什么是沼泽图?

发布于 2024-07-07 14:48:38 字数 126 浏览 12 评论 0 原文

有人告诉我沼泽图,解释说它们可以通过测量给定产品的传入缺陷率和传出修复率来预测代码质量。

不幸的是,我无法找到有关这些图表的更多信息,我想知道这是否是特定于一家公司的术语。

你能解释一下什么是沼泽图吗?

Someone told me about swamp diagrams explaning that they were useful to predict code quality by measuring the rate of incoming defects and outgoing fixes on a given product.

Unfortunately, I am unable to find additional information on those diagrams and I am wondering if it is a jargon term specific to one company.

Can you explain what a swamp diagram is?

如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。

扫码二维码加入Web技术交流群

发布评论

需要 登录 才能够评论, 你可以免费 注册 一个本站的账号。

评论(1

雨夜星沙 2024-07-14 14:48:39

您可以在本文中看到有关“调试和CESR 的性能特征”,pdf 第 5 页或第 14 页。 该文件 1988 年。

(CESR 是康奈尔电子存储环,旨在提供质心能量高达 16 GeV 的碰撞电子和正电子束。)
(显然,我只是从这里的文章中复制内容,我只是一个编码员,而不是物理学家;))

现在沼泽图的有趣之处在于重新分配比率方面,您可以轻松地根据两个轴查看数据组(“沼泽”)。

如果您的两个轴是:

  • 传入缺陷率和
  • 传出修复率

您可以可视化修复的性质:当发现大量错误时修复过程是否有效?
这可以告诉您很多有关所发现缺陷的性质的信息(“很多”或“沼泽”可以指许多非常简单的错误,因为在大量文件中重复出现愚蠢的拼写错误并且很容易纠正,但其他很多也同样重要,但修复得非常缓慢可能表明存在更多的结构性问题,可能会影响您的应用程序的架构选择:您想要专注于那个沼泽)

最后,不,它不是某个公司的内部术语,只是一个指旧图表的旧术语。 ..在给定领土上实际的地理沼泽重新分配。 (如这篇关于“早期波利尼西亚人破坏森林" 从 1977 年开始显示,返回几页查看该地图)


正如 David Segonds 所说注释,现代名称将是“二项式趋势线”(尽管比率方面在这种图表中往往会有点丢失)。
您可以在此按年龄划分的选民投票率图表。

沼泽图的另一个现代示例是这个图表,将每个县显示为椭圆形,其大小为椭圆与两次选举中该县的人口(更准确地说是选民投票率)成正比。 (奇怪的是,似乎只有与这种数据表示相关的政治示例才会弹出;)

[免责声明:以下只是一个示例,绝不说明此处的任何类型的政治观点;)]

http://andrewgelman.com/wp-content/uploads/2008/09/1896c.png

You can see an example of a "swamp diagram" in this article about the "THE COMMISSIONING AND PERFORMANCE CHARACTERISTICS OF CESR", page 5 of the pdf or p. 1988 of that document.

(CESR is the Cornell Electron Storage Ring, designed to provide colliding electron and positron beams up to center-ofmass energy of 16 GeV.)
(obviously, I am just copying stuff from the article here, I am just a coder, not a physicist ;) )

Now what is interesting about a swamp diagram is the repartition ratio aspect, the way you can easily see groups ("swamps") of data against two axis.

If your two axes are:

  • the rate of incoming defects and
  • the rate of outgoing fixes

You can visualize the nature of the fixing: is that fixing process efficient when a lot of bugs are found or not ?
And that can tell you a lot on the nature of the defects found (that 'lot' or swamp can refers to many very easy bugs due to a stupid typo repeated in lots of file and easily corrected, but that other lot quite as important but fixed really slowly may indicate a more structural problem affecting perhaps the architectural choices of your application: you want to focus on that swamp)

Finally, no, it is not an internal term from some company, just an old term referring to old diagrams of... actual geographic swamp repartition over a given territory. (as this article about "Forest destruction by early Polynesians" from 1977 shows, go back a few pages to see that map)


As David Segonds says in a comment, the modern name would be a "binomial trendline" (although the ratio aspect tends to get a bit lost in that kind of diagram).
You can see an example in this Graph of voter turnout by age.

Another modern example of a swamp diagram would be this diagram showing each county as an ellipse, with the size of the ellipse proportional to the population of the county (more precisely, the voter turnout) in the two elections. (weird, only political examples seem to pop-up in relation with that kind of data representation ;)

[Disclaimer: the following is just an example, and in no way illustrates any kind of political opinion here ;) ]

http://andrewgelman.com/wp-content/uploads/2008/09/1896c.png

~没有更多了~
我们使用 Cookies 和其他技术来定制您的体验包括您的登录状态等。通过阅读我们的 隐私政策 了解更多相关信息。 单击 接受 或继续使用网站,即表示您同意使用 Cookies 和您的相关数据。
原文