Did you just hear that? No? That was the sound of the Microdata Specification just quietly croaking in the background. In this edition of “While you weren’t paying attention” something very interesting just happened. And, I’m just going to get this out of the way, as I hinted that this would be coming in an article last year – I told you so. I also gave a very important take-away: It is not ok for marketers to not know about markup. I might add to that now and also say: It is not ok for marketers to not know about web standards and why they exist – and are often times very fluid (ehem, murky).
Last Thursday, August 15th, Manu Sporny released an article called: The Downward Spiral of Microdata. Manu is the chair of the RDFa Working Group, and has been heavily involved during the standardization initiatives of both Microdata and RDFa. Here’s what you need to know:
- Support for the Microdata API has been removed from Webkit (Safari) and from Blink (Chrome) a few months ago
- Microdata has also been ripped out of the main HTML5 specification at the W3C, “with the caveat that the Microdata specification will only continue ‘if editorial resources can be found’. Translation: if an editor doesn’t step up to edit the Microdata specification, Microdata is dead at W3C.”
Manu continues in his article to state that “Microdata doesn’t have an active community supporting it. It never really did”. Why this is an issue may have escaped most of us – it means that if you weren’t watching web standards, you may have just informed your clients to use Microdata – which is basically unsupported and dying.
What this means for Schema – well, we don’t know yet. It is very intriguing that a specification like Microdata would be released as a kind of open specification, but Schema.org (unlike the W3C) doesn’t really have open support – just 4+ million webmasters implementing Microdata markup on their sites to try to gain search rankings or at least some distinction in the SERPs. …UM…woops! Then there’s the part where if you’re going to release an open specification, but not open up the website that supports it for developers to continue to add and help expand it – well, you’ve just shot yourself in the foot. Schema will probably continue to support Microdata for some time, but how long is very unclear.
There was also an interesting exchange on the RDFa discussion list between Dan Brickley and Manu Sporny (yes, I lurk), where it has been pointed out, one of the main issues with Schema.org is that many web developers have asked the Schema.org group to at least upload the Schema.org site to github for people to contribute to the site. This at least would have helped make Schema.org adoption rise, and build a real community around a fractured vocab landscape, which was, I think, the whole point of the Schema.org exercise to begin with – to help develop common standards. Currently I believe there is about as much diversity in ontologies and vocabs out there for semantic markup as there are species on the planet.
This is about to start looking like an episode of the Walking Dead – because, if Google/Binghoo still insist on using Microdata for Schema, well that means Microdata could get resurrected at the W3C – since this is the only real strong usage that Microdata has seen. No one else is consuming data marked up via Microdata.
On the flip side, RDFa Lite is on the map, does everything that Microdata did and more, plus other Linked Data benefits to boot. A big issue is that the debugging tools offered for use aren’t very good either, and oftentimes incomplete in their validation. The syntax is seen by some to be for more “professional” developers, which some see as a major criticism. Microdata did seem easier for junior developers and architects to understand. But I don’t really follow that argument – itemprop and property are pretty much implemented in the same way. Given that there were at minimum two possible trajectories this could take, it would have behooved even junior developers to grok new syntax until the standards pendulum swings one way or the other. Google and Bing do not define these standards, they were trying to make the decision of which vocabs to choose easier – it really wasn’t up to them to say Microdata is the way to go and delete RDFa (or hide its application to this situation). For example – how many of you even realized this site existed: https://schema.rdfs.org/ ? Probably not many search marketers do, I would dare to say. And then there’s the fact that most of us have been waiting for all kinds of examples to make their way into Schema.org for over two years.
In the end, it’s all of our faults – if you want to have standards, and you want to have community backed standards, you really need to have a community that contributes to them. If you just spent the last two years going with the losing syntax in this markup battle, you maybe should contribute to its existence as a standard.
But let’s not forget where this is all going – eventually keyword-based search will go away as well, and we will all be querying (whether we know it or not) concepts and things. If you can’t grok that, at minimum, you should start looking for an occupation outside of the search world.
For those of you interested in where to get involved in web-standards, or just learn about them, this is the list of active W3C groups. Apparently today, HTML5+RDFa 1.1 will become an official W3C Recommendation (i.e., World Standard). It has massive amounts of support and development happening.
For those of you who are search marketers, you may be interested in an informal poll Aaron Bradley is running. Super great discussion and fairly balanced views demonstrated by a great number of thought leaders here.
Until next time – pay attention people.