introduction-fr

Introduction aux Microformats

Related: press, presentations, podcasts, suggested-reading, testimonials

Que sont les Microformats?

Les microformats sont les "dictionnaires" du contenu sémantique XHTML. Ils ont pour objet de codifier le contenu HTML riche-en-information couramment utilisé, de telle façon à ce que le contenu sémantique puisse être extrait par une machine. En bref, ils sont la manière la plus simple d'architecturer des données (ou formats de données) destinés à être réutilisés sur le Web.

Pourquoi les Microformats

Pourquoi en sommes-nous venus aux microformats?

En bref, les microformats sont la convergence d'un certain nombre de tendances :

1. une étape à venir logique dans l'évolution du "web design" et l'architecturage de l'information.
2. une façon pour les éditeurs autonomes de publier eux-mêmes une information plus riche, sans devoir s'en remettre à des services centralisés.
3. la reconnaissance du fait que les efforts "traditionnels" en matière de métadonnées ont soit échoué, soit pris tellement de temps pour engranger une quelconque adoption, qu'une nouvelle approche était nécessaire
4. une façon d'utiliser le (X)HTML pour les données (data).

Evolution du Web Design

Au commencement (1990), il y avait le HTML, et il convenait bien. Il était simple, minimal, et utilisé pour baliser semantiquement les données visible de l'utilisateur (texte) et les partager sur le World Wide Web.

Ensuite vint la guerre des navigateurs (1994-1999) qui vit les éditeurs dominants de navigateurs took their turns en introduisant des étiquettes "innovantes" pour la presentation, donnant aux auteurs/designers typiques du web ce qu'ils voulaient : un semblant de controle sur la presentation de leurs pages web. Le résulat : HTML 3.2 "standardized" these defacto presentational innovations.

The introduction of CSS1 (1996) and the semantically richer HTML4 (1998) brought a glimmer of hope, but it wasn't until years later (2000-2001), with the introduction of fully compliant (or almost) implementations of CSS1/HTML4 (IE5/Mac, IE6/Windows, Netscape 6) that it became practical for web designers to depend on CSS in their web pages. Leaders in the community began to furiously adopt and promote CSS (even if it took a hack or two) and the efficiencies and enhanced productivity that separating presentation from markup brought them, yet remained a small vocal minority.

The introduction of the Wired News redesign in 100% CSS, and the beautiful CSS Zen Garden (2002-2003) was CSS's tipping point. With the clear and obvious presentation of visual beauty and broad creativity, designers world-wide "got it" and realized that this was the future of web design. The presentational markup of <FONT>, <TABLE>, and spacer.gif were tossed aside by any and all self-respecting web designers, who discovered the near infinite flexibility of <div>, <span>, and the 'class' attribute. A few in the community even began adopting some of the more semantic elements in HTML: <p>, <h1>...<h6>, <ol>, <ul>, <li>, <em>, <strong>. Leaders in the community exercised the semantic limits of strict HTML4 (experimented with XHTML) and documented best practices.

As the community followed rapidly in the footpaths they had worn, the leaders began to run into the limits of semantic (X)HTML. Other subcultures were attempting to rewrite the world in their own language(s) (RDF, "plain" XML, SVG), yet not having much of an impact on the World Wide Web, which required human presentable data, compatible with the browsers people already used. Social Software and Blogs, written by this new generation of web designers and programmers, began to take off.

Natural patterns emerged from the way people used blogging systems, putting things into lists, for example lists of other bloggers (known as blogrolls), and annotating them with information representing relationships such has having met, friends, family, etc. The first microformat, XFN, was designed to match these behaviors, and introduced to the blogging community (2003-2004), who adopted it within weeks. The GMPG was formed as a home for XFN, and documented a few key design principles later adopted for microformats. The key notion, that semantic (X)HTML could be extended, had been introduced and accepted by the community.

By understanding, using, and combining semantic (X)HTML building blocks, as well as determining that semantic (X)HTML could be validly extended via new rel, meta name, and class values, defined in (X)HTML profiles in the XMDP format, the community began to design and develop many more microformats (2004-2005). More patterns emerged from the blogging community, and each aggregate human behavior drove the design of simple, adaptive microformats to meet its needs. Creative Commons licensing became popular and rel-license was proposed. Outlines and lists: XOXO. Contact info: hCard. Calendars and events hCalendar.

Using these new found building blocks, the web design and information architecture communities were no longer limited by the predefined semantics of HTML4 (nor did they have to compromise human presentation and ease of authoring which other attempts sorely lacked). 2005 may well be the year that microformats became the next step in the evolution of the web.

L'Attrait de la Simplicité

• Microformats are a simple effort which has appealed to many frustrated with previous complex efforts. One parallel that can be drawn is to REST in the web services world, i.e. see this comparison of microformats and REST. See the REST wiki for more about REST. Related microformats and microprotocols.
• See also: Web Services and the Innovators Dilemma by Justin Leavesley

Pages en rapport

• Recent press interviews and articles are also a good introduction.
• See microformat presentations for more background and introductory material on microformats.
• Listen to podcasts about microformats.
• More suggested reading
• So you wanna implement microformats?

Articles

• Digital Web Magazine: Microformats Primer by Garrett Dimon
• An Extra SIDE to Web Standards Based Design - How Semantic Information Design Ethics (SIDE) and a few readily available techniques can help heal the Web, by Jon Tan. See also his CSS Zen Garden "Leggo my ego" entry.
• Andrew D. Hume has written a blog post introducing microformats and another one on usable microformats.
• Practical Microformats with hCard by Drew McLellan
• Jesse Skinner's introduction to hCard
• Shaun Shull's great post on How Microformats Affect Seearch Engine Optimization
• Blog Business Summit: Microformats in Plain English: the Promise of Simple Business to Business Data Exchange by Steve Broback

Références diverses

These are various intro-related links/articles which I haven't figured out yet how to incorporate. You may find them of interest. - Tantek

• Data First vs. Structure First
  • Tantek says: In many ways it is actually *far* worse than that post conveys. The "typical" programmer literally loves spending far more time worrying about and designing the structure for structure's sake, than data, and even less so, "real world" data (current behaviors etc.). Hence we have taken the directly opposite tack with microformats when looking to solve a problem.
    • Zeroeth, define the real-world problem. If you can't do this, then stop.
    • First, look at real-world usage (data).
    • Second, what previous standards are people actually using today? If there is more than one, then lean towards those with the better adoption.
    • And only after those first two do we bother to pay attention to theoretical standards, those that have been invented (whether by individuals, committees), but haven't seen much if any actual adoption.
• 2000-03-21 Dan Connolly on human-consumable information: (strong emphasis added)
  • I believe that one of the best ways to transition into RDF, if not a long-term deployment strategy for RDF, is to manage the information in human-consumable form (XHTML) annotated with just enough info to extract the RDF statements that the human info is intended to convey. In other words: using a relational database or some sort of native RDF data store, and spitting out HTML dynamically, is a lot of infrastructure to operate and probably not worth it for lots of interesting cases. We all know that we have to produce a human-readable version of the thing... why not use that as the primary source?