On 2007/05/28 02:44 (GMT-0400) Philip Kiff apparently typed: > Felix Miata wrote:
> I included the 2nd link to the Briggs article because I thought that perhaps > the first link might not have been understood since it went directly to the > a page of Briggs's images. I realize that you have spent considerable time > studying this issue, but your explanation of Briggs's technique seems > misleading to me. Under Briggs's technique, the body font-size is set to > 76% and then the p font-size is set to 1.0 em. All other elements are then > sized with ems. This should not produce tiny fonts on most people's > systems: that is the whole purpose of his going through the exercise of > producing all the screenshots using different browsers and operating > systems. Although the screenshots date back to 2002, they do include IE 6, > and I doubt there are differences in font-size rendering between IE 6 and 7 > that would make Briggs technique suddenly unusable. Context was largely my point. Start by catching up with other bits that Briggs has to say on http://www.thenoodleincident.com/tutorials/typography/incremental_differences.html and http://www.thenoodleincident.com/tutorials/box_lesson/font/index.html and note his rather strong bias against the defaults, and the date of the original writings: "most browsers default to a text size that I have to back up to the kitchen to read" "the browser defaults are huge, like 200% of program toolbar font. Absurd." The windoz UI default is 8pt, while its browser default is 12pt (the Linux desktops I've used seem to have standardized on 10pt as the UI default). Even though it appears he's exaggerating, as 12 is 150% of 8 and not 200%, those numbers are of nominal sizes, not real sizes. Size is a function of area, which is determined by both height and width. At 96 DPI an average 12pt letter lives in a box of about 128px (8px wide, 16px tall), while an 8pt letter in about 72px (6px wide, 12px tall), or 77.7% bigger in real size for IE content default compared to windoz UI. So he's exaggerating only somewhat for the difference, but he's way off base for his characterization, even back in the period. The UI doesn't need to be and shouldn't be as large as the content. Content is unfamiliar territory, and generally there's a lot more of it, and it's commanding a lot more effort and attention. UI is mostly just little bits grabbed here and there. They're presumably familiar, and command little time. Your back won't suffer the same pain of leaning forward to see UI that it would leaning forward through whole web sites. The eyes can usually adjust readily to the difference between UI and content. Smaller they should be, in order not to distract from content, and to distinguish from content. When you focus on the results represented by his screenshots, the validity of the samples are primarily valid for the context of the pre- and early-CSS period, when display PPI didn't vary a whole lot from one local environment to the next, when sub-16px sizes were presumably still reasonably legible for most users, and when 16px was indeed "too big" for the average user. When deviating merely 1% from his recommended 76%, the consistency at sub-100% that was his purpose breaks down. Today we have considerably wider PPI variation and significantly smaller average size of a pixel. He effort has traveled considerably down the path between highly practical value to wholly academic relic. The major point that remains valid is that setting a size in body cascades down into everything else, but that's the inherent nature of CSS. > Briggs's method will produce pages where fonts appear similar to what they > appear like if you use 12pt text as your base font-size. Surely you meant 12px. > This is the size > that is still used today by millions of websites. No doubt some people find > that size too small, but that is still the norm on the web these days. I Obviously you meant 12px. If the majority of sites were using 12pt we wouldn't be having this discussion, as 12pt is what most ordinary users prefer. http://psychology.wichita.edu/surl/usabilitynews/2S/font.htm > don't quite understand the issue with the different dpi displays. Won't > that have the same affect on all browsers, regardless of what method is used > to size fonts -- unless you use pixel sizes, of course? The method of sizing via body remains valid. The presumption that the defaults are too big no longer fits. > I agree wholeheartedly. Different viewports and preferred sizes are > perfectly OK. But if a designer finds a way to make sites appear almost > identical across all major browsers and platforms at a screen resolution of > 1024x768 on a 17" monitor with everything else set at default settings, and > those sites are STILL scalable for other users, then shouldn't that be OK > too? When done right, there's no need to depend on a particular size as a starting point, and thus no reason to shift overall text size up or down by any perceptible amount. Any such shift made is wholly arbitrary. Visitors should find the page ready to use on arrival, and not need to fiddle with adjustments before it's suitable for personal use. >> 65%-80% produces a uniformity of substantially reduced accessibility >> and usability that 100% basing does not do. Whether 65%-80% is >> intended to disrespect visitors is irrelevant; only the fact that it >> does is. > Again, I would question the use of the term "disrespect" here. Respect is a > human value and feeling, and it certainly does depend on notions of intent, > and on human cultural constructs. That is why it is possible to disrespect > someone without intending to. You (designer) don't know my preferred size. You do know browsers come equipped with adjustments so that browser operator whim can be accommodated. You do know that some people do make some adjustments, and that others don't. Of those that don't, you have no idea how many don't because they find no reason to, and how many would make some change if they knew they could. You intentionally make a body rule setting size to 80%, which you know is to be applied to every visitor, even though you have no idea of the actual physical size suitability of 100%. The above is a very good definition of disrespect, and it is a quite intentional act on your part. Application of the word to that scenario was not my invention. Thinking of the problem as an issue of respect likely happened when I first visited http://www.w3.org/QA/Tips/font-size. > According to your view, the majority of > websites out there "disrespect" their users. I don't think that most users > feel this sense of disrespect the way you do. Probably not. Nevertheless, those materially beyond their teen years typically want to know why most web page text is so tiny. I tell them it's because most web designers are under 40 and/or have average or better vision. I tell them it's because designers assume things they shouldn't be assuming. I ask how many with poor vision do they suppose would work a job in front of a computer screen most of each day? I ask how many do they suppose in the youthful realm that is the internet have the wisdom of many years to incorporate into their effort? They usually understand that voluntarily working with computers is still the realm of mostly the younger half of the adult population, plus a not insignificant number of pre-adults. The classification as disrespect comes primarily from the intentional application of arbitrary adjustment upon an unknown and unknowable starting point, reinforced by the W3C recommendation language. It would be nice if the WCAG working group would deal with this issue directly instead of sidestepping it with their adjustability language. -- "The path of the righteous is like the first gleam of dawn, shining ever brighter till the full light of day." Proverbs 4:18 NIV Team OS/2 ** Reg. Linux User #211409 Felix Miata *** http://mrmazda.no-ip.com/ ******************************************************************* List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm Help: [EMAIL PROTECTED] *******************************************************************
