwide implemented on all Adroid devices.
#
ChiefRA
so I do believe spotting the bugs and taking the time to report them helps.
#
barnabywalters
ChiefRA: cool, nice that you got to see the whole feedback loop!
#
ChiefRA
barnabywalters :) it took me 7 months, but I've lived to see the result on my Galaxy S5 :D
#
ChiefRA
that's why I proposed that if you can, you should implement other vocabularies into your validator barnabywalters :)
#
barnabywalters
interesting that validator.nu is funded my Mozilla: http://about.validator.nu/##I+would+like+to+thank
#
barnabywalters
ChiefRA: you’re right, I think there’s some value in a more generic syntax validator as well as vocab logical validators. I might try putting one together
#
ChiefRA
barnabywalters please by all means, please tell me about it when you do, as I will be testing it to report porblems :D
#
ChiefRA
Error: Almost standards mode doctype. Expected e.g. <!DOCTYPE html>.
#
ChiefRA
From line 1, column 1
#
ChiefRA
to line 1, column 121
#
ChiefRA
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">↩<html
#
ChiefRA
=)) validator.nu has a LOONG way to go to be near perfect.
#
tommorris
ChiefRA: why? it's accurate. You don't need anything the old-school DOCTYPE
#
tommorris
it literally serves no purpose at all.
#
ChiefRA
I believe you, but if I change it to HTML5 doctype, I get a LOT of errors.
#
tommorris
So, fix the errors.
#
ChiefRA
which have to be fixed, and the coders are way too busy to do that.
#
tommorris
The browser doesn't care about the DOCTYPE string.
#
ChiefRA
and it's a whole platform.
#
ChiefRA
I can't fix it manually for each website.
#
ChiefRA
it has to go through a long process.
#
ChiefRA
so I skip it for now and try to patch things up, like using mf instead of schema etc.
#
ChiefRA
Don't get me wrong, i love mf.
#
ChiefRA
the only thing i cry about is that hlisting is not YET supported by Google.
#
ChiefRA
I mean it sees it, but that's about it. It doesn't treat it in any way in search resulta.
#
ChiefRA
results*
#
tommorris
You can use both microformats and Schema.org (either as microdata or as RDFa).
#
ChiefRA
I know. but that involves changing the DOCTYPE :)
#
ChiefRA
so here we go again into the same problem :)
#
tommorris
The DOCTYPE doesn't do anything, so there's not really a problem.
#
ChiefRA
damn, it's the first time I heaf about this term.
#
ChiefRA
thanks :D
#
ChiefRA
I'm smarter now.
#
tommorris
Derives from Richard Feynman's excellent "cargo cult science" analogy.
#
tommorris
so, the only reason people use a doctype other than <!DOCTYPE html> is for validation. but the validation doesn't tell you anything interesting except that it validates to a DTD/XSD that has no bearing on how HTML is actually interpreted by the browser.
netweb joined the channel
#
tommorris
so you can get stuck in the loop of "well, I have to have the DOCTYPE because I need to validate it and I can't change the DOCTYPE because then it won't validate"
#
tommorris
well, the validation isn't telling you anything worth caring about, and the DOCTYPE isn't doing anything interesting.
#
ChiefRA
yeah, i believe so...
#
ChiefRA
did you guys knew about this? https://code.google.com/p/aump/
#
ChiefRA
it's kinda old if I look at the archive date of creation.
#
barnabywalters
ChiefRA: I hadn’t come across it, do you use it?
#
tommorris
I could say <!DOCTYPE html PUBLIC "Tom's Magical HTML variant - http://tommorris.org/magic/pixies.dtd"> but that doesn't mean WebKit or Gecko is going to treat the page any differently. All it cares about is the "DOCTYPE html" bit to not chuck it into quirks mode
#
ChiefRA
nope, I just stumble upon it now, searching for hlisting into the Google code, for some real examples.
#
ChiefRA
tommorris totally agree with you :)
#
ChiefRA
mentality it the FIRST thing that has to change
#
ChiefRA
it's*
#
tommorris
ChiefRA: so, if you want an hListing parser, nag me and/or barnabywalters and hListing can be added to php-mf2 and/or mf2py
#
ChiefRA
I do want it if possible, but I would love to have it as a validator of somewhat
#
ChiefRA
if not, I can manually debug it with the help of the parser output.
#
ChiefRA
ba da.
#
ChiefRA
tommorris and barnabywalters if you will create the Hlisting parser please leave me a PM :) I'll happily use it :)
#
ChiefRA
I have to go now, see you all later. and thanks :)
#
barnabywalters
ChiefRA: bye!
#
tommorris
ChiefRA: I'll file a ticket on mf2py.
#
ChiefRA
ok, thanks :)
#
barnabywalters
well the first thing which needs to happen is for a h-listing mf2 vocab to be created
#
ChiefRA
I can do that.
#
ChiefRA
I can submit it on the proper page.
#
ChiefRA
thanks, I submitted a ticket to get my Github password back.
#
ChiefRA
:)))))))))
#
barnabywalters
okay I vastly improved the php-mf2 testing interface as the first steps towards making it into a syntatic validator: http://waterpigs.co.uk/php-mf2/
#
barnabywalters
at the very least, Codemirror does basic HTML syntax checking
pfefferle and pfefferle_ joined the channel
#
ChiefRA
back
#
tobyink
front
#
ChiefRA
aside :)
pfefferle_, adactio, philipashlock, gRegor`, TallTed, trendynick_ and pfefferle joined the channel
#
@cishetboy
THANKPETE IM LAUHGING OS HCARD CHARLOTTe
(twitter.com/_/status/491976746422767616)
tantek joined the channel
#
Loqi
tantek: barnabywalters left you a message 6 hours, 25 minutes ago: thanks for the heads-up about hproduct, for some reason the backcompat mappings were missing. Pushed to master, will add to the next tagged release
jschweinsberg, chiui and uf-wiki-visitor joined the channel
#
tantek
good morning - thanks barnabywalters!
#
barnabywalters
morning tantek — no problem
#
tantek
tommorris: AFAIK the data-vocabulary stuff is *not* the same as schema
#
tantek
it was *yet another* attempt by Google to create a hierarchy of types
#
tantek
which schema rewrote yet again
#
tantek
the data-vocabulary.org stuff was pure political posturing - when launched in 2009 with claimed RDFa support, Google wasn't even parsing RDFa yet - it was just a nod in attempt to seem neutral with respect to syntaxes. and no one was yet publishing any rdfa data-vocabulary either.
#
tantek
in 2009 the only thing that google parsed that actually worked (at launch) was microformats
#
tantek
they did eventually get the rdfa parsing working, but at launch it was vapor
#
tommorris
interesting. should really dig out the sources for all this stuff and make a Wikipedia article on the history of web data or something
#
tantek
tommorris: I have a bunch of stuff cited re: the even older attempts by google and others in the recent microformats.org blog posts
#
tantek
Google's legacy of structured data attempts could make for its own article
#
tantek
from Google Base, to gData, to data-vocabulary they've had so many attempts
#
tommorris
I've been listening to the web history podcast. very early stuff: Mosaic, Netscape, early IE etc.
#
tommorris
tantek: I was once contacted by a publisher who wanted me to write a book called "The Case for GDATA". I wrote back and told them I didn't think there WAS a case for GDATA.
#
tantek
tommorris: feel free to start documenting them each on their own pages (by name of technology) on the microformats.org wiki
#
tommorris
Apparently, I was right.
#
tantek
so you don't have to deal with notability nonsense
#
tantek
since anything regarding past structured data research is perfectly reasonable and certainly in scope for documenting on microformats.org
#
tommorris
knows notability rather well as he's had to decide on it hundreds of times. ;-)
#
tantek
but may be too "inside baseball" for wikipedia
#
barnabywalters
tommorris: Re: GDATA that’s hilarious
#
tantek
just saying, I know you're a WP admin and all and know all that stuff
#
tommorris
yep, it's worth having places outside of WP for stuff that falls out of scope - am also an admin on wikiqueer.org for LGBTQ history stuff that's not quite WP-worthy.
#
tantek
that way you can also show that you documented it first more thoroughly in a more topical area / site
#
tantek
and WP can have summaries instead
#
tantek
whereas on microformats.org we can be thorough with documenting all the past history of overly hyped formats and such
#
barnabywalters
wow, I had never heard of “ROR”: http://www.rorweb.com/
#
tantek
so many articles talking about how Google Base was going to kill this that and the other
#
tommorris
barnabywalters: there was a strange fetish for putting metadata inside RSS feeds for a while. see also: GeoRSS
#
tantek
also a fetish for calling anything XYZRSS, e.g. see MediaRSS
#
tantek
another Yahoo deadend IIRC
#
tommorris
I thought MediaRSS was still used by some podcast clients?
#
tommorris
Didn't Apple nick some of that for iTunes Podcast support
#
tommorris
oh, wait, no, they made their own XML NS - xmlns:itunes="http://www.itunes.com/dtds/podcast-1.0.dtd
#
tantek
of course they did, XML culture is all about make up your own snowflake babel namespace
#
tommorris
on the µf wiki, do we need a deadlink/archive template btw? a lot of the historical stuff 404s now
#
tommorris
on WP, citation templates have the ability to put in "archiveurl" and "archivedate" parameters, and it explicitly notes that it's a link to an archive version. could do something similar or could KISS and just use links to archive.org
#
tommorris
edited /hmedia (+86) "fixing MediaRSS links to archive.org"
(view diff)
#
tantek
tommorris: happy to take your guidance on that one - whatever you think is both easiest to edit/add, and frankly, copy paste to wikipedia
#
tommorris
I'll play about with some later on
#
tommorris
it's quite useful to be able to tag broken links and then go through and fix them all using Special:ExternalLinks
#
tommorris
edited /hmedia-fr (+86) "replacing MediaRSS links to archive.org"
(view diff)
#
tommorris
edited /alternates-examples (+42) "replacing MediaRSS links to archive.org"
(view diff)
#
tantek
tommorris: perhaps just make a /wiki/MediaRSS page and then link all the references on other pages to that?
#
tantek
and then move all the archive / dead links / references to the /wiki/MediaRSS page?
#
tommorris
Eh, all done now. ;)
#
tantek
just saying, if MediaRSS is really that dead, we might as well make a new canonical place to document things about it
#
tommorris
Indeed.
#
tantek
documenting old formats is something microformats.org can still help with
#
tantek
especially the trail of wreckage of xyzRSS and abcML attempts
#
tommorris
edited /rdfa (+383) "expanding"
(view diff)
#
tommorris
edited /xoxo-opml-issues (+4) "linking [[OPML]]"
(view diff)
#
tommorris
edited /OPML (+297) "RFC 822 (and not even that!) vs. ISO 8601"
(view diff)
#
tantek
wow tommorris is really having a good time tonight ;)
#
tommorris
I have a lot of grumpiness to channel. I should probably take it out on the Stairmaster instead.
#
tantek
oh no, please go ahead and channel it on obsolete formats of yore. good times.
#
tommorris
you'll love the next one I'm just about to document
jschweinsberg and ivc joined the channel
#
tommorris
edited /Twitter (+0) "fix"
(view diff)
#
tommorris
edited /twitter-syntax (-33) "all the old tr.im shortlinks are dead. url shorteners really are a solution looking for a problem..."
(view diff)
philipashlock_ joined the channel
#
tommorris
edited /rel-tag (-33) "removing dead tr.im shortlink"
(view diff)
#
tommorris
edited /optimus (-33) "removing dead tr.im shortlink"
(view diff)
#
tommorris
edited /naming-principles (-32) "removing dead tr.im shortlink"
(view diff)
#
tantek
tommorris: did you see /wiki/tr.im ?
#
tantek
please be sure to create the alternate pages to increase findability for those dead tr.im links!
#
tommorris
oh, list them in equivalents?
#
tantek
create and list equivalents yes
#
tantek
for findability
#
tantek
and then replace (rather than remove) the dead shortlinks on the individual pages
#
tommorris
ah, I see it
#
tantek
tommorris: yeah I may be being a bit overly methodical about it, but I'm hoping to make it so that if/when people google for old tr.im URLs that we get top results for the relevant replacements
KartikPrabhu joined the channel
#
tommorris
edited /wiki-better-than-email (+759) "/* what if someone starts an edit war on an issue */ adding a discussion about Wikipedia's BRD"
(view diff)
robmorrissey joined the channel
#
tommorris
edited /User:TomMorris (-110) "tidying up"
(view diff)
#
tommorris
edited /tr.im (+125) "/* equivalents */ added some equivs"
(view diff)
#
tommorris
tantek: just reading up on the rel parsing rules, it might be an idea to deprecate rel-tag
#
tommorris
or rather deprecate uses of it where it is obviously intended to not be document-scoped.
#
tommorris
µf2 h-entry spec already discourages other entry-level rel stuff including rel-syndication and rel-bookmark
#
tantek
tommorris: agreed. please add to /wiki/rel-tag-issues - feel free to cite this in the logs too
#
tantek
KevinMarks won't be too happy about it but I think we lost that battle (rel-tag scoping) a while ago
jschwein` joined the channel
#
tommorris
edited /rel-tag-issues (+1388) "added issue: Formal deprecation of object-level rel-tag usage"
(view diff)
#
tommorris
tantek: ^^
#
tommorris
pour one out, ladies and gents, for the rel attribute.
#
tantek
well, for using the rel attribute with anything else
#
tantek
microformats2 parsing model has already adopted/enshrined the rel attribute as being useful at page level
#
tantek
tommorris - that works
#
tommorris
so, of the microformats that are rel-based, we've got XFN - which isn't really going anywhere because of widespread implementation
#
tommorris
and rel-nofollow - again, not really going anywhere because of widespread implementation and quasi-standardisation by Google et al.
#
tantek
there were also usability problems found with mixing rel/property in rdfa which provided the impetus to drop use of rel (or make it optional) in the changes from 1.1 to 1.2 I think
#
tommorris
I haven't kept up on RDFa development, alas. ;)
#
tantek
worries that he might know some aspect of RDFa development more than tommorris
#
tantek
tommorris: on the contrary, subsets of XFN have become essential/useful like rel=me and re=contact
#
tantek
and rel=author has become more useful
#
tommorris
I should clarify - by "not going anywhere" I mean, we're not going to suddenly change it, it's too entrenched.
#
tantek
nofollow is a deadend yes
#
tantek
due to abuse - it was a very bad design
#
Loqi
fo sho
#
tantek
oh sorry
#
tantek
Loqi, sometimes your delayed timing is pretty funny
#
tantek
anyway - issue documentation looks good
#
tommorris
now that the age of Technorati has gone and we're in the age of hashtags, rel-tag is deprecatable without causing too much pain
#
tantek
yeah - it was good market research. sigh. pour one out for scoped rel-tags.
KartikPrabhu joined the channel
#
KartikPrabhu
tommorris: agreed with the rel-tag scoping issue. Can't decide if that means rel-tag is completely useless
#
tommorris
KartikPrabhu: well, it's not necessarily useless if the scoping matches with the document. ;-)
#
KartikPrabhu
yes but when would anyone do that?
#
KartikPrabhu
specially with tag...
#
tommorris
permalink pages
#
KartikPrabhu
why use both rel-tag and u-category?
#
tommorris
there isn't really a reason to use both, but if it's just on the permalink page, at least the resulting parse is moderately useful
#
tommorris
it's when the tags are on the homepage that you get a problem
#
KartikPrabhu
will have to check my usage
#
tantek
KartikPrabhu: easier for more parsers to support page-level rel handling
#
tantek
also less work for the author
#
tantek
always good to enable / empower simpler (even if more limited) solutions
#
KartikPrabhu
it is more work for author if you say use rel-tag on permalink but not in feed page
#
tommorris
incidentally, we were discussing earlier with ChiefRA and barnabywalters the issue around microformats validation - checking to make sure that you don't use rel-tags inside h-entry (or even inside <article> and other semantic-level elements) is something worth checking
#
tantek
same reason we have rel=in-reply-to and u-in-reply-to
#
tommorris
we should probably have a page for useful validation rules. will write one in a sec.
#
tantek
tommorris: sounds more like linting than validation
#
tommorris
yep, indeed.
hadleybeeman and jschweinsberg joined the channel
#
tommorris
hadleybeeman: http://microformats.org/wiki/PDF mentions gov.uk's standardisation of PDF. ;)
#
tommorris
edited /pay-to-read_standards (+246) "a quote from timbl"
(view diff)
#
hadleybeeman
If you want to be pedantic, tommorris, the wording you want isn't "recognised by the UK government's policy paper;" take it from "PDF/A must be used for static versions of documents produced for download and archiving that are not intended for editing, in particular PDF/A-1 and PDF/A-2."
#
hadleybeeman
But, overall — glad we could help. :)
#
tommorris
edited /PDF (+128) "thanks hadleybeeman"
(view diff)
#
tommorris
speaking of ISO standards, can anyone remember what the US military standard version of ISO 8601 is?
#
tommorris
because that's published free-to-read unlike the ISO version
gRegor`, KevinMarks and krendil joined the channel
#
tobyink
ANSI INCITS 30-1997 (R2008)?
philipashlock and tantek joined the channel
#
@prettvodd
RT @cishetboy: THANKPETE IM LAUHGING OS HCARD CHARLOTTe
(twitter.com/_/status/492082338629943296)
#
tantek
never heard of the hCard OS
KartikPrabhu and tantek joined the channel