请不要混淆标准和开源软件

有些人想把自由和开放源码软件的理念与标准结合起来，实际上把讨论变成了“开放标准”。这混淆了两种截然不同的想法，一旦你超越了它们都涉及到技术社区的合作。首先，它们在市场中为不同的经济目的服务。标准的存在是为了支持和鼓励能够跨标准规范定义的边界互操作的多种实现。一个只有一个实现的标准，无论该标准得到了标准组织的认可，都是失败的。自由/开源软件项目本质上是单一的协作开发的技术实现，不管它们是否实现了一个标准。正式的标准由标准开发组织(SDO)开发，SDO可以是全国性的(如ISO、BSI、ANSI、AFNOR、DIN)、基于行业的(如IEEE)或基于技术的(如W3C、IETF、OASIS)。每个SDO遵循规范开发、投票、维护和退役的各种严格流程。自由/开源软件项目可以在正式的过程中开发，如果由一个基金会(例如Eclipse或CodePlex基金会)进行管理和赞助的话，但是这些项目通常都是非正式的精英组织或者是仁慈的独裁组织。SDO的文化也非常不同。在本质上，它们必须发挥委员会的作用，而不管它们代表的是什么共同利益。 Many of the experts that participate are employees of large vendors. It's a game of technology diplomacy, and in the end, after the meetings are over, the job as with all diplomats is to expand one's area of economic influence and defend sovereign territory. The collaboration culture inherent in FOSS projects is about sharing the common costs involved developing a component all can use. There is a place where standards and FOSS overlap beautifully. If you have one true FOSS implementation that runs universally then you don’t need a standard specification. FOSS can however provide the reference model around which a new standard is created. Standards typically occur when a vendor technology matures and the incumbent starts to over deliver on features consumers don’t want to buy. Consumers begin to complain about price and competitors begin to discuss a specification at an appropriate layer of abstraction. Good standards are based on existing practice and experience. The starting point chosen for a standard is generally a technology base with which all the participants have experimented. When competitors chased the DEC minicomputer market, they did so with the UNIX specification. Everybody had a UNIX variant, and the licensing landscape was well understood. At this point in history, vendors can choose a FOSS project as the starting point for specification, and indeed that is what happened with the打开文档格式(ODF)规范它与OpenOffice项目有关。标准和自由/开源软件项目在市场中扮演着不同的经济角色。它们差别很大，但可以很好地相互补充。他们的另一个共同点是，他们都容易受到专利滥用的影响，但那是另一天的专栏了。