[Wylug-discuss] a webdesign query / mini rant

[chris]

i spent most of friday and today redesigining the page to be strechey! 
it now works. so no more forced resolution and no more px size values!!

it took me a long time to figure out what was goign wrong to brake 
positioning in MSIE but building it up from scratch solved most of the 
problems so it was probably user error!

>
>Hi Chris,
>
>I'd like to backup Smylers on this and to clarify a point that I made
>earlier.
>
>When I said that it was difficult to get stretchy designs right, I
>didn't in any way mean to suggest that people should use fixed or
>pixel-based layouts.
>
>Like Smylers, I consider stretch design to be an intrinsic part of cross
>platform interface design and a requirement for usable web pages.
>
>The main point I was trying to make was that of the two layout methods
>(float vs position), the latter is more robust, because of the
>catastrophic way in which float layouts break.
>
>If you have borders on your column divs, stretchy positioned layouts can
>also break due to insufficient container width, but they do so less
>dramatically.  
>
>Typically, columns at the left and right will slightly overlap columns
>in the centre of the page. In some circumstances, the container will
>simply expand beyond the right edge of the viewport (browser window).
>Both are ugly, but neither is catastrophic.
>
>With good knowledge of CSS fundamentals, and a certain amount of
>experience, you can make both float and positioned stretchy layouts
>very pretty and pretty much unbreakable across browsers. 
>
>This is partly a matter of avoiding settings which prevent related
>widths from scaling proportionately to one another (e.g. mixing pixel-
>or em-width borders with percentage widths on the rest of a box, or
>unecessary use of the width property when margins and offsets will do).
>
>Its also partly a matter of knowing when to use the less obvious
>features of CSS (e.g.  using auto widths and min/max dimensions to let
>the browser re-size margins or content to keep stacked boxes within
>their containers).
>
>Having said that stretchy designs are good, and quite achievable with a
>bit of effort/learning, I would like to question the wisdom of _always_
>using percentage widths on columns.
>
>It's a long established fact that people find it hard to read very long
>lines of text (> 8-12 words), even when they don't have to scroll
>horizontally.
>
>Setting widths in ems and also setting maximum widths (ems or
>percentages?) would allow horizontal stretchiness without taking it to
>extremes. Remember, if mad people really want 30 word lines, they can
>always override your settings in a user stylesheet (browser
>preferences).
>
>Having made that caveat, I think its worth saying that I can't think of
>any sane justification for setting a pixel-based height on a page.
>
>It's a general principle of CSS design, that you can't sensibly
>constrain the height of text boxes, because users will always have
>different font-sizes/screen-sizes, so you can't predict the total space
>that everyone needs for the same textual content. 
>
>Since horizontal scrolling makes text much harder to read than vertical
>scrolling, the obvious thing is to allow 'excess' text to flow down the
>screen as far as it needs to.
>
>Dave
>
>
>
>_______________________________________________
>Wylug-discuss mailing list
>Wylug-discuss at wylug.org.uk
>http://mailman.lug.org.uk/mailman/listinfo/wylug-discuss
>  
>