Potential of Microformats and Future of Web Pages

If you know what “ microformats ” are, then I suspect that you have a good knowledge of Web 2.0 services and their related technologies. In the past, I have touched on microformats several times in this column, and recently, a front-end engineer, Yagura, wrote an article entitled “Support for Microformats and Future Outlook” in the Web Standards Blog. In the present column, I would like to reintroduce the basics of microformats, give a short summary of the current trends surrounding microformats, and finally give my personal opinion on the future outlook of Web pages.

The term “microformats” consists of the words “micro” and “formats.” The former implies “very small,” and the latter is literally the word “format” in plural. Hence, the term implies “very small formats” when translated directly into Japanese. While there are numerous ways of representing a piece of information, there exist a number of formats that are widely used for representing specific types of information. Think of a business card, for example. It typically consists of the person's name, email address, and information about the organization to which the person belongs, such as the postal address, phone number, and facsimile number.

However, even if we mark up such specific information using (X)HTML, the browser or a user agent cannot understand whether a piece of information represents a name or phone number because the elements required to enable this have not been defined in the current (X)HTML specifications. On the other hand, if new elements were to be incorporated in future specifications, a new problem would arise: it would take an extremely long time to standardize a specification, notwithstanding the actual process of making changes. Even if a new technology format different from (X)HTML could be developed within a short time frame, the learning cost and interoperability between the existing technologies could pose new problems.

This is the reason why the concept of microformats has been proposed. The technique is based on (X)HTML and is designed such that it provides additional semantics that are not included in the (X)HTML specifications without violating any of the specifications. By utilizing specific attributes and attribute values corresponding to different types of information, the meaning that microformats convey can be extracted mechanically. The design of microformats is not developed by the World Wide Web Consortium (W3C) ; rather, it is developed at a grassroots level where everyone can participate freely through discussions (mainly in English) in an extremely open arena. Strictly speaking, this technology does not form a part of the Web standards; however, I believe that the technology specification can be cast as being very close to a Web standard. I have interviewed Mr. John Allsopp, one of the central figures in the microformats community, to hear his thoughts on the relationship between microformats and Web standards. The video recording of the interview can be accessed from the links given below.

Currently, many different types of microformats have been proposed, including those for which the specifications have been finalized and those that are still in draft form. Some examples of microformats include hCard , which can hold personal and organizational information; hCalendar , which can be used for storing calendar and events-related information; XFN , which can be used for describing human relationships; and rel-license , which can convey license information. There also exists a microformat called rel-tag used for tagging information in Blogs and social bookmarks, since tagging has recently become popular. In addition, if you are interested in search engine optimization (SEO), you would have seen the microformat rel-nofollow used for instructing search engines not to evaluate a link destination. Please refer to the microformats Wiki for a current list of microformats. There is also a Japanese translation created and maintained by volunteers.

Is the use of microformats becoming more widespread? The answer could change depending on personal opinion. However, considering the fact that large Web service vendors including Google , Yahoo! , and Technorati , and more recently, personal Blogs, have started using them, it can be said that this technology has gained a certain level of popularity and recognition. Having said this, it is rare that the benefits gained from microformats can be experienced by Web browsing due to the limited support provided by Web browsers. In this respect, this technique has unfortunately not yet evolved out the “chicken or the egg” stage of development.

Let me introduce an implementation example of microformats. If you use Mozilla Firefox as your Web browser, try installing an add-on called Operator and then access the page “ Visit Us .” You will see that the Operator add-on will automatically detect the microformats embedded inside the page, allowing you to extract the contact information of the Mitsue Links headquarter and its studios. This information can be used for various purposes; for example, it can be easily incorporated into Outlook or Address Book applications in an electronic business card (vCard) format. Certainly, you can achieve the same outcome by manually copying and pasting the information from the Web browser; however, I think you can see from this example the merit of being able to extract specific information mechanically.

As stated above, by utilizing microformats, you can increase the machine readability of specific types of information. This suggests that even static Web pages could become an application programming interface (API—interface used by applications). Like John introduced above, Mr. Tantek Çelik also plays an important role in the microformats community. In his lecture notes entitled “ Microformats at Stanford ” from the lecture that he gave at his alma mater in March this year, he mentions as one of the definitions of miroformats, the fastest and simplest way to provide an API for your website.

A semantically and grammatically valid Web-standards compliant Web page provides a certain level of machine readability and therefore can function as an API. However, this can be considered an API only in the context of a Web browser and Web search engines, and the machine readability that can be provided is limited by the semantics of the (X)HTML specifications. However, by introducing microformats that have been standardized by the community, you can expect to gain richer semantics and the associated convenience in the form of a highly functional API.

Furthermore, the learning cost required for implementing microformats is inherently low since it utilizes the existing standard technologies. Microformats are not the only initiative to make information on Web pages machine-readable, but it is expected that their use will become more widespread due to their low cost of learning. If the support for the web browser side continues to progress, then this technology could evolve from the “chicken or the egg” stage of development in the not too distant future.

Ultimately, information must be interpreted by humans and not by machines; based on this assumption, consideration for the overall design and optimization of the mechanism by which information is presented is indispensable. However, I also believe that in the future, Web pages should contain a higher level of functionality through the provision of APIs. It may also be necessary for the API consumer (i.e., applications) to match the pace in development; however, the two can be considered to be the wheels of the same car. As usual, we would like to focus on how to generate value for business in relation to the process of marking up information and continue our investigation and research into the effective use of microformats and metadata and develop new services for our customers.