<< Back Home
CSS Essays : "The Holy Font Size War"
or "Feel Free to Set Any Font Size You Like!"
There is a raging debate online right now. First let me set the backdrop for the war and then set forth the battlefield itself, the monstrous ramifications the w3c has set forth with their recommendations, and finally my ideas about who is right and who is wrong. Yes, I will try and sort out the mess for you. Keep in mind, though Ive got the podium, font sizing and your decisions as regards text and how you choose to size fonts in your web pages using CSS is ultimately up to you. (And that ends up being my whole point). Lets begin...
New Movement, New Confusion
There is a new movement online to change the very nature of how websites and pages are built and displayed. With all the billions of web pages out there on the internet, such a change will very soon mean that most site owners will have to begin to investigate how best to make the move to style sheet driven web pages. Such a change represents a huge effort for anyone involved in the development of any sort of technology solution using the Internet as a delivery mechanism, because at some point most of you will have to figure out how best to deliver that content using style sheets, and that choice will sooner or later, force you to interpret some standards, which CSS experts today are still not sure about.
Many web designers have been used to using tables and other html code solutions when designing web pages. Dozens of web development software packages today still support this table-based HTML design. The software vendors are clearly way behind the curve on helping this process move forward. This has helped keep most sites in the dark ages and today millions of websites world-wide are still neck deep in this old-fashioned means of designing web pages using HTML table-based layouts. But the new CSS movement is taking shape today, driven by devoted designers and coders online, which does bring us all, for the first time, some superior solutions that are fully tested and proven as solutions we can all use with some success. Most of these "experts" agree that CSS will bring in a new era of web design once more of the "best practices" are shown online in active sites as viable solutions. Such solutions will help prove the advantages CSS brings, such as separation of content from presentation, improved code maintenance, search engine improvements, and accessibilitiy advantages. The gurus now have proven in numerous demonstrations online that CSS works, and has clear advantages for those who take the plunge and start building Style Sheet driven web applications online. But some very subjective opinions remain as how best to apply style features in web applications, and new debates are raising their ugly heads as to whats good and bad design using style sheet solutions.
So, as the new CSS (style sheet) standard has moved many of us to completely rethink and redesign how we code our web applications and websites, so too it has brought some confusion and debate about the best way to apply these style sheet solutions correctly. Web authors not only have more power and control over their layouts and text, but so too a wide range of users is now given more control over the way they experience websites, and as such, both users and authors now are given more power to change or modify the look and feel of website designs, based on settings in their browser. But there is confusion about users, and authors, and the role each one should take, especially as regards font-sizing using style sheets and modifying text in web pages using styles. This is further compounded by many authors misinterpretation of PC and Mac operating systems, browsers, and font-sizing. So, these new exciting movements have brought new confusion. At the heart of that lies font-sizes, and so we bring forward a new war based on text and how its best sent to the users browsers, and how authors should style web pages. To fail, in terms of how you deliver fonts and text sizes, means complete website disaster, as readability is still a huge issue online. To succeed and use CSS to improve your websites readability can mean lots of kudos from your audience and web designers alike. But again, to fail with font design using the new style directive means information is lost to millions of potential readers online and one could face harsh criticism from the CSS and user communities online for that failure!
The BattleField
The battlefield is set, and the battle lines drawn. The opponents take center stage. Im ready to give you the players and their points of view....but first, lets ask some simple questions any web designer might ask:
"As a 'web author', how should I set 'font-sizes' for my web pages using style sheets?"
And one may then ask, "As 'web designer' and 'author' of my site, how much control should I give up so that users of all types have as much accessibility as possible to my site and its text?"
...and...."As 'web designer' and 'author' of my site, do I have to give up some freedom as to how I design my site? What can and what can I not do using style sheets and font-size properties?"
Finally, "As 'web user' and viewer, how do I set a my 'preferred font size' for all websites I surf online?"
We know the "web author", is the person responsible for building a style sheet that is attached to a web page so its styled as they intended. The author "should" have full power to style anthing in their page, as long as the browser supports it. Thats the general rule of thumb with style sheets and the new standard. We also know that when viewing a simple web page, the "web user" has the power to "change" their font "size" by using various properties in their browser. But most people are confused as to what those setting really do. We also are confused as to how changing font-sizes by the author using a style sheet is reflected in how various users with various "font sizing" selections view our page? Should the authors design be honored, or should users be able to "resize" that content as they need, and when? As web designer, do you have the freedom to size fonts as you like, or are you joining the new and growing ranks of those who harshly condemn that practice? We know users are more empowered to modify or "change", or even "set" font sizes and "preferences" as are authors in how they change fonts and text using the new CSS style standard. This raisies the spectre of how authors are supposed to "honor" those "preferences. And what does honoring or "respecting user preferences" or "honoring preferences" really mean?
Few people have articulated those question, much less the answers very clearly and thoroughly taking in consideratuion all the variables involved. Thats no easy task, and dont expect me to tackle that completely in this article. But some things could be made clearer, and that should begin with correcting incorrect assumptions.
First of all, most designers have made very bold assertions based on browser and css assumptions (and biases), and that lack of clarity and objectivity has fueled the debates raging now. Thats why we have drawn the battlefield, as its a war begin on misconceptions, bias, anger, confusion, context, and relative relationships....not facts. So, the battlefield has been drawn. And that battlefield has two major forces at war:
The "Purists" are those who believe the web author is required to "limit" styles such that supposed "user preferences" regarding font-sizes, especially, are honored and respected, so that author's sheets dont distort or change the user's preferences in the browser. They believe that no style rule should be used that "changes" or shifts the basic setting a user applies via the browser (usually the "View" section), and that one must practice restraint and extreme limitations, in some cases, on how and which style rules are applied in style sheets. Doing so, they believe that one can eventually create a new "web standard" for how all web authors design text, supporting web text that looks or supports as many user browser font settings as possble among all users. Its about a philosophy of making sure one supports those users, always, even if that means giving up some CSS features allocated to us by the W3c for web authors. There are deviations, but these are the general philosphies for this camp.
The other camp (what I call "The Purveyors" of standard Css font-size practices) is just as concerned with respecting user preferences, but sees user preferences, as set by the user in the browser, as "influencing" font-sizing in a web page, where such sizing is ultimately controlled by the author's style sheet. The author's preferred font size should be honored, they will argue. The CSS standard, they may argue, is designed that way, with the freedom to do what they like with sizes, they will claim. As long as some careful practices are used such that "resizing" and readability is offered, they argue such practices "respect user preferences", despite the argument against that fact. Some in this camp firmly believe that users never had the power to "set" fonts entirely anyway, but are resizing fonts always relative to the authors sheet, when present, and that any font resizing used by users can and should be honoring the author's sheet anyway. (This will be explained later). There are deviations, but these are the general philosphies for this camp.
These two camps basically have tried to articulate how one sizes fonts using style sheets, how user preferences are set, what they mean, and how one does not respect those preferences, if they even exist. Again, all that depeneds on what camp you are in. This is the battlefield where the warriors are aligned, ready with swords drawn to attack each other in endless quarrels about which practice is best and which is a violation of everything pure and holy! The battle line are drawn and the war rages on the battlefield night and day (the internet never sleeps, as you know).
The Dragon Awakens
In the midst of the carnage on the battlefield a sleeping Dragon has awakened. The Dragon is the W3c.org. Why are they a sleeping dragon awakened? Because they are the monster that has created the confusion, I feel, then gone to sleep on the issue, waiting for the vendors to support standards and explanations which in the case of font-sizing and acessibility, have ended up being very "half-baked" and added to the confusion, browser to browser and from version to version. The W3c is the great bloated Black Dragon of Shadow and Mist, belching out greasy smoke clouds over the battlefield of truth and shrouding useful "best practices" that would help both parties sort out the mess. By refusing to take a real stand, wisely, the dragon watches safely over the battle as it rages on. But maybe fainally they are attempting to clarify (with more "fog and mist").
At the w3c.org, on some pages of the xhtml 1.0 recomendation pages and CSS 1 and 3 pages, we see them clearly using examples showing how one may freely use a variety techniques to resize text, which includes changing fonts in paragraphs (what in most pages is essentially the default text thats supposed to be honored in the users settings), body text (which in many agents is equivalent to changing the "default" or starting text assigned by user settings, according to purists), and using both relative sizing techniques as well as so-called "absolute" (not keywords) sizing using pixels. Beside the fact that some browsers, like Internet Explorer, dont allow text resizing on elements assigned to pixel sizes, its a beast of a trick trying to sort out what the Dragon has to say (w3c) on font-sizing. Seems as though they have given us the ok on any practice, and thus the vendors as well, to pretty much sort out font sizing as we see fit.
(Is that the way it should be?) In the accessibility guidelines, they do recommend relative font-sizing, yet they have provided the full ability to defy that. This sort of "double-speak" means the dragon is the ancient evil in the battle, slowly waking and blowing steam in our faces, as we penetrate the dark cave of the www.w3c.org, digging through page after page looking for the truth (golden nuggets of truth!). Will the "dragon" awaken its gnomish "members" to help us with the hidden treasures within....help us brave the mist and shadows its obviously created concerning correct font-size application in web pages? Some quotes from the w3c pages Ive provided below to show some of the varied opinions on fonts and sizing:
W3C Quote:
"...a good design should look good without requiring the user to enlarge or reduce the text size." - (true, but what does "look good" refer to? Does this imply that authors are empowered to set an initial page or body or default font size in the page different from the browser or the user's font size, that any user may resize, but should not have to? Is "enlarging" setting a text size thats larger or making the authors size larger relative to it?....hmmmmm)
W3C Quote:
"[You may] set a base font-size for the document and use absolute size when defining the font size for a particular element within the document" - (This clearly implies that authors may set or are encouraged to set font sizes for a document before the user has even resized it, and that size might be different than what the user has set. Does that not defy the concept of honoring the the user's preferences?)
W3C Quote:
http://www.w3.org/TR/CSS-access
"CSS allows users to override author styles. This is very important to users who cannot perceive a page with the author's chosen fonts and color. CSS allows users to view documents with their own preferred fonts, colors, etc. by specifying them in a user style sheet." - (another interesting quote from the CSS Accessibility recomendation. Here they seem to say, yea, go ahead and change font sizes if you like, all users...but as above, seems like they dont recommend that. Also, seems to imply that user style sheets should be honored in this case over author sheets, but here, that appears to be done or is SUPPOSED to be done in a browser via the user style sheet! This is a critical concept I will bring up below in my arguments later. Still, its not very clear what the w3c see as correct font application in web pages by authors or users. It appears to be a subjective and still very open question.)
Font Size Chaos
With the dragon spreading black mist all around the battlefield, the knights play dirty and hurl threats, personal attacks, cyber-insults in an attempt to degrade their opponents into submission.....such is the passion of this debate at times. (who knows why?) But allot of chaos surrounds the who debate and battle on the ground. Very intelligent scientists dig up proven theories on why they never fool with changing font sizes in their sheets, and how foolish it is to do so. Then we have the vast galaxy of general PC-based website authors, using a completely different font foundary from Macs, various monitor resolutions, and at times, horrible old Internet Explorer version 5, and who claim, at times, that their font is even bigger when not set to default and argue the necessity of changing sizing further as authors of a web page. Warm and fuzzy Mac people proclaim themselves highly qualified font size Purists (despite the fact their pretty little Macs make up maybe 3% of market share world-wide) and claim their web experiences and font-settings are defiled in Safari browsers by those who change the font size in the body tag, and how foolish web authors are who tamper with fonts. The tiny minority of elderly who have figured out how to change text "size" in their browsers are content to enlarge the text in Internet Explorer, and see no loss of usabilty, and sit silently happy with their choice despite the differences site to site surfing gives to their carefully chosen font setting in their browsers. They claim to us text is readable enough, despite poor eyesite. Then we have the vast world market that has never touched or changed their so-called "default" font setting in their browser and who stare confused as to what the debate is about on web blogs across the web. Finally we have the educated and partially blind elite, who are savy enough to both read the w3c recommendations of "user style sheets", then read the documentation on how to build a user style sheet, and who in the end, figure out how to build a user style sheet (with something similar to the following, "body{font-size:large !important;}") and end up laughing at us all as they seem to have found the correct and only true means of setting a "preferred user font size", and are now seeing what they want to see as the Dragon had originally given them power to do.
What a MESS! Seems like the battlefield is not only filled with confusion, but now different types of warriors....the whole argument seem to now take a relative point-of-view, meaning: Font-sizes seem to be relative to each of our own unique experiences, and its almost imposible to set "standard" practices for font-sizing, when you really really look hard at the variety of experiences, both in terms of browsers, platforms, demographics of users, and preferred browsing preferences each of us has experienced. (The Dragon is truly the all-knowing and powerful, objective beast we should have expected in this war, and is looking over the battlefield from a safe distance in his cave, and seems to have crafted the whole war and debate, or at least has been smart enough to have left it alone! Maybe when we look at those subjective, stupid little cyber-soldiers in the font-size batttlefield, we realize the Dragon is pretty wise after all and seems to have some special purpose in this war.)
What I have tried to give you in the paragraph above is an example of the current confusion in the conflict that is the "Holy Font Size War". Different users and authors with different expectations and different levels of enlightenment, as regards controlling and viewing fonts and font sizes, seem to fill the field of battle. As Einstein has said, concerning the whole fabric of the universal order of things, and certainly now applying to the font-sizing war, is it's all relative!!! ....its relative to what your trying to achieve, what operating system you are using, what your point-of-view is regarding the medium and how you are using it, and what your need is as far as viwing fonts. Seems impossible from that stand-point to even pretend to know or "assume" you know what any user will prefer. That applies to both camps! Purists will take that concept and run with it saying that thats plenty of proof NOT to set or change basic page font settings using CSS. Purveyors will say that thats plenty of proof for any author to use whatever font-size solution in any page as they see fit, as one cannot presume to know how to plan or limit one's font-size designs. Both are right, yet both are wrong (...but one is very wrong) !
Clearly this confusion and trying to resolve it is the beginning of how I explain my point (and side) on the subject, though, the side I take in the war and how I have chosen to implement styles and font-sizing I will save till later. Are you still with me....good. Ok, I'll try and make my point as clearly as possible. Let's proceed.
A Tale of Two Webs
First, lets sort out the two main camps in the armies at war. The first takes a "purist" (not necessarily user-centered) point-of-view and believes that web authors on all counts should try to honor and respect browser "settings" such that web authors should tread cautiously in how they use style sheets. The other side, or what a purist might call the "Dark Side" (Purveyors), are those who believe, in general, that the Dragon (w3c) has given us the great gift and "design empowerment" using CSS to change any and all font sizes using author style sheets for any tag in any web page we design as we see fit. And that, despite what the purists say, can still do that and actually respect user preferences. The purists, or Knights of StyleLight, as I call them, have a noble argument. Basically, what they are saying is, that if a typical user goes into the area of their browser, like in Internet Explorer, under "View/Text-Size", and changes the font "size", as they call it, to whatever that size is "set" as, it should be respected by the author's style sheet in the authors website, such that if a user sets "large" or in some cases "larger", that "font size" is seen exactly as "sized" for that website, or any other website, for that matter, by any number of authors on the web. This translates to the fact, that if you set your "View/Text Size" feature to say "Larger", your web page design should show "larger" text no matter what the author has styled the text as, or as the purists say, a "large" font size. (Does this mean that "larger" is a large font size setting, in say pixels, in the browser or text that is "larger" relative to the authors's styled font-size settings?....more on that later) According to many purists, though, this applies to general page text, not 'secondary' textual page elements that refernce the page font sizing. And so, this is not to say you cant change say the header element (h1) text to enlarge bigger or smaller relative to the user's chosen size (which it does by default). Just dont use tricks or make changes that affect the basic font setting applied by the user, or the "default" font site of that page and violate the honored setting that the user has set.
Now the "dark camp", are seen as messy, "lazy", confused, and ignorant by the purists. They claim this group is incorrecty and freely modifying page text sizes such that those changes are messing with the user's "font preferences". If you change the body font-size or change basic page text bigger or smaller than anything other than the default text size for the page (as controlled by the user's browser "View" font "size" settings, many claim), you are violating the users supposed font-size settings, and forcing them to go in and have to change their browser font-size setting again, bigger or smaller, they argue. They also argue people with poor site are now forced to do even more radical font-sizing if your text is, say, too many steps smaller than the "default". That may be a valid claim.
Lets look at the dark side, or author-based, Purveyor camp. They are actually a minority in number compared to the purists, yet, most designer do evil styling as authors, so are a majority in my opinion. What this says is that most web designers, including some of the purists themselves, are doing things the WRONG way by default, as the purists will argue, and using font-sizing that either does not perfectly honor or respect the user's initial setting (in some form) or blatantly changes those settings. Thats bad if so many designers do that by default, not knowing that changing the BODY tag to say body{font-size:80%;}, is a huge no-no to a purist, who will come down hard on those designers. By setting a font size on the BODY tag different than the default, which is medium or 100%, or worse, use pixels or some "absolute" font seting, one is clearly practicing lazy or sloppy font-sizing, and should be spanked severly! (Thats again according to the Purists)
This gets me to my last part of this section, and that is how confused both camps are in what they define as critical font-sizing philosophy and user preferences. The purist camp is the worst. I ask the question: At what point are you truly respecting or disrespecting "user preferences"? And what are "user font preferences" anyway as regards font-sizing? Let me state by opinion of this and see if it makes sense to you. I will first give you an example:
h1 {font-size:150%;}
First lets look at the Purists and their confusion. Most purists will look at the rule above and say, hey, thats not a problem to most of us, as you have not changed the user's default font set by modifying font-size in the body tag or the major paragrph or copy text in the page (yet). Others will say, also, you are using relative sizing, so thats also ok. But I say, you ARE disrespecting user preferences, as 150% is not always the default relative font size for the heading tag in the browser you are using! For that tag to be exactly as I expect, it should not be changed from its size, relative to my font size setting, if no style sheet was applied. In other words, if I stripped out all styles from that web page and stuck a header h1 tag with text, how big is that text actually relative to the default font size for that browser and that default font?
The answer: Its all RELATIVE! It depends on the browser, and the operating system. Right? So how can one claim any single implementation of font-sizing is right, and another is wrong as regards the so-called "user preference" for font-sizes? How can one change even this innocent preference for this heading font size, claim it is or is not supporting user preferences, when we just dont know what that size or relative size should be?!? Netscape 7 actually sizes that default h1 tag a little differently that IE 6 when using the default font size and text for any given end-user in the browser (on PC's, that is...its all relative, remember?). But back to my point, and that is, first, if you are going to be "pure" about honoring settings in the users browser, why not go all the way and NOT use a style sheet at all, as thats the only way to truly send the user whats "set" in their browser, according to purist theory. Otherwise, its just not pure, right, and its certainly not
TRULY respecting user preferences, right? Second, how can we honor any user's preferences when its all relative? That h1 may be exactly what the user expects, or it may be "good enough" in terms of what they expect as far as relative font-size goes, or in some agents, like a PDA, actually the exact same "default" size that agent applies anyway to h1 header elements. But then again, Safari on a Mac may make it way too small. No one can stake a claim in this war, then as to whats right or wrong concerning "respecting user prefernces" as thats clearly a matter of degree, and now as everyone knows, as its all relative. What Im not saying is lets get black and white here and everything is light or dark? No....what Im saying is, lets clarify whats a user preference, before we stand behind a Mac Safari browser and whine about seeing text thats too small. Maybe that font setting is the default for another browser, or another user has set the same small text in their "preferences", wherever those may be, and it works for them. Does leaving one CSS selector void of font size styling, but applying font-sizing to others, broaden the usabilty or respect for preferences? In the ultimate purists sense, no! How can it? Using the purist's argument, how can any style that modifies ANY tag using styles respecting any so-called preference? It cannot and does not, unless one completely abandons style sheets, taken to the extreme.
And thats obviously not what the Dragon (w3c) intended for the peons below in their heated battles....the beast wants us to use styles. And like the mighty dragons twisted tail, using styles freely in any and all forms will and should have the possibility to twist, change, modify, and resize fonts and text at every possible degree or level of implementation in a website.
But then we come to the dark side....the authors who modify everything and change font-sizing without any responsibility or concern for how it affects end users. If I set a 9x font in the BODY tag or say a paragraph tag that is used widely through my site, allot of people will have a hard time seeing and reading my text, as will IE users, who cannot resize pixel text using the conventional browser preferences. Many browsers will also have problems resizing text that deviates so far from the user's default text. If a Netcsape user had stumbled on that page with really small "font settings", the text just might be tiny blurred lines at the bottom of their browser....completely unreadable !
body {font-size:9px;}
Since Internet Explorer, as of the time of this writing, is about 90% of market-share, thats a real problem! Authors who blindly modify fonts like this, and in large degrees or "steps" in the main body or paragraph text of the page, such that text is either unreadable, or worse, where useres cannot change its "size", such authors are clearly setting themselves up to create sites that actually fail completely and are virtually unreadable to a large group of people online. Thats the whole purist's dictum to them: DONT RESIZE MY FONTS! Without going into all the evil ways one may screw up fonts for users, it nevertheless must be reminded that authors are "empowered" by the dragon and w3c to make those changes nevertheless, so have the power and freedom to do those things as they like in any combination as they like, else it would never have been designed that way (there is a reason for everything, huh?). It turns out that there are legitimate reasons for ANY AND ALL applications of font sizing one may wish to apply using styles, and that there may be reasons for you or I to, say, set a 9px font in the body element of a style sheet at some point in our styling lifetime. As an example, Ive actually had a menu system that required such a tiny font, and for IE, such that that group should not modify that text. Purists, many of the less pure, that is, will say, sure, I have too, and I actually use Zeldmann's old article to back up my argument for using pixels, as it addresses font shifts among macs and pc's, etc. Again, so we come back my original argument on the current confusion.....and that is its proven that font-sizing is ALL RELATIVE to the user, the author, whats to be achieved, etc. And to try and articulate a rigid set of practices, is utterly foolish. To say, that authors can and should use relative font size units, or avoid say, "pt" units, except for print, etc, is valid and a good examples of "best practices", and so is ok to suggest and support, as the dragon has and continues to provide to us. But to say that, as a broad practice, one can or cannot or should not do this, or all sites will work great if all users will just do the same thing, is wrong, and considered "worst practices" in terms of CSS education and implementation. That does not allow authors to escape unscathed. Many of you continue to do some things that are really really bad for all people concerned, like assigning percentages or fractional em units (ex: 0.8em) assigned to a primary selector, like a div or paragraph, such that as those tags nest, they compound in size, multiplying themselves till they are tiny in size, etc. And the whole use of pixel-based fonts is problematic, with the clear advantage or percentage-based font sizing features now improving designs online. I can safely say, in general, I would not use the 9px body rule very often, if ever, in a sheet I would design, but to say it does not have its place, is helping to confuse authors, who are more and more scolded by purists when they use such rules, when they are legitimate means of assigning fonts, user preferences or not. The w3c has given us that ability for a reason....so it CAN be used if needed, and its not evil, as some are now saying! Does that not mean the Purveyors are not creating horrible font-sizing....yes they are creating bad CSS designs and fonts! Wheere the Purists win this debate is so many designers in the dark element of the css battle not fully
understanding whats best to use. Again, theer is no single best practice, but the dark camp suffers most from lack of education (look at my little article, Im sure some of you experts have noted many errors....we are human, right?). But in summary, using and choosing to use a 9 pixel font property for the body tag in most browsers means you have not only shifted the browsers font to something very difficult to read, but IE users will have some problems resizing, and thats a sure sign that the freedom brought by the Purveyors of CSS have clearly failed. Education in this group is lacking, big-time!
Who's to clear up all this confusion and lack of education? Clearly, the w3c is also confused on the whole font-size debate, as it says one thing then another, and there are confilcts in the literature, which are not good. And so the dragon is a symbol of entropic chaos as regards font-sizing....meaning.....the battlefield is getting more confused and chaotic as the world wide web advances and they dont seem top say there are any rules as to when you apply font-sizing, when you should not, and which rules you should or should not apply. The dragon has left that to us to decide. The browser vendors who parse our pages also seem to have agreed keeping the debate open and given us the power to say, modify the default font size for the page. That is changing the font size in the BODY tag, or in core page text, as expressed usually in paragraph or div tags (the areas the users preferences are most targeted towards in terms of changing text readability). So what binds the web author who changes font-sizes...what rules must he apply? Answer....nobody really know, except one general rule....use relative font values whenever possible, which includes font keywords(small, large, x-small), percentages (120%), or em values (.8em), where useful. Now, both camps that use those tricks in various forms run to the front lines and take broad slashes with broad swords at their opponents, as this is where the arguements get heated. Yet, this is the very place where the most confusion occurs, as noone really has defined their position clearly before they begin. This is where both forces face horrendous attrition rates, where some purists leave their position and come to the dark side when they suddennly realize they have to change the basic size of paragraph text to something other than 100%. And the Purveyors of Darkness are given "holy status" and come to the light side when they realize that changing body text sizes spell trouble in many browsers, and so leave it blank and abandon the challenge of resetting any basic font sizes in their web projects beyond relative sizing for secondary tags like headings, table cells, and forms. One may become a Purist that way, Im told. Both sides have broad understandings regarding font-sizing, and as has been mentioned, this adds to all the other layers of confusion when debates rage online between both sides. Someone calls me a Purist when I dont change the default font-size in my pages, but Purists call me a lazy ignorant piggish SOB when I change my text in my divs to a percentage of whatever the user's font size is.
All this leads me finally to my argument!
The Dark Knight Arises from his Stylish Crypt!
Thats right....know you know. Im a member of the Dark Knights, am a Purveyor of All Standard CSS Practices (as good and worthy) and Im a self-proclaimed five-star General in that army and never going back to the Light. In other words, I fundamentally believe, after much research and reading and testing and practice, that not only are
all font-sizing techniques legitimate means for web authors to set font sizes in pages, using relative sizes in most cases, and not only a very good pratice, but those techniques are handed to us by the Dragon himself as gifts of gold for us to use to empower and enrich our website solutions. And I firmly believe, at some point, everything handed down as tools by the w3c will find their place and use, despite "user preferences" and self-proclaimed restraints by web exerts against some CSS practices. All these tools will be needed and used and do and can serve and respect user preferences, if used with understanding and skill in the hands of a careful and senstive CSS expert! Am I that person and confident to argue the value of, say, using a point (pt) unit for a font-size in print sheets...no, not yet. But I honor and respect that tool and its purpose in designing style sheet s for those designers that need them, despite the fact they may not always work well with a given user or type of user online. Will I condemn an author who has set a pixel-based font smaller than my default for the body or text tag in their page, screwing up my default IE font or "resize" settings" in the process? NO, because that author may have a reason for that! Do I believe in correct font-sizing values and implementations....YES! Of course. And the tough question....do I believe in honoring or respecting user preferences? Of COURSE! Im a web designer, how can I NOT CARE? Thats what we all do. We design text and pages so they are accessible, usable, readable, and (hopefully) easy to comprehend. Why would anyone say Im NOT doing that? Its ridiculous. All pious purists seems to be telling us that Im a naughty boy for "soiling" their "font size preference" in their browser. How do I know what they are using, or how can I pretend to know, or design for such a setting, when another user may feel the oppposite....that I should have designed smaller text or larger text for their experience? But as I said, I do have values, and "font morals" and so I believe some practices with CSS are better for general design and text layout purposes than other, no doubt. I will be articulating dozens and dozens of those in future articles, dont worry. And Im sure I will be flat out WRONG on many points (Im human like you, remember?). But I do believe, for example, that, as long as one is implementing font sizes using using relative units (%,em,etc.) and other font-size "best practices", wherever I find them, and one is sensitively and intelligently considering ramifications of any font-size choice in their style sheets, one can and will be perfectly respecting user preferences, wherever those may lie! And saying that, I stab into the heart of my comrad! (keep reading if you are baffled by that last point!). To say I OR ANYONE ONLINE KNOWS HOW BEST TO TELL ANY AUTHOR HOW TO DESIGN FOR THEIR USER EXPERIENCE IN THEIR BROWSER WITH THEIR SETTINGS OR ANY OTHER USER'S EXPERIENCE IS WRONG!....and so I am the Dark Knight. Let Cascading Style Sheet and Font-Sizing chaos rule the online world forever! You are the author....design and size fonts as you see fit in your web pages! I will defend that right with my dark sword of chaos as long as the web exists!
Come to the Dark Side
I'll refrain from metaphors and colorful comparisons and get to the point.....the way font-sizing was designed by the w3c and the vendors and their browsers, its obvious that they have given us the power, as web authors designing author sheets, to change our sites (and the web world as we know it) and certainly, as they defined us, as web "authors", our sheets are to be respected and honored by the user above all other things in the browsers, until such time as the user does one of two things:
- The user "changes" their font size RELATIVE TO THE AUTHORS SHEET (not setting size) using, say, in IE6, the "View/Text Size" feature of the browser, which clearly does NOT set a font size. As IE shows (example: using the setting term, "Larger") in that feature, that is the area where the user DOES NOT SET font size but clearly enlarges or reduces the font size of the AUTHOR's style sheet (see below)!
- The user intentionally sets a "user style sheet" as defined by the w3c and the browsers as a means to give users "control" in manipulating, changing, resizing, or even setting font size such that it cascades over authors font sizing. This is the TRUE area for setting font sizes that are to be honored, and thats clearly controlled in the users camp, in the user style sheet area of the browser. In IE, this is found by going to "Tools/Internet Options/" General tab, and clicking "Accessibility".
(*note: there are other secondary accessibility settings for fonts in the browser, like "Ignore Font Sizes Specified in Web Pages" in the Accessibility section of IE, which is much like a quick shortcut to a user style sheet that is styled to completely override author sheets for font properties. This is next to the user style sheet area, interestingly enough, in IE. This has very little bearing on how author sheets are designed, though, or the confusion surrounding how authors and users interact with a styles web page design.)
Ok, so what does this all mean? This means that Purists who have argued for limitations using style sheet font-size properties are flat out wrong on the concept of changing font "sizing" in the "View/Text Size" area of say Internet Explorer, as those settings increase of decrease the web page text, which in the absence of a style sheet or style rule defaults to the users default font size, or in the presence of an author's sheet (most sites), is relative to the authors font size as set in their sheet.
The purpose of the w3c and CSS and the author's sheet is to empower the author to set the font size for their site any size they need, yet respecting users by not setting fonts too big or two small for readability purposes only, using fonts with similar foundaries or attributes or families such that they are represented correctly (or substituted correctly) on all agents, and using relative style units in as many cases as possible such that as text resizing occurs in these areas, those sizes are flexible and size relative to those changes. Thats what is termed "respecting user preferences". One is respecting the ability to see large or smaller text when the user selects that in the "View/Text Size" area of IE (similar to Netscape and Mozilla), for example. Yes, that is honoring those preferences, because that feature does not set font sizes, but sets large and smaller text relative to whatever the current font size is. What the browser does, as regards fonts, is provide a medium for the author to display styled pages as they see fit. Second, all user preferences outside the user style sheet concept, modify those settings relative to the authors sheet...not the other way around. So, there actually is no user preference to respect until a "user style sheet" is attached by that user that has rules to cascade over all author sheets such that a new and permanent and "set" font is established by the user (see below). Only by making fonts bigger or smaller relative to the author's sheet occurs. Its the still the author's responsibility to make sure those "styles" are readable and usable to the end-user, such that the information is available, and "resizeable", as far as fonts go and other pre-determined sizeable css sections of the site.
But most importantly from all this, its not the purpose of ANY AUTHOR to second-guess or pretend to assume they know anything about a user's preference, especially with the wide array of settings, user types, and agents out there. Nor is it the author's responsibility to address any font-styling that supports any user prefernce, beyond the ability for the user to resize those fonts in some fashion, period. To say, its better design and more sensitive and accessible web design to design font-sizes so they have the best chance as possible to appear sized "as if the default font size for the user's agent was not overwritten and honored", a noble idea, but actually quite silly and makes no sense in light of the fact that thats done using a user style sheet. If author style sheets and author font styles were meant to be designed that way so that the user requires some special predetermined font-setting in all author sheets so their "settings" are "respected", then the author-user style sheet system as outlined by the w3c.org would have been designed that way. Why should any author "cripple" his style design for something he has almost no control over or influence over. The user style sheet and author style sheet system handles that interaction. Until that user sheet is built and applied, the author's sheet and font-sizing MUST be honored over the browsers or any user defaults...not the other way around. The user has complete control to change that if they wish. The resizing feature does not do that, though. The user sheet is the final layer that cascades over author sheets and allows final and full "respect of user settings" by the end-user, if so desired. The authors sheet takes precedence over all settings until then and all user settings up until that point apply changes RELATIVE TO THE AUTHOR's SHEET until the user elects to override that sheet via the accessibility area of most browsers where a user style sheet may be built and applied, seting true user font-size preferences.
Now lets take the two extremes away from this conecept. First, if Im a user, how do I "set" a font size and have it be honored? Am I "setting" a font-size, enlarging or modifying the authors, or overriding that setting? Or does the system work such that the author is supposed to design some magic css code that sends me styles but doesnt apply them for some things when I change something in my browser? I dont think that makes much sense, does it? Does the user have settings, absolutely. Does the user have control over font sizes, absolutely. Is there font-sizing settinsg to be "
respected"? Absolutely. And that last one is totally contrlled by the user and has no power or influence or second-guessing influence over authors or how they designt their sheets whatsoever.....and thats done using a "USER STYLE SHEET"!
REVELATION : HAVE YOU EVER BUILT A USER STYLE SHEET IF YOU WANT YOUR PRECIOUS FONT SIZES RESPECTED?
Thats my whole point to the purists....THE USER CONTROLS FONTS SIZING VIA A "USER STYLE SHEET" PERIOD....not the author abandoning font size features the w3c and browser has empowered them to use to style and design their web pages with. Some Mac Safari users complain that not only does that not apply to their browser in terms of "setting font sizes", but that because of the differences in how PC's and Macs size fonts in the first place, such settings are creating even worse effects on their machines. Again, I ask....have you bothered to look at the specs for Safari and looked into creating a user style sheet, if a "set in stone" font size is so critical for your viewing pleasure?
The w3c, again, has clearly outlined that power in nearly all agents and for all kinds of users, such that even applying the new "!important" override feature in user styles sheet rules, one may safely overpower any rule in an authors sheet, such that a font-size is truly set, not only for the current web page, but ALL WEB SITES ON THE WEB. Now thats empowerment! And thats the way the style sheet system was designed!
The Mystery of the "View/Font Size" Feature Revealed!
Now the some of the Purists on the batttlefield come on with their little debates about how the percentages of users who actually use the "View/Text Size" in the case of IE are very small, or the "Increase/Decrease Text Size" feature in say Firefox is widely used so should be respected in your style choices. Others will also say that noone knows the "user style sheet" even exists, and its more reason to try and design more accessible text based on the text resizing feature in "View/Text Size" in IE or "VIEW/INCREASE TEXT SIZE" in Firefox. Just because someone knows or does not know something can be used as an argument for both camps. But in the end, so what? What does it matter if my audience is not fully educated on that property of their browser, or that those that are and who have changed font sizes to be bigger, on average, are not being honored. Again, its all relative, right? Whats too big for some will be too small for others. You or I as web authors cannot begin to make assumptions for all users. All we can do is design for our needs, and understand our chosen audience. If thats the world, which is usually is, then you have to decide if you want to affect those with or without changes to any font-size setting in their browser, which ones, and what you want to use as far as font-sizes. How important is your choice of font size for any given tag in your styled web page to the percentages of those who have say, enlarged that text one "step" higher using the "View" area of their browser, for example? Thats up to you to decide....not some purist! That should and does not limit us to any single font-sizing choice....it just influences us. That level of influence is perfectly aligned to your point of view and your website design, period!
Read what the "VIEW" section of those font "sizing" areas actually say! Those areas allow you to make text larger or smaller, not set text size. If they were designed to set text size, they would do that, and on making text "larger", as in IE6, one would instead see something like a numerical value, like "Time Roman 20px". etc. Those areas do just what they say....make the text larger or smaller! On the other hand, if a user is so desperate to have, say, the basic default text for every page be the same, build yourself a Useer Style Sheet and do that in the body tag of that sheet! Every page you visit will be overriden by your sheet and though all the other styles are honored in the author's sheet, the core font size will be what you want. Sorry, but increasing or decreasing font-size in the "View" area will always satisfy what it is meant to do....NOT set a font size but resize font text relative to the authors sheet, honoring and respecting the authors intended size, but making it bigger or smaller as needed relative to teh author's intended initial page size, which is what the author intended anyway! Its not our job as web authors to try and predict what resizing a user will perform, but to simply make sure, if thats a featre we want to allow a user, that doing so keeps them within th intended "range of sizes" we intend for our user base. That, again, does NOT have to be within the range that anyone tells you it has to be. It can be gigantic or teeny tiny and unreadable, if you have a legitimate reason for that. (Most designers design relative unit-based text to allow text to resize within the 1.2 step range set in most browsers, even though the default or starting page font may be changed by the author to begin with in any number of tags in the author's page and style sheet).
In summary, seems to me that the w3c and the browser manufacturers agree on this one, that for the most part, if a user wants his "preferred font size" honored, that there is no font to honor really, and that the authors design is buil by default to "override" over the browser's default values, unless the user overrides the authors values using a user style sheet. Clearly the power is with the user, and using user styles sheets if they want that power. This is not to say that certain areas of some browsers have features like aliasing text, zooming the whole page design (Opera), or "ignoring" font sizes, styles and colors, as found in IE 6 and their "Accessibility "preferences section of the browser. Those are spelled out clearly and override the authors sheet by design if chosen. Again, no assumed "honoring" by the authors code need take place. The user has the control, if they choose. Using the user stylesheet, one may also use relative sizing that interfaces with an author's sheet or override it such that a set font size is set for various elements if required by the user.
This fact, again, opens up the possibility that the "other" means of "setting" font-size, in the "View/Text-Size area of IE, for example, is really NOT about setting a font size, but really, CHANGING font size relative to the authors sheet. I realize each agent is different in that interprettion, like "zoom" in Opera, but all exibit that feature such that it does NOT "set" a preferred user's font size, but simply enlarges or decreases whats in the authors pages already. This area and the user style sheet areas of the browser allows different levels of "user control", and its obvious to me that the View/Text Size area in IE, for example, is NOT a means for users to control text size, but modify it relative to the author's intended size. So, those members arguing that we all must set BODY font-size to 100%, or leave it alone, or design pages that dont change basic font size, is incorrect in those arguments. Internet Explorer (not the best browser), states the following in their browser's help section concerning the View/Text Size feature, supporting my argument by stating this:
"To display Web page text larger or smaller:
On the View menu, point to Text Size, and then click the size you want." - IE Help
*Ie names these "Large" and "Small" but thats NOT what they do....they clearly dont set font size, so there seems to be no way of "dishonoring" how font-size occurs from an authored style sheet point of view! This brings me to the actual point of this text "resizing" feature of most browsers....and that is, one does not, nor are we expected as authors, to make assumptions about what is usable or not by our audience 100% of the time, as there are so many ways that simple interactions between the author's web designs and the "enlarging" ability of the "View" font-resize feature may be applied. Everyone that argues that "user style sheets" are "userless" or rather, useless, have an argument, as most people don't know what that feature does in their browsers. That does NOT free us though, to disrespect the system as its designed by the vendors or the w3c. Its built that way for a reason. BUT....again, if I had a severly handicapped friend that consistently was unable to surf the web enlarging the text using this view feature, then one could simple build and assign a "user style sheet" for them and depending on what styles are applied in that sheet, deliver ALL FONT SIZES essentially the same for them, regardless of author sheets and font-sizes set there. Thats the power of the user at work! Why should authors worry about this "View" feature when clearly the power is with the user, if required, to override all font styling if they so wish. Why should any web author, knowing that, give up their precious web designs and text layouts such that users who cannot comprehend those features, are given some "half-baked" design, with giant fonts, or some semi-designed web page layout solution, which may solve one groups browser font issues, but for other risks creating new ones. I for one would not appreciate those designs who use such "crippled" web designs with giant fonts shown to me. Neither would the vast majority that never touch this "View" feature in their browsers? Nor would I appreciate a pixelated font sent to my IE (by Safari Mac designers), which for some people with eye problems, could potentially be difficult to size in IE on my PC. My argument is users are doing fine enlarging text as they do now, the rest of us are fine who dont, and those who occasionally do so are compelely (and with great intuitiveness) completely understanding of what those features do, despite what some of the highly educated font-size "purist web designers" seem to argue is vast irritation and confusion. Take the case below:
FONT-SIZING AT WORK : An Example
I watched my wife's father the other night surf the web....he has problems on his huge monitor seeing text at times. I watched how he had learned to improve that issue using the "View/Text Size" feature in IE 6 on his XP machine. I suggested he consider reducing his monitor resolution, but he preferred to "enlarge" the text via the browser and the "Larger" setting in that area. Did he expect that to be "honored" as he surfed the web, and was he constantly changing that? No! Clearly, as he surfed the web, some pages used relative fonts, so his enlarged text just was as it was intended....not a "set font size" meant to override the author's setting, or to be honored as such, but an expression of that size, resized relative to it. He never changed it, and he seemed content that its ability to enlarge most of the text on the web was expected and good enough for him in assisting him with readability of web pages he regularly surfed. That seemed to be good enough, even though not every page expressed his "larger" font the same. Those sites that set the text smaller, in the body, then using relative em's for the rest of the page started out smaller, but still were bigger. Those pages that started out really small, he never touched, but could easily have changed his font size if he needed, but the "large" font setting in IE he used seem to be enough and I never see him change that. That feature clearly did not allow him to "set" a font size, but "enlarge" the font size relative to the authors intended font size. I may go back to him, and if his vision is worse, set him up with a user style sheet with say, body{font-size:120% !important} and that should solve his problem permanently for most pages (I believe it might destroy allot of layouts and page design as well! Thats the correct trade off....honor the author honor the user, or something in the middle, which is enlarging author fonts as possible, not "setting them"). This not only makes sense, but most users are content to resize (not reset) the authors chosen size as they need, and in most cases, never have to change it, though no two fonts are EVER the same size using this feature! And why should they ever be. If Im a purist and set my fonts to match exactly what the user has set in the View area...so what! Did my father-in-law need that or expect that to occur? No! Font sizes change like leaves in the wind and always will! But if I ever needed that feature, even at the expense of wrecking a web page design, I could do it using a user style sheet. This enlarge-author-font-feature and the power of the user style sheet shows the elegant beauty of the system that the "Dragon" (w3c.org) has given us, and shows, that the dragon isnt so evil after all. And we need not be at war, and are free, truly free, to use style as we see fit! So, authors! authors! authors!, of all persuassion....style away, and resize fonts as you see fit. Just know what they do in which browsers and for which platforms, etc. Know your audience, and design wisely and as the Dark Knight of Style says, you will go far!
One Final Word...
I cannot end this discussion without trying to help people make sense of font sizes in their browser as they test using styles sheets.
First of all, the whole point of talking about the chaos and seemingly open protocols and standards for how the w3c recommends applying font-sizing and the techniques involved is to show that even though the Dragon and the Purveyors are right, and the Purists are wrong.....there is really is no right or wrong with style sheet techniques, especially when applied to trying to respect user preferences, standards, or even best practices with style sheet solutions. Why? Because again, its all relative. Sure, if I avoid setting by BODy tag to any font size for the page, use relative font sizing for headers, and avoid pixels and absolute sizing techniques, I have the best chance of all my users seeing their browsers "default" text size matching what it would look like with no author style sheet applied to main page text. Big deal! Its all relative....I may need to set pixels for menu text I dont want stretching till it breaks a special layout, or I may want smaller page text to start with, that users can resize or override with user sheets anyway, or I may want point units for my text for a print display feature, or I may be applying a special page for a special mobile device that only recognizes a specific pixel size value for text, or I may want to use a special pixel value for a specific font that is required at a certain value for a Mac intranet project Im building. It all depends. To make things worse, fonts on Mac and PC could use different font foundaries which look differently, or my XP users could have enabled the anti-aliasing of web text, or by audience has monitor resolutions primarily set at 640 x 480 pixels, so I dont want my fonts too large as they will be huge! Its all relative. Combined with the incremental increases and decreases of text in the "View" area, resizing of text by OS, resolution setting differences in monitors, one soon realizes that absolute text sizes is an absolute impossibility!. And that applies to setting font sizes for any assumed setting by an end-user. Why speculate that one group needs one text size, and another needs another. Even worse, why address that by relying on the default browser font size, when that can be resize larger or smaller anyway by the user, or even cascaded over by a user sheet. Its you duty as a designer to design pages AS YOU SEE FIT, and not confined by some imaginary assumption on what users need or require, as far as font sizes go. The w3c has not attempted to resolve the chaos, nor oultine the possibilities, nor in their accessibility recommendation, completely outline all the scenarios that might apply or consequences to using one font size or system or CSS application over another in a thorough fashion, as that power is with you, the author! Explore it and do not fear the shadows.....use the font sizes you require!
As I have stated above, I see value in using relative font size units, and there are plenty of advocates on both sides of the batttlefield advocating that. Do a search on Google for font-size and CSS and you can read all the options. Just dont get suckered into the argument that by changing this or that in setting your font-styles for your web project, you are positively affecting some group or another.
FONT SIZE DECISIONS SHOULD ALWAYS BE BASED ON WHAT YOU NEED TO ACCOMPLISH!!!
...not on some rigid guideline, practice, or what some expert online tells you, including me! And Im not scared to say that, as the Dark Knight. I give you freedom to style your fonts any size you need or like, and its ok. Let all the people that get irritated with you go build a user style sheet to fix that problem....CSS was built for you and I, to give us possibilities, not what someone on the web that has proclaimed it as his area of expertise. Styles can be anything, and so can font sizes. Feel the freedom. As a proud American (sorry, I'm throwing in the demographic argument now), I invite that feeling....always! But especially with style sheets!
So, always base your decision on the type of user group you are serving, and what the demands are for your web design project from the clients point of view.....always. Dont throw the baby out with the bath water just because someone states you are a fool for using this font size or another. As many arguments he can make you should be able to make for it. Just remind them, if the font size issue comes up, that if they want to see your sheet at a certain preferred size, to build themselves a "user style sheet". Thats my official recommendation!
Finally Testing Recommendation
Its my fundamental belief, that you cannot begin testing or comparing font sizes, either between agents or between a Mac and PC, until you make sure you understand whats setup for all platforms each step of the way and how they affect font size. Until you go through these steps, using the default sizes of the browsers first, only then can you begin testing using your chosen font size, no matter what that is (note: for all these recommendations, go online and find the information/solution that applies to you and undertsand fully whats going on before you move onto the next one):
- First check monitor resolution. That can be 600 x 800, 1024 x 768, etc, and usually expressed in pixels. You need to make sure multiple machines you may be testing on with multiple monitors all have the same setting, or reflect in the hardware setting, the same relative sizes for the screen size provided. If you are testing a web page on different boxes, just make sure you have some graphic in the web page as a visual reference point to determine what size is actually being delivered. You cant test font size till thats determined. Example, I tested a web page with just text on a Mac and Windows Platform and the Mac looked bigger, until I realized the PC resolution was 1024 x 768 and the Mac monitor was showing that same page within a 600 x 800 resolution. When I put a PC next to the MC with the same size monitor from inventory in our testing lab, and the same resolution, I saw clearly that the font sizes were the same in the pages and my font size hacks were working correctly as intended. So, dont let monitor resolution side track you.
- Next check your page on different operating systems. Macs and PC's as you may know are different. The browsers in each OS is also different as are the font foundaries, and other font characteristics. Dont let all that confuse you. First, go online and read about font differences between Macs and PCS (using platforms with the same monitor settings, as above). Test without style sheet font sizes first using default fonts for various families of fonts and note differences. PC's are supposed to increase font sizes 33% over Macs in some cases, and this is also tied to the 96ppi and 72 ppi difference between PC and Macs. This is further confused by styles that use say points versus pixels. But first test the raw fonts themselves on the same monitor resolution, then test using pixel-based font sizes in styles sheets. Then finally, try different fonts. And then, try relative fonts. Try keywords as well, as those are mapped to values in the browser's font size structure, which again may be different for the same browser on different platforms.
- Next explore browser "version" differences. (Not that IE for Mac and IE for PC are different in this issue). IE 6 with a valid xhtml doctype will be different than IE 5 in terms of font sizes when one changes the default font-size in the "View" area to anything other than the default. So dont let that sidetrack you. If you have tested the first two bulletes above and understand what font-sizes and hacks to apply to remedy that, then on PC for IE, you may have to apply the "Tantek hack" or other hack to fix this bug for fonts in IE for PC.
- Next, after you have understood the platform/OS differences, and IE and other browser version differences, go in and look at each of the browsers and test each in detail using your chosen font family and size, as desired. Are they the same relative size among browsers? Is the difference readable, sizeable using teh view area, and do you want it resizeable? Does your target audience use one of these browsers more than the others (ie Apple Safari), and if so, do you need to set a special font size for them?
You might have to go back and set up a "style sheet free page" and test in your primary target browser for that audience, noting differences between others, then change the font size in the body tag for that user group with your font-family and test again. Note how that influences the default font size in most browsers. Then go in and use the resizing or "View" feature of the browser and understand how that affects the size of fonts bigger and smaller for your target group.
- Once you have your core font size tested and working in all the above scenarios, check how fonts size in nested tags with text inside other tags. There are bugs with how IE (5) and others in how they apply "em" values to nested text tags (say divs) when the default text is changed either by the author in say the body tag or by the user enlarging or decresing the authors font using the "View" feature. Try percentages, then try font "keywordss", like small, and x-small. Ive seen great success with keywords, even though how they "map" to various fonts on various browsers and OS's is highly variable. Its still a "relatively consistent" alternative that guarantees your page fonts will map to specific sizes in the browser, will map to usable and readable pixel screen version for various families of screen fonts, are resizeable and correctly resize using the "View" area of most browsers, and finally cannot be resized past a certain point in terms of the default font size for your web page. All pluses and a very good alternative to your other font units.
- Finally, try making your text bigger and smaller in your styles for various browsers, then resize those using the view feature of your browser. Are they still readable? Now go in and change some of the relative size values for the ":secndary" tags in your xhtml tag style set. Headings, who's size is relative to the default font size you set in say, your body tag, is a good test to run. Review how this performs in all sizes for that font family you are using. It may be that if this is too large a relative size, and extends too far beyond its default value, you may see text that is either way to large at the largest font resizing value, or unreadable and too small. It may also break your designs. Try this relative font sizing and testing with the structure of your site as well, or parts of that structure, such that setting say an "em" value to a column or div allows it to expand as the text is made larger or smaller.Test this on various monitor resolutions as well, now, as making columns stretch on smaller or larger resolutions may mean some problems for certain viewers. You may want to "nail" down some text to absolute sizes, like pixels, to make sure your target audience does not get jumbled layouts. Again, making sure the text (and the layouts that hold that text) is readable and acessible for both your target audience and all secondary audiences as well, is your main goal with all these tests! You cant make everyone happy, though. And you often will find you cannot, with CSS. And again, no hard and fast rules here, as it all depends on you the author and what your web design is trying to achieve! It's all relative, remember!
If these tests perform well for your target groups, yet work well on all lesser agents and platforms, you have done well my friend and are designing very usable websites using CSS and have experienced the full freedoma nd flexibility of Cascading Style Sheet web design without the restraints or restrictions that some may require of you in using CSS. Your audience is your audience and your design and font-sizes are your own, and have a purpose thats only relative to you and what you are trying to accomplish. If that means changing the body tag's font size, or using absolute font sizes, so be it! Happy designing!
-"The Dark Knight of Style", Mitchell Stokely, USA
