Microformats: you’ve heard about them, you’ve seen them and you may even be using them – what are they, what are they going to become, where are they going and why do I care? I recently had the privilege of interviewing someone extremely knowledgeable on the subject to answer just those questions, Brian Suda of suda.co.uk.
Microformats seem to keep coming up in conversation and popping up more and more in the mainstream web. But are any of us ready for it? Are Microformats too premature for all of the exposure; perhaps even too premature for the pressure that we feel from “innovators” trying to push this new “technology”.
Brian is co-author of the hCard format, creator of the popular Microformats Cheat Sheet and now author of his newly written O’Reilly Short Cuts’, “Using Microformats”. I still have many questions, doubts and concerns about Microformats since my earlier post on the subject – so, I figure why not take the questions right to the source.
Martin: Brian, thank you for taking the time to answer some of my questions. I’ll get right to the point – what the heck are these crazy things called Microformats?
Brian: If you read the definition from the microformats.org website it says:
Designed for humans first and machines second, microformats are a set of simple, open data formats built upon existing and widely adopted standards. Instead of throwing away what works today, microformats intend to solve simpler problems first by adapting to current behaviors and usage patterns.
But what does that mean to the average web developer and designer? Well, microformats are away to encode semantic meaning right into HTML. The idea is not new, CSS gurus have been harping on folks for choosing “good class names” for years. The style you call “blue-box-300” which makes a blue box of 300 pixels makes sense today, but come next redesign, that is legacy code you are porting around which makes no sense when it describes a red box that is 80% wide. A better name would be something like “alert-box” or “error-callout”. We’re all guilty of doing things like this, but if we use more semantic names then lots of things begin to happen. First, we separate design and layout from content, which is what CSS was designed to do, but more importantly, if we standardize these semantic values then we can begin to “interact” with different web pages and we are sure we mean the same thing.
Part of the yearning for “human-readable” data is a historical one. Firstly, most search engines completely ignore the metadata inside the <head> element. This data was “gamed” by spammers and was usually out-right incorrect. The old saying applies “out-of-sight, out-of-mind”. Hidden metadata is much more likely to be “stale” and incorrect than data you are staring at on a daily basis. The second reason for making data visible because of the various files we keep on the server but do NOT view through a browser window. Files like our vcards or icalendar files, those require specialized software and are not viewed in the browser, so they are likely to be incorrect because you are not staring that data in the face every day.
The reason machines take a back seat to humans is because we are focusing on the publishers. A handful of folks will build the tricky applications to parse and use microformats, but the largest percentage of people will want to publish them – so the idea is to make it EASIEST for publishers, because those are the people who ultimately be reading and writing the microformatted data day-to-day.
Martin: A simple question, but important: who are Microformats for?
Brian: Microformats are for everyone. I have demoed microformats as several conferences and there is a sliding scale of people. Some people are extremely technical – rocket scientists, figuratively and literally, who work daily on complex semantic systems – to people who design beautiful handcrafted websites and have studied classical literature. Then there are the rest of us in-between. Microformats apply to each and ever one of these folks.
Martin: The concept sounds intriguing and I love all things “new” on the web. However, what are the benefits of using Microformats today? Is it more work than it is worth?
Brian: Certainly not, one of the simplest microformats is the rel-tag. One of the reasons, or so I’d like to think, that tagging has taken off is the direct feedback you get. I write a blog post, tag it. Submit it to a site like technorati, and within minutes I can then see my post listed under that tag category. That round-trip and direct feedback is what makes it worthwhile. We are starting to see more and more of that with the more complex microformats as well. Conference schedules are being marked-up with hCalendar so attendees can sync the data straight into their calendar applications. As events are updated on the website, the data in the calendar app is updated as well.
Martin: So, Brian, I get the gist of Microformats, however, where is the focus on Microformats right now? Is it on building the foundation for this new technology or is it actually using these within our sites? If I am not the pioneering type – should I wait till innovators like yourself have finished laying this foundation?
Brian: I’m not sure things will ever be “finished”. Much of the foundation has been laid and more and more large corporate sites are implementing microformats. If they have taken the time to evaluate the pros and cons, and they decided to implement microformats, then I think it is passing the ‘early adopter’ stage.
One organization told me that it took more time arguing about whether or not to implement microformats than to just do it.
The focus of microformats is always changing. The microformats site is now a year and a half old. In that time we’ve created several new formats, hammered out a pretty good process and have been evangelizing to get more and more sites, both big and small, to implement them. If I had a crystal ball I would say that the next 12 months will be focused more on getting the word out, documentation, and applications rather than new formats. My PDF book was just the start, I know there is another book in the work due out early 2007 and probably more on the way. Conferences have been interested and have been asking for more presentation about microformats – so evangelization will be big in 2007. Then, as always, as the formats mature and are in use more and more, the documentation, examples and uses will continue to grow.
Martin: So, you’ve sold me and I’m ready to start implementation. However, even among standards there are exceptions that make an initial understanding complicated. I see everyone implementing Microformats classes on different XHTML tags; is one way better than another? How does one know how or where to do it?
Brain: This flexibility is part of the beauty of microformats. There are two things to remember when adding microformats into your HTML. First, use the most semantic HTML element for the purpose and second, we don’t want to force the way publishers have to encode their information. In many of the examples on the microformats site, in the books and on blogs use <span> and <div> elements because those are semantically agnostic. In some instances you might have an event title in an <h1>, microformats don’t force you to change that to any other element.
I see how it can be confusing to someone new, they only see <div> and <span>, but it is certainly possible to use microformats on any element. That flexibility then leads to more confusion, but eventually you will have that ‘ah ha’ moment and understand that the data is the most important, not the mark-up.
Martin: Even better than how to start is where to start; where do I start with Microformats? Is it on my about page, my resume or contact page – is it all? It seems a little overwhelming. What if this hCard information isn’t information I want to share, like my email address for example. Can I be private and still utilize Microformats?
Brian: With hCard, and other microformats, very little is required. So if you are worried about privacy it is perfectly acceptable not to publish things like telephone numbers, addresses and email and still be valid microformat.
As for where to start and what to encode, an About page or contact page is a logical start. You known yourself best, so when you see data, even simple data like you name, it can be encoded with microformats. A friend of mine was over-joyed when he mention the Italian Renaissance in a presentation and was able to mark it up with hCalendar. Once you start to play with adding semantics, you begin to see them everywhere.
Martin: I’ve heard a lot of people debate the use of XHTML and classes to convey meaning – what are your thoughts on this?
Brian: The HTML class attribute has been assumed to be only for CSS. This is NOT true. According to the W3C HTML spec, it is a general attribute for user-agent processing. That is a very technical way of saying: you can use the class attribute to add more semantics into the document. Then the browser can connect those elements with associated data – sometimes that is a CSS style, but in the case of microformats it is semantics that correspond to people, places, organizations, events, reviews, etc.
Martin: I am hoping that this interview helps those in similar situations like myself – excited about Microformats but intimidated, unsure and slightly confused. You’ve been a great resource to myself with your Microformats Cheat Sheet and your new book, “Using Microformats”. Could you elaborate on these resources a bit? Why did you write them, who did you write them for and what will I get out of them?
Brian: Thanks, I’m glad they have been as useful as they have. The microformats website is a great wealth of information, but you sometime really have to know where to look. My attempt with the “Using Microformats” book was a condensed version of the microformats site and presentation I have given. It was all the important things you would need to know about in a short concise format with a few scenarios of “how cool microformats could be in the future if …”. It is ideal to give to someone who says “what are these microformats things” they can read the book and feel that they are confident enough to browse the microformats site and look for deeper answers, find what they need and/or ask the right sorts of questions on the mailing list.
The cheat sheet really surprised me. The microformats site does have a master list of all the possible values. In fact, it has everything the cheat sheet does and more. I just took that information and distilled it down to a single sheet. There was a quote somewhere, which I have forgotten, but it basically said something to the effect “if your spec if more than a page long it is too complicated to implement” so I tried to get everything to a single page. This also goes back to the question about “who is using microformats” hard-core tech-folks are comfortable reading lines of Regular Expressions in a dry verbose spec, but sometimes more left-brain design folks want a quick and dirty reference – and I think the cheat sheet fits this appeal. Ironically, I have written the hard-core-tech XSLTs to convert hCard to vCard and the cheat sheet and the cheat sheet has almost twice as many bookmarks on del.icio.us
Martin: It seems as though early adoption towards Microformats has been swift and successful – especially with companies like Yahoo! supporting the effort. What more is to come? What does the future hold for Microformats?
Brian: Microformats are easy to implement. One of the first big corporate sites to implement microformats was Avon. Yeah, that Avon – the women’s cosmetics empire. We got an email one day from their web developer. He had taken their template for the Avon sales person and added hCard to their contact details. Then when he republished the site, over 40,000 new hCards were available. Yahoo, Upcoming, Flickr, Eventful, Meetup and more have all been able to leverage microformats by editing their publishing templates and within minutes thousands to millions of instances appear!
I have no idea exactly what the future holds, but I do know that many smart people are in the microformats community and they bring back a lot of these ideas to their respective companies. I’ll be really excited when I can one day read a job description and see the company explicitly asking for “Microformats experience”. I know personally, I have been volunteering with all sorts of companies to help them get microformats into their sites. Corkd.com, linkedin.com, ma.gnolia.com and more have all been very open and receptive to working in microformats. I think they see it in several different lights; there is the altruistic reason, because they want to be a good net-citizen, they also see it as an opportunity for some publicity, but it also allows them to expose a lot of data without the need for a separate API, which makes the developers happy.
Just recently, there has been a new plugin for FireFox 2 called “operator” (link here) which detects and extracts various microformats found on a page. This is a plugin for FF2 so they can work out the bugs and streamline the code for possible inclusion for FireFox 3. Having microformat detection baked right-into the browser has loads of cool possibilities. Imaging as you browse a local database of microformatted data is collected and saved. Then you can simply search you local data for “all event in my zip code” and the browser would show you all upcoming events that match the criteria.
Of all the ideas about the future possibilities of microformats, I’m not interested in the ideas I can think of, but instead of all the possible ideas I can’t – that is what excites me. The “oblivious development” on the part of designers and developers allow all sorts of applications and mash-ups that no one can predict.
Martin: Brian, I’d like to thank you for taking the time to answer some of my questions. What really makes this industry an amazing is the passion that it can ignite. I think industry influencers such as Brian really help us push the envelope on what is possible and what can be possible.
I am more and more becoming a huge advocate for Microformats. I highly suggest that everyone visit Brian’s site at http://suda.co.uk.
You can learn more about Microformats online at: http://www.microformats.org.
Curious about the redesign? It's more of a design satire then a reflection of personal taste: Read More
Interesting interview. I suspect microformats will become more widely known in 2007. I wasn’t even aware of them until I searched for details on Technorati Tags. I’ve just launched TagBuildr, a tool that creates tags using the rel-tag microformat but keeps the links on my own site. This way, I can stop giving links to Technorati. Plus, the TagSummary pages that these tag links use should be useful for my blog readers.
Richard — I think you are right. 2007 is going to be the year of Microformats.
TagBuildr is very cool. I’ve only tried it once but plan on playing with it later this evening a bit more. Thanks.
I’m inspired now.
it’s awesome to see more and more companies and sites taking up Microformats. unfortunately it’s still in the phase where it’s cool in theory but close to worthless in practice. there’s so many possibilities, but so few implementations to date. of course, they’re so easy to add that you may as well just put them in anyway and let them wait until enough applications are available to make use of the data.
Thanks for the comments! I think we are going to see more and more from Microformats in the upcoming year. I am excited to start playing with them!
Thanks for conducting and sharing the interview. I hadn’t realized microformats were starting to hit critical mass. Seems like it’s time to play with them.
Romer!can,
I completely agree — definitely seems like this is the time to start “playing” with Microformats.