South By Southwest Interactive 2011

Jump to: navigation, search

Microformats events, content and so forth at SXSWi.


The Future of Microformats Panel


140 Character Summary for SXSW documentation

What is the future of structured data on the web? A look at new implementations, tools and hacks, plus updates on the microformats project.

Social media hashtag

hashtag: #futuremf

Brainstorming the Panel

Figure out the direction, content and general flow of the panel.

Subjects to Cover

  • Summarize the state of Microformats
    • “Are microformats dead?” (not necessarily in those terms…)
      • I'm curious about the self-consciousness behind this question. Is it an actual concern? One you have heard? Or just an ironic opener? Tantek 19:28, 30 January 2011 (UTC)
        • I've had it asked of me repeatedly, for a while now. The blog is neglected, there've been no new formats promoted (which is an external impression of progress), none of us work actively on it, supposedly stable specs like hAtom remain at sub-1.0 versions, the mailing lists are deserted. It is an entirely legitimate impression that the effort has folded into irrelevance. — BenWard 22:14, 4 February 2011 (UTC)
        • I've genuinely been asked if HTML 5 (various new elements) / Microdata replace microformats Phae 14:21, 31 January 2011 (UTC)
          • If the context is one of comparison, then let's handle that in comparison discussions (see below). - Tantek 23:35, 31 January 2011 (UTC)
    • Major implementations (to date)
      • Show an overview slide with logos of all the sites publishing microformats
      • Google search results - screenshots of rich snippets
      • RelMeAuth
        • There's no way this can be framed as a ‘major implementation’. It's a brilliant experiment, a great proof of concept. It's not a major implementation. Given Glenn's new hacks too, maybe we need to break this into two sections; actual major implementations and then hot new experiments that demonstrate how the data which is now out there and stable can be used. — BenWard 22:14, 4 February 2011 (UTC)
      • Could demonstrate Twitter Tweet Button using rel=me to infer data from embedding pages.
        • It's a tiny small use, but it is also all over the web. Not worth much time, but a nice frame of the concept of using microformats for progressive enhancement. — BenWard 22:14, 4 February 2011 (UTC)
      • Glenn Jones has a few neat things too
        • Possibly demo each very quickly? Glenn was asking about doing a demo too.
    • What microformats are there which are stable, ‘complete’?
      • This might be a good time to discuss potential thoughts about more formally defined microformats spec stages, e.g. draft (brainstorm consensus, ready for publishers), specification (stable multiple publisher support, implementations recommended), standard (multiple interoperable implementations) - I've brainstormed this a bit with Kevin Marks, perhaps worth further discussion at a dinner and capturing on its own wiki page).
  • Cover structured data in general (perhaps briefly, just to clear up confusion)
    • do a slide comparing microformats, microdata, RDFa (when you use it, which vocabs in use) - I think this might be a good lead in for the "Microformats 2.0" introduction.
    • What's microdata? What's it for? Where is it supported?
    • RDFa: Depending on further checking of the figures cited in Peter Mika's blog post (, what vocabularies are emerging in the RDFa space?
      • If the figures are wrong though, or there aren't any major vocabularies, then can probably skip over this. — Ben
        • Yeah, I'm a little tired of the deceptive marketing by the RDFa crowd (that article, and Ben's and Kevin Marks' trivial critiques (c.f. their twitters) are a prime example) and frankly in such a short session we shouldn't give such deceptions any time. However I *would* suggest a backup slide to debunk it (e.g. showing adoption of the class attribute or meta keyword in comparison to RDFa) when the inevitable one RDFa question comes up from the one RDFa person in the crowd. Tantek
        • We check the figures and if they are actually backed by a growth in particular vocabularies using RDFa, we mention them because they are relevant to structured data on the web. — BenWard 22:14, 4 February 2011 (UTC)
  • What's happening this year?
    • Any new vocabularies expected to be stable?
    • vCard 4.0, and how that will tie into hCard
      • updates to hCard in general hCard 1.0.1, 1.1 - Tantek
    • HTML5, and how the time element affects hCalendar, hAtom, etc.
      • Good to mention X2V/H2VX update to support HTML5 and time/datetime here. - Tantek
    • “Microformats 2” (datatype prefixes, generic tree parsing.)
      • If this is mentioned at all, assuming Tantek is even continuing this work at all(?), I am firmly of the view that it needs to framed as an exploration around a use case, not an accepted progression of microformats as a whole. — Ben
        • Yes it is continuing :) I agree with the general thinking behind Ben's framing, that is, we need to stay practical. Direct use cases are one subset of real world problems (which is how microformats started) that microformats solve. What's driving Microformats 2.0 *is* the solving of specific real world problems that sufficient developers have encountered (i.e. not just hypothetical) that it's now finally worth spending the time to solve them. And yes, the goal would be to have Microformats 2.0 "ready to use/parse" for introduction at this session, possibly with some publisher/parser support already available (thinking Twitter, Facebook, X2V, Operator offhand). - Tantek
          • I don't see how it's feasible to take ‘Microformats 2’ from its current rough draft documentation, without consensus, into a deployed and consumable state in the space of a month. Especially given how slowly and sporadically things move here. — BenWard 22:14, 4 February 2011 (UTC)
          • Additionally (and, I should stress, unrelated to my personal scepticism of the µf2 proposal) there's no way Twitter will be able to ship anything related to it before SXSW.
    • If we're covering individually led efforts though, Ben would like to frame up the concept of rich data containers and expanding parsing rules for those (vcalendar TV listings and the like.)
  • Community Management? Being that this has been (is?) one of the larger challenges the microformats community has had, is it worth bringing up, lessons learned etc.?
    • It's been on my mind a lot recently - particularly the reduction of traffic to mailing lists and potential stagnation of some proposals. I think we've got a tricky balance to strike. Phae 14:33, 31 January 2011 (UTC)
      • I think what essentially happened, to be blunt, is that the trolls "won". We had a sufficient # of trolls who overwhelmed the cognitive surplus of the admins (and the rest of the community, none of which were experienced with community management), that people (including myself) lost a lot of motivation and daily participation, IRC, etc. dropped off. I've been thinking a lot about this lately, and I think we actually have an opportunity to fix this, that is, to set up much stricter (and upfront) participation guidelines and consequences, and then act swiftly, and fairly merciless to ban those that are even minimally disruptive, because frankly, we don't have the staff to deal with educating (i.e. teaching better online manners etc.) those who are either new, immature, or just obsessive (i.e. can't help themselves from sending repeated emails to the lists, they're welcome to get such training on W3C lists where there are paid staff to deal with such folks). At that point, we can re-engage editors to step up activity, and then as admins very vigilantly patrol the lists and keep the clear of trolls. I would rather have higher quality traffic from a few productive folks, than noise from drive-by emailers. We need to act more like a meritocratic, hierarchical open source project (e.g. Mozilla) than the semi-free-for-all perceived flat "democracy" that so many standards lists (e.g. W3C) have. - Tantek 23:35, 31 January 2011 (UTC)
      • I think it's a bit much to pin all of this on trolls at this point. We banned them, they went away. After that—which was multiple years ago, remember—no one has done any major spec or community work with any visibility. There hasn't been any activity to draw people in or stimulate interest. We need to avoid playing some sort of excuses game here and focus on positive future effort, reasons why people should consider participating — BenWard 22:14, 4 February 2011 (UTC)
      • Relatedly, hNews might count as a valid effort which did have activity around it, but which was hosted around a different place than — BenWard 22:14, 4 February 2011 (UTC)
  • Accessibility? Since the value-class-pattern has solved these problems, is this worth mentioning at all?
    • May be worth a mention, if only because I think the date-pattern debacle put a few souls off. Phae 14:33, 31 January 2011 (UTC)
      • Phae, that was 2 years ago - literally - we covered accessibility and the value-class-pattern at the SXSW 2009 session. What's left to do is debunk the out-of-date Wikipedia article on microformats and any other place the debate resurfaces. Not sure why it would be worth spending time on this at SXSW 2011 except in response to Q&A perhaps. - Tantek 23:35, 31 January 2011 (UTC)
        • Fair enough, just thinking out loud :) Phae
        • We don't even need to do that concerning dates. At this point it's sufficient just to show HTML5 time elements in action. It may be reasonable at this point to consider updating all the specs to use time in the examples rather than old forms. — BenWard 22:14, 4 February 2011 (UTC)

Subjects not to cover at all

South By Southwest Interactive 2011 was last modified: Friday, February 4th, 2011