One can read only so much before it comes time to join a discussion and ask some questions. There are two major venues for discussions about CSS, but each is concerned with a specific type of discussion -- so make sure you go to the right place.
This Usenet group, often abbreviated as ciwas (pronounced "see-wass"), is the gathering place for CSS authors. A number of experts in the field check this newsgroup regularly, this author among them, and all are there for one primary reason: to help new CSS authors over the hurdles that learning any new language will generate. The secondary reason is for the spirited debates that Figure 9-10.
Finally, there is overflow:auto. This allows user agents to determine whatbehavior to use, although they are encouraged to provide a scrollingmechanism when necessary. This is a potentially useful way to useoverflow, since user agents could interpret it tomean "provide scrollbars only when needed." (They maynot, but they certainly could, and probably should.) occasionally erupt over some aspect of CSS, or a browser's implementation thereof. Rather unusually for a newsgroup, the signal-to-noise ratio stayed fairly high for the last few years of the 1990s, and will with any luck continue in that vein.
Anyone who wishes to be involved in discussing the future course of CSS, and to clearing up ambiguities in the specifications, should subscribe to this list. The members of the list are all, in one fashion or another, interested in making CSS better than it is already. Please note: www-style is not the place to ask for assistance with writing CSS. For help with CSS authoring problems, visit ciwas instead. Questions beginning with "How do I ... ?" are frowned upon by the regulars of www-style and are usually redirected to a more appropriate forum such as ciwas. On the other hand, questions that begin "Why can't I ... ?" or "Wouldn't it be cool if ... ?" are generally welcome, so long as they relate to some ability that appears to be missing from CSS.
Messages to www-style are only accepted if the sender is already subscribed to the list. In order to subscribe, send email to firstname.lastname@example.org with the word subscribe in the subject of the message; to unsubscribe, send email to email@example.com with the word unsubscribe in the subject of the message.
Copyright © 2002 O'Reilly & Associates. All rights reserved.block-level status helps explain why when an element is floated,other content flows around it.
Remember that if you float atextelement, its width will tend toward zero. This is exactly theopposite of the normal horizontal rules, wherewidth is increased until the seven propertiesequal the parent's width. A floatedelement's width will default to auto, which underlining for the B element. Navigator, Explorer, and Opera all do this, if there is an explicit text-decoration: none to cause the suppression of underlining. This is, of course, what an author would tend to expect, and that's why the browsers do it.
The caveat here is that browsers (or any other user agents) might oneone character wide -- assuming that to be the browser'sminimum value for width. In practice, it'smore likely that the browser will make the floated element as narrowas the longest word in the element, as shown in Figure 8-29.
The names of these properties make them fairly self-explanatory. Herewe have one possible solution for the example given in the previous
Before we get there, however, there are a few more things to cover.Remember that absolutely positioned boxes can have backgrounds,margins, borders, and padding; styles can be applied within them,just as with any other element. This can make them very useful forthe creation of sidebars, "stickynotes," and other such effects. One example is the ability toset a "change marker" on any paragraphthat has been edited. This could be done using the following stylesand markup: