Difference between revisions of "microformats-ja"

From Microformats Wiki
Jump to navigation Jump to search
(→‎microformatsとは何か?: 北村暁さんの訳を基に翻訳 )
(→‎the microformats principles: 北村曉さんの訳を基に翻訳)
Line 28: Line 28:
* 上記のいずれでもない
* 上記のいずれでもない
== the microformats principles ==
== microformatsの原則 ==
* solve a specific problem
* 特定の問題を解決する
* start as simple as possible
* できるだけ簡単に始める
** solve simpler problems first
** 最初に簡単な問題を解決する
** make evolutionary improvements
** 進化的に改良を行う
* design for humans first, machines second
* 第一に人間のために、第二に機械のために設計する
** be presentable ''and'' parsable
** 見栄えよく、''そして''機械的に解釈できるように
** ''visible data'' is much better for humans than ''invisible metadata''
** 人間にとって''見えるデータ''''見えないメタデータ''よりも断然良い
** adapt to current behaviors and usage patterns, e.g. (X)HTML, blogging
** 例えば(X)HTMLやウェブログのように現在の振る舞いと用法パターンに順応する
** [http://tantek.com/log/2003/0813t1158.html#handauthoring ease of authoring is important]
** [http://tantek.com/log/2003/0813t1158.html#handauthoring ease of authoring is important]
* reuse building blocks from widely adopted standards
* 広く採用された標準に少しずつ追加していく
** [http://tantek.com/presentations/20040928sdforumws/semantic-xhtml.html semantic], [http://tantek.com/presentations/2005/03/elementsofxhtml meaningful (X)HTML]. See [[SemanticXHTMLDesignPrinciples]] for more details.
** [http://tantek.com/presentations/20040928sdforumws/semantic-xhtml.html semantic][http://tantek.com/presentations/2005/03/elementsofxhtml 意味付けされた(X)HTML]。詳しくは[[SemanticXHTMLDesignPrinciples]]を参照。
** existing microformats
** 既存のmicroformats
** well established schemas from interoperable RFCs
** RFCによって規格化された方が良い
* modularity / embeddability
* モジュール化する / 埋め込めるようにする
** design to be reused and embedded inside existing formats and microformats
** design to be reused and embedded inside existing formats and microformats
* enable and encourage decentralized and distributed development, content, services
* enable and encourage decentralized and distributed development, content, services

Revision as of 18:30, 18 August 2006





  • 新しい言語
  • 無限に拡張可能であり、無制限なもの
  • 皆がその振る舞いを変え、そのツールを書き直させようとする企み
  • 今日既に働いているものを無駄にする真新しいアプローチ
  • すべての分類学(taxonomies)、オントロジー(ontologies)および他のそのような抽象化のための万能薬
  • 全世界を定義したり、広範囲に悪影響をあたえるだけのもの(defining the whole world, or even just boiling the ocean)
  • どこかの個人や組織にコントロールされているもの
  • 上記のいずれでもない


  • 特定の問題を解決する
  • できるだけ簡単に始める
    • 最初に簡単な問題を解決する
    • 進化的に改良を行う
  • 第一に人間のために、第二に機械のために設計する
    • 見栄えよく、そして機械的に解釈できるように
    • 人間にとって見えるデータ見えないメタデータよりも断然良い
    • 例えば(X)HTMLやウェブログのように現在の振る舞いと用法パターンに順応する
    • ease of authoring is important
  • 広く採用された標準に少しずつ追加していく
  • モジュール化する / 埋め込めるようにする
    • design to be reused and embedded inside existing formats and microformats
  • enable and encourage decentralized and distributed development, content, services
    • explicitly encourage the original "spirit of the Web"

  • Related Principles we re-use from other design paradigms
    • DRY (Don't Repeat Yourself)
    • Least Surprise
    • Pareto Principle (80/20)
    • Data Integrity. One of the common objectives which many of the principles help achieve is data integrity.
      • Visible data = more accurate data. By designing for humans first and making the data presentable (thus viewed and verified by humans), the data is inevitably more accurate not only to begin with (as errors are easily/quickly noticed by those viewing the pages/sites), but over time as well, in that changes are noticed, and if data becomes out-of-date or obsolete, that's more liklely to be noticed as well. This is in direct contrast to "side files" and invisible data like that contained in <meta> tags.
      • Not repeating yourself (following DRY) - means there are fewer chances for inconsistency
      • Multi-language integrity. Perhaps not a principle, but many of those involved with microformats have found that consistently using UTF-8 helps ensure that the human text content itself is not corrupted, especially when using non-ASCII7 characters.



more thoughts on how microformats are different

There are plenty of existing formats that are nearly totally useless/ignored.

They're not totally useless though. They're useful in that they illustrate what at least someone thought might be useful, which unfortunately is typically a lone-inventor working a-priori without any domain expertise.

Or there is the other extreme. Lots of corporate inventors working with plenty of experience, over-designing a format for what might be needed some day. In particularly bad cases, the corporate vendors collude to prevent openness and/or adoptability by the open source community. Media standards often suffer from this kind of deliberate "strategic" positioning.

We seek to combat all of those problems with the microformat approach.

  • We're not lone-inventors; we're a community.
  • We don't work a-priori ("from reason alone"); we require documentation of existing examples, previous attempts at formats. See The microformats process.
  • When lacking domain expertise, we seek out the domain experts to provide it, and we immerse ourselves in examples and prior art from the domain (see previous point).
  • We do our work in the open with open discussion forums.
  • We're a diverse mix of corporate, independent, hobbyist, enthusiast.
  • We don't over-design. We under-design, deliberately, and then only add things when they are absolutely necessary.
  • We adopt very liberal copyright/licensing (CC,GMPG,IETF,W3C) and patent positions (RF,IETF,W3C).
  • We ruthlessly self-criticize based on our principles in order to keep to the above.

Some ask what the purpose of the (intended) standards is.

Why do you need purpose? More often than not, premature focus on purpose tends to distort data formats towards a particular application which may not be all that relevant. Hence rather than focus on a-priori purpose, we focus on modeling existing behavior, with the knowledge that additional structure will yield plenty of interesting uses, most of which we will not be able to a-priori predict.

This is obviously a very different approach than traditional data format efforts.