Posts in the CSS Category

I have in my hands a physical copy of the second edition of Cascading Style Sheets: The Definitive Guide, bound with a RepKover lay-flat spine and everything.  So I figure it should be shipping out to folks within the next week or two.  If you’ve pre-ordered, there ought not be long to wait!  (And if you haven’t, then what are you waiting for?)

As I mentioned yesterday, the ‘diagnosis’ favelets I used during my SXSW04i presentation generated a lot of comment, so I now have the underlying style sheets on a “Favelets” page in my “Tools” section.  For those of you who know how favelets work, just grab any or all of the style sheets you want and go for it.  For those who need some assistance, I wrote a “Favelet Creator.”  You plug in the URL of a style sheet you want to have applied to whatever page you’re viewing and the name of the favelet as you want it to appear in your toolbar.  Then you drag the resulting link into your favorites toolbar.

All this really does is create a javascript: link that, when invoked, will dynamically write a link element into the head of whatever document you’re viewing.  That link points to a style sheet, and so the styles are applied.  As an example, you could point it to a style sheet that sets borders for tables and table cells.  When you click on the favelet, all of the tables and table cells in the currently-viewed page become visible.  Figuring out exactly how a table-based page is laid out thus becomes a snap.

So if you don’t like the styles I created, you can write your own (or modify the ones I provided) and create your own diagnostic style sheets.  The favelet creator should make it even simpler.  Either way, I hope these will be helpful.

• Return of the Fish was published on Friday, March 19th, 2004.
• It was assigned to the Books, CSS, Tools, and Writing categories.
• There have been no replies.

Having taken some time to decompress, play with Carolyn, and generally recover from the trip, I present some thoughts, impressions, and memories gathered at SXSW04 Interactive.

• There’s still a lot of interest in social networking, and as much interest in doing it right.  The attention XFN gathered (see D. Keith Robinson’s comments about hallway buzz) indicates that people want to make assertions about their social connections, and that they aren’t satisfied with the currently popular mechanisms for doing so.
• The interest in favelets/bookmarklets caught me totally off-guard; apparently, of everything I said in my ten minutes, the use of favelets for diagnostic purposes was the things that fascinated people the most.  It’s easy to forget that such a simple thing can be unknown.  I’ll be posting some of the ones I used in the near future, but in the meantime, you can’t go wrong with Tantek‘s favelets.com or Jesse Ruderman‘s Bookmarklets (“ancestors” is flat-out brilliant).
• Once I managed to pry him away from his fans , I chatted with Scott Andrew about the latest goings-on in the CSS world.  I mentioned that Dave Shea and Doug Bowman are (deservedly) getting a lot of invitations to speak these days to cool events in other countries with travel expenses paid and everything.  “That’s the rock star treatment,” said Scott.  I groused a bit that I didn’t get that kind of treatment.  He looked at me for a second and said, “Yeah, but you know, you’re the guy who’s been around forever and inspired all the new up-and-comers.  You always get credited as an influence but you never get to the top of the charts.  You’re basically the Neil Young of CSS.”
• I got a picture of Photo Matt as he was taking this picture.  This fact amuses me completely out of proportion to its actual significance.
• The ratio of Mac laptops to everything else at the conference was about 10:1, maybe higher.  I say “everything else” because I know of at least a couple of Intel-based laptops that were running Linux, not Windows.  Brian Alvey came up with the idea of inventing a glowing Apple-logo sticker that non-Mac users could stick on their laptop lids in order to blend in.
• While standing outside the Webmonkey party, I saw a guy ride up on a bike with what looked like a heavy metal pole over his shoulder.  He pulled to a stop, dismounted, and put his burden down while he locked up the bike.  I was completely floored when I realized that he’d brought his own bar stool.  Now that’s a self-sufficient man.
• I ended up at the Iron Cactus two nights in a row, and both nights I found myself shaking my head over the sign outside advertising low-carb margaritas. That’s right, folks, they take out the alcohol and pass the inflated profit margin on to you!  It’s kind of a brilliant sales tactic, really, and I applaud them for overcharging customers in an honest yet sneaky way.

Frankly, the whole low-carb mania is starting to seriously tick me off, as it did Nick Bradbury; and yes, I know people for whom it’s worked.  A cousin dropped 70 pounds and halved his cholesterol on Atkins, and my father has been very happy with the related South Beach Diet.  That’s no excuse for the herd mentality I keep encountering.  For example, Schlotzsky’s, whose sandwiches I love, now has low-carb options.  How’s that work?  They take away the distinctive sourdough bread (to use their own marketing phrase) and put the sandwich contents on a bed of lettuce.  Um, isn’t that just a drastically overloaded salad?

• SXSW04i Wrap-Up was published on Thursday, March 18th, 2004.
• It was assigned to the CSS and Speaking categories.
• There have been no replies.

As a followup to yesterday’s entry, I thought I’d share some details on what will be in Cascading Style Sheets: The Definitive Guide, 2nd Edition (now available for pre-order!).  Here’s the Table of Contents, or at least a core subset of it:

1. CSS and Documents
2. Selectors
3. Structure and the Cascade
4. Values and Units
5. Fonts
6. Text Properties
7. Basic Visual Formatting
8. Padding, Borders, and Margins
9. Colors and Backgrounds
10. Floating and Positioning
11. Table Layout
12. Lists and Generated Content
13. User Interface Styles
14. Non-Screen Media

1. Property Reference
2. Selector, Pseudo-Class, and Pseudo-Element Reference
3. Sample HTML 4 Style Sheet

Owners of the first edition will notice that the chapters have been rearranged a bit.  Thanks to the expansion of selectors in CSS2, it made sense to rearrange things so that they got their own chapter (which you can read in beta form from O’Reilly), and the parts about how CSS relates to document structure were folded into the chapter about specificity and the cascade.  The consolidation of floating and positioning into one chapter really helped cut down on redundancies, although that is the longest and most-enfigured chapter in the book.  (In second place, “Basic Visual Formatting.”)  “Table Layout” talks about how tables are laid out and styled, not how to do layout with tables.  The last two chapters are basically overviews with some detail, since user interface styles are almost certainly going to change radically in CSS3 and non-screen media support is limited or largely theoretical at this stage.  There’s still enough detail to satisfy, I think.

As for the other chapters, they’re largely the same as in the first edition in terms of topical coverage.  They’ve just been updated and expanded to match what’s in CSS2.1.  As an example, “Text Properties” covers everything that it did in the first edition, now updated for 2004; plus it adds information on text-shadow, direction, and unicode-bidi.

The technical reviewers for the second edition were Tantek Çelik and Ian Hickson, who were just as tough and thorough as I’d hoped.  Ian’s one of the people who pounded the inline layout model into my head until I got it when writing the first edition, actually.  I got similar treatment from both reviewers over the interaction of generated content with non-generated elements this time around, not to mention when I tried to figure out the value syntax for text-decoration.  It used to be simple, but oh no… they had to go makin’ it all fancy.

Anyway, I hope that will give some idea of what lies ahead for those of you who do me the honor of purchasing the book.

I occasionally toy with the idea of setting up a Cafépress store with CSS-related merchandise.  If anyone out there has bought stuff (particularly T-shirts and other articles of clothing) from Cafépress, kindly let me know what you thought of it in terms of quality and durability.

• Fishing For Style was published on Tuesday, March 9th, 2004.
• It was assigned to the Books, CSS, and Writing categories.
• There have been no replies.

This past weekend, the folks at O’Reilly and I wrapped up the final edits and adjustments to Cascading Style Sheets: The Definitive Guide, 2nd Edition.  The in-stock date is near the end of this month, so it ought to be physically sitting on shelves by the beginning of April, maybe sooner.  The page count listed on the O’Reilly page (as I write this) is an early estimate and too high; the actual page count will be closer to 550 pages.  There are a few reasons for this drop in pages:

• The support charts have been dropped.  When the first edition came out, it made sense to include that kind of information in an appendix, so we did.  As I recently wrote on www-style, the world is much different now, and the day of nifty support charts may well have passed.  In the CSS realm, anyway.  To even present a simple yes/no support chart for CSS2 would have been a dozen pages long, and a nuanced chart with notes would easily have run five times that long.  I still have notes and warnings about particularly egregious problems sprinkled through the text, though.
• The “CSS In Action” and “Look Ahead” chapters were also dropped.  There is plenty information available these days on how to actually use CSS, so we decided not to be redundant.  As for looking ahead, even a high-level overview of where CSS3 is headed could be a hundred pages long, and out of date the minute we printed it.  Better to wait and see where things end up than make a lot of ill-informed guesses.
• By rearranging the way information was presented, I was able to cut a lot of redundancies that bedeviled the first edition.  I also cut out some material that seemed important back in 1999, but has long since become irrelevant (like notes about what IE3 does or doesn’t do).
• The figure count has been scaled back.  There are still a few hundred figures throughout the book, but I went to some effort to combine several points into a single figure when I could, and not illustrate every little point I made.  You really only need to see so many examples of “boldface text,” you know?
• The text doesn’t spend time on things that were in CSS2 but aren’t in CSS2.1, and that nobody will likely ever support.  This means that some paged-media properties like marks weren’t described, and I didn’t waste time on the CSS2 marker-styling features since they will almost certainly die out and be replaced by a different approach in CSS3.  I did cover properties like font-size-adjust and text-shadow, but not in major detail.

So the second edition is an update of about 380 pages of the first edition, once you subtract out the stuff that was cut.  Every chapter of the first edition was reviewed and, in most cases, significantly overhauled even if it wasn’t expanded (for example, the Fonts chapter didn’t gain a lot, but it was still reworked to reduce the number of figures needed and to clarify some points).  There are four all-new chapters, five chapters with significant additions or revisions, and five more that were lightly to moderately revised.  So it’s practically a whole new book.

That’s even more true of the book I have coming out in mid- to late April from New Riders: More Eric Meyer on CSS, a sequel to Eric Meyer on CSS.  And when I say “sequel,” I really mean it: this is a collection of ten entirely new projects, so it is not a new edition of the older book.  You can own one without the other, although of course you should buy both!  Baby needs a new pair of shoes, after all.  (Okay, that’s a lie; she’s too little to be wearing shoes.  But you know what I mean.)  I’ll have more details as they become available.

• Making Book was published on Monday, March 8th, 2004.
• It was assigned to the Books, CSS, and Writing categories.
• There has been one reply.

I freely admit that I’m kind of a zealot when it comes to words and text—heck, this quiz classified me as a “grammar fuhrer.” (Note: the rating of “M” for the linked quiz may be too weak; “A” might be better.  The results you get may well offend you terribly.  Consider yourself warned.)  I love to read, and wish I could do it more often.  I enjoy writing, for that matter, because there’s a sublime joy in crafting the right sequence of words and creating the right tone as a result.  (Speaking of writing, I’ll have some things to say about that tomorrow.)  In what I regard as an exceedingly rare moment of lucidity on Rush Limbaugh’s part—perhaps he’d been off the drugs for a few days—he once said, “Words mean things.”  I absolutely agree, and not just in the narrow sense.  To me, the right words mean something far more than a string of communicative markers.  They create a meaning that is much greater than the sum of the parts.  The right words at the right time can literally change a life.

I say all that to frame what I’m going to say next.  The Man In Blue recently posted some thoughts on the use of text in Web design—or, more precisely, his opposition to its use for branding purposes.

By utilising run-of-the-mill text to render your identity, it makes it very hard to distinguish yourself from anyone else  delivering a message.  Granted, usage of plain text can itself be an identity, but I think that  Eric Meyer has pretty much cornered the market on that one.

I’m not sure if that’s meant as a compliment or a criticism, but it actually doesn’t matter.  I took it as a compliment, and I think the things I said above help explain why.  Still, I’m going to speak in some defense of using “plain” text for design.  Not a full defense, because Mr. Blue (a.k.a. Cameron Adams) is quite correct: branding is important, and visual identity is an important part of that branding.  As he says, Coca-Cola spends millions every year on branding.  I’m sure Pepsi does the same, as do a great many other companies.  As they should.  He goes on to say:

Although it might pain the purists, sometimes only an image will do.

He couldn’t be more right.  It makes a lot more sense for me to share a picture of Carolyn smiling at her mommy than it does to try to describe the same thing.    If I were in charge of Amazon.com or Wired or some other branded organization, there’s no way I’d replace the logo with plain text.  Besides, an img with alt text is as accessible as plain text—a little longer to download, but not enough to make a significant difference.

For me, I stick to text because I’m not a visual designer.  I don’t have a CD full of fonts that I can use to make my text look different, graphically or otherwise, and I don’t have the patience to search free font resources to find one.  It doesn’t make sense for me to spend time flailing toward what would probably be an amateurish result when I  can just style some text and move on.

But that doesn’t mean that I settle for the default presentation of text, either.  An example is the “Cut your costs…” text at the top of Complex Spiral Consulting’s main page.  The relevant bits of its CSS are:

font: bold 166% Arial;
letter-spacing: -1px;

Yep, just boldfaced largish Arial—but the negative letter-spacing pulls the letters together just a touch, and significantly alters the visual impact of the text, making it seem weightier than normal.  Conversely, a positive letter-spacing would spread text apart, giving it a more open feel.  Most visual designers grit their teeth over this kind of thing, because it’s the crudest form of kerning imaginable.  True enough, but it still has a desirable effect, and typically one that’s underexploited.

The other reason I stick to text is that it’s almost infinitely more flexible than raster graphics like GIF and JPEG.  If I decide to resize the “meyerweb.com” in the masthead, all I have to do is fiddle with a line of CSS.  For that matter, I do change the size of that text on sub-pages, just as I do the masthead itself.  I can also change the text color to suit the masthead graphic.  For every one of the 26 mastheads I’ve created (this week’s is #6), I’ve changed the text color to blend better with the background.  Doing this was, again, a change of a couple of lines of text.  It would be even better if RGBA color values were widely supported, so I could make the text colored and translucent, but never mind that now.  (Side note to Dean Edwards: how about adding RGBA support to IE7?)

And, of course, the user can resize text, which I regard as a benefit, although a lot of designers regard it with pure horror.  If I had the name of this site as a graphic, then in any browser except Opera, resizing the content display would leave the graphic text unchanged.  That bugs me.  If there were universal vector-graphic support, say for SVG, then I could use it to create any font-and-logo combination I was able to dream up (or hire someone to create for me).  Even with the widespread availability of the Adobe SVG plugin, it’s still not enough.  I know, beyond any doubt, that a Web browser will support text rendering.  I don’t have the same confidence about SVG or even Flash, both of which can be scaled.

So I stick with text.  Realize that this is not what I would tell a corporate client to do.  When I worked with Macromedia, I didn’t tell them that they should replace their logo graphic with plain text stating “Macromedia,” and it wouldn’t even have occured to me.  If I were to work with Adobe or Microsoft or Apple or Red Hat or anyone who’d established a visual identity based around an image, I wouldn’t even consider telling them to replace it with plain text.  I’d agitate for sensible alt text, of course.

But for this site, or for Peter‘s site—both of which are personal sites—I don’t see anything wrong with using text.  (Then again, I guess I wouldn’t, having a corner on this particular market and all.)  In a lot of ways, I think it’s preferable, reducing bandwidth consumption and server hits.

And, y’know… it’s text.  C’mon, everybody, sing it with me:

Text is natural, text is good Not everybody loves it But everybody should

Or, if you groove to an older, smoother beat:

I can tell you, darling, that it’s textual healing Mark up, mark up, mark up, mark up, let’s design tonight Wake up, wake up, wake up, wake up, text will do it right

Thankyew!  Try the veal.

• Textual Healing was published on Sunday, March 7th, 2004.
• It was assigned to the CSS category.
• There have been no replies.

Back in January, I reacted to Peter Nederlof‘s whatever:hover by musing that it would be nice to see behaviors used to extend IE in other useful ways, like adding generated content support and so on.  Dean Edwards, regardless of whether he saw what I had to say, is doing that and more with his in-progress IE7 behavior suite.  Can you say, “will add support for attrribute selectors, multiclass selectors, and adjacent-sibling selectors to IE/Win?”  Oh yeah… I thought so.  And that’s just the beginning.  He has generated content on his list of things that will be supported, and a whole lot more besides.  The behavior is currently alpha, but it’s everything I could possibly have hoped for and more.  I’m going to be keeping a close eye on Dean’s work, and will be putting it to use as it moves out of beta.

In a similar vein, Dean also created an XBL binding to let Gecko-based browsers use Microsoft behaviors.  I think he just might be a genius.  Thankfully, he’s using his powers for good instead of evil.

Hopefully, I can get one of you XSLT gurus to do the same on my behalf.  I have a problem that’s proven beyond my ability to grasp.  Basically, I have a list of events that include start and end dates; here’s the basic markup that drives it.  I can get a list of upcoming events, no problem; I just pass in information about the current date when I run the XSLT and do comparisons.  What I need is a list of recent events, where “recent” is defined as occuring within the past three months, even if those months straddle a New Year.  I also want to get at least the most recent event even if it didn’t happen within the last three months.  And, of course, I want any results sorted in reverse chronological order.  I cannot figure out how to do all that in XSLT.  Any pointers or takers for this one?  I could really use some help.

(Yeah, yeah, doing it with some database or other would be a snap.  I’m trying to do it in XSLT.  Think of it as a creative design constraint.)

On a totally different note, here’s an interesting pair of articles from SF Gate: Gay marriage momentum stuns both backers and foes and Where Is My Gay Apocalypse?  Thanks to Jeff Veen and Simon Willison for the pointers.

• Hack and Slash was published on Saturday, March 6th, 2004.
• It was assigned to the CSS, Culture, Standards, and XSLT categories.
• There have been no replies.

Okay, maybe Tantek’s right and the CSS I devised yesterday wasn’t the greatest (note to self: avoid writing journal entries at 4:45am).  And yes, it would be more elegant, at least on the markup side, to use the href values to determine how to style links.  It feels a touch clumsy, for some reason, maybe because the selectors end up being so long and I’m used to short selectors.  Go check out what he has to say and suggestions for better selectors, and while you’re at it go take a look at substring selectors to get ideas for how to do even better.  (I don’t think anyone supports *= yet, so you’re likely to have to use ^= instead.)

Back in high school, my best friend Dave and I devised a scenario where water shortages in the American southwest became so severe that states literally went to war with each other over water rights and access, fragmenting the United States in the process.  It never really went much of anywhere, just an idea we kicked around, and that I thought about trying to turn into a hex-based strategic wargame but never did.  It’s always lurked in the back of my head, though, the idea of climate-driven warfare.

According to Yahoo! News, a Pentagon report asserts that climate change is a major threat to national security; well, actually, to global security.  And that if the global climate crossing a “tipping point,” the changes will be radical and swift.  In such a situation, economic upheaval will be the least of our concerns—we’ll be more worried about adding to the climate shifts with the aftereffects of nuclear exchanges.

I actually read about this on Fortune.com a few weeks ago, and although now you have to be a member to read the full article at Fortune, there’s a copy at Independent Media TV.  The Fortune article characterizes the report as presenting the possible scenarios if global climate shifts occur, but not claiming that they are happening or will happen.  It also says that the Pentagon agreed to share the unclassified report with Fortune, whereas the Yahoo! News article says the report was leaked after attempts to hush it up.  For that matter, the Yahoo! News article makes it sound like the report claims that The Netherlands will definitely be uninhabitable by 2007, and so on.  According to the Fortune article, that was one aspect of a scenario, not a concrete prediction.  This is probably due to the Yahoo! News article being a summary of an article in The Observer, which is a production of The Guardian and claims to be the “best daily newspaper on the world wide web.”  Uh-huh.

So I guess I’m saying read the Fortune article, as it gives more information and takes a more balanced tone—not that it sounds any less disturbing, really.  The fact that the report was commissioned at all suggests that the subject is being taken seriously at the Pentagon, which is not exactly a gathering place for leftist wackos.  I’ll be very interested to see what reaction, official or otherwise, is triggered by this report in the weeks to come.  My fear is that it doesn’t matter any more, that whatever accusatory words might get thrown around will just be insignificant noise lost in the rising wind.

• Gathering Stormclouds was published on Friday, February 27th, 2004.
• It was assigned to the CSS, Politics, and XFN categories.
• There have been no replies.

If you thought XFN or VoteLinks were the last (or only) word on lightweight semantic link annotation, think again.  Tantek writes about the idea of adding a license value to indicate a link that points to licensing terms.  In his post, the expression of this idea is centered around Creative Commons (CC) licenses, but as he says, any license-link could be so annotated.  Apparently the CC folks agree, because their license generator has been updated to include rel="license" in the markup it creates. Accordingly, I’ve updated my CC license link for the Color Blender to carry rel="license", thus making it easier for a spider to auto-discover the licensing terms for the Color Blender.

Tantek also said of the idea of applying CSS to documents that uniquely styles license-links:

I wonder who will be the first to post a user style sheet that demonstrates this.

Ooo, me, me!  Well, not quite.  I don’t have a complete user stylesheet for download, but here are some quick rules I devised to highlight license links.  Add any of them to your user stylesheet, or you can use these as the basis for your own styles.  (Sorry, but they won’t work in Internet Explorer, which doesn’t support attribute selectors.)

/* simple styles */
*[rel~="license"] {font-weight: bold;}
*[rel~="license"] img {border: 3px double; color: inherit;
  padding: 1px;}

/* add a "legal" icon at the beginning of the link */
*[rel~="license"]:before {content: url(legal.gif);}

Here’s my question: should the possible values be extended?  Because I’d really like to be able to insert information based on what kind of license is being referenced.  For example, suppose there were a c-commons value for rel; that way, authors could declare a link to be rel="c-commons license".  Then we could use a rule like:

*[rel~="c-commons"]:before {content: url(c-commons.gif);}

…thus inserting a Creative Commons logo before any link that points to a CC license.  At the moment, it’s highly likely that the only rel="license" links are going to point to CC licenses, but as we move forward I suspect that will be less and less true.  I hope we’ll soon see some finer grains to this particular semantic extension.

If you don’t like using generated content for whatever reason, you could modify the rule to put the icon in the background instead, using a rule something like this:

*[rel~="c-commons"] {background: url(c-commons.gif) no-repeat;
  padding-left: 15px;}

The usual reason to avoid generated content is that IE doesn’t support it, but then IE doesn’t support attribute selectors either, as I mentioned.  So don’t add any of these rules to an IE user stylesheet.  Use Firefox, Safari, Opera, or one of the other currently-in-development browsers instead.

In other news, I was tickled pink (or maybe a dusky red) to see that for sol 34, one of the “wake-up” songs for the Spirit team was The Bobs’ Pounded on a Rock.  My hat’s off to you, Dr. Adler!  I’ve been listening to that particular album recently, mostly to relearn the lyrics.  I’ve been singing to Carolyn when I feed her, and some favorites of ours are Plastic or Paper, Now I Am A Hippie Again, Corn Dogs, and of course Food To Rent.  It’s awfully cute that she smiles at me when I sing to her, mostly because I know one day she’ll grow up, learn about things like “being on key,” and stop smiling when I sing.

In the meantime, though, she’s perfectly happy to rock on!

• License To rel was published on Thursday, February 26th, 2004.
• It was assigned to the Carolyn, CSS, semantic web, and XFN categories.
• There have been no replies.