C|Net has an article about the future of Internet Explorer. Much of the article is speculation fueled by vague quotes from various Microsoft employees about how wonderful web browsing in Longhorn, the next version of Windows, will be.
Wade through the fluff, though, and there's a pretty good history of the browser wars and a reasonably accurate description of the state of the market today. Most significantly, the article makes the point that Microsoft isn't so interested in the web as a publishing medium; they've got their sights set on the web as an application platform. Exactly right, and that's where the action will be over the next few years.
According to ThinkSecret, a revived GoLive 8.0 will join Creative Suite v2.0 to be released in early 2005. The news item reports Adobe is working to improve the handling of CSS content including a toolbar for adjusting CSS layouts.
Also getting a makeover is the grid element for CSS DIV authoring for easy switching between DIV and T-t-table layout views.
I’m looking forward to seeing how it handles CSS in comparison to Macromedia’s Dreamweaver. Will GoLive have a chance against Dreamweaver? Maybe it’s too late? Maybe not - after all, Firefox managed to shake up IE and Dreamweaver is not as dominant in Web authoring as IE is in Web browsing.
Fellow WaSP Dave Shea has cracked the nut of making min-height work in Safari. Ironically, he does it without using min-height. Well, almost: there's still a 'phantom' min-height in there to get Opera to do the right thing, but that's it.
Nice work, Dave!
The wags over at /. are speculating that Google may release a browser. Their evidence includes recent hires at Google and the fact that Google already owns the Gbrowser.com domain.
Personally, I'd say this one lands somewhere between 'unlikely' and 'I'll have some of whatever you're smoking' on the bullometer, but there's some fun speculation in the comments nevertheless.
Update: UK-based IT rag The Register is reporting on Google's purported browser effort too, though apparently based largely on the same New York Post article.
Zeldman has announced another Happy Cog redesign. This time, they've worked their magic on the Kanas City Chiefs' site, and have supplied a before image and a discussion of the work.
Too bad Zeldman & co. don't remake defenses, too. ;-)
Remember the bad old days of 2000-2001? Back when Netscape Communicator still had appreciable market share on most sites? If you were working with CSS layouts then you probably don't; your brain has buried that period deep in your subonscious mind. It's just too painful a memory to deal with.
Mine had done, anyway. And the backward-march of my hairline had slowed markedly as a result. I'd been sleeping peacefully through the night, with no wee-hours screaming about JavaScript stylesheets and double-loading. Then Eric Meyer brought it all flooding back with his post about a crash-inducing bug for IE/Win.
So it's official, then: IE/Win has inherited Netscape 4.x's mantle as the red-headed stepchild of the browser market.
Here's hoping the IE team rights the ship post-haste. It's a massive project, but if anyone can do it, it's those guys.
MacRabbit is a new CSS editor for the Mac. I haven't had a go with it yet, but the extraction feature alone, which allows you to suck down the CSS for an arbitrary site using a custom user agent string so you can 'spoof' different browsers and circumvent browser-sniffing, looks like it might be worth a shot.
Hat tippin': Jeffrey Veen
Update: Alert reader Joe Lewis notes that the program itself is called 'CSSEdit'. 'MacRabbit' is the company that develops it. Joe also gives the program a glowing review, saying he's been a happily-licensed user for several months already.
Thanks for clearing up the confusion, Joe!
Spread Firefox has reached its goal with five days to spare. Any guesses as to the number of downloads at the end of day 10?
FirefoxIE has been updated to work with Firefox 1.0PR. See the release.
Microsoft's Internet Explorer share has dropped from 95.6% in June to 93.7% this month. This is the most notable slip for IE's dominance in over seven years.
Mozilla browser share has grown from 3.5% to 5.2%.
The Spread Firefox campaign is on, with a promotion to encourage one million downloads in ten days. They're already halfway there!
Of course, WaSP continues to promote a better browsing experience via our Browse Happy web site.
The numbers might be small, but the rewards for those who have made the switch are indeed mighty.
Source: BusinessWeek Online, A Firefox in IE's Henhouse.
Why John Gallant, of course. The ex-pool repair guy is now an internationally recognized CSS guy instead. You can read all about his transformation (oh, that's a bad pun) in the recent “Ten Questions for John Gallant (Big John)” from our great friends at the Web Standards Group.
Aleksandar Vacic has posted what looks to be an exceedingly thorough guide to Z-index positioning. He's a good man, and thorough. He goes throgh the effect of z-index values on both absolutely- and relatively-positioned elements in a variety of browsers. Definitely one for the bookmarks file.
Update: Well, that didn't take long: Dave Hyatt has posted a correction to Aleksandar Vacic's guide to z-index positioning. Being the alert CSS wonk that he is, Aleksandar has corrected his guide accordingly.
The Mozilla Foundation has released Firefox 1.0PR (that's 'Preview Release', or 0.10, one of the last releases before it goes 1.0), Thunderbird 0.8 and version 1.7.3 of the Mozilla Suite.
Especially noteworthy in this release is rudimentary support for ATOM and RSS feeds. Neither RSS nor ATOM is a by-Hoyle web standard, though ATOM is being standardized by the IETF. To my knowledge, despite the use of RSS by dozens of big-name publications like the BBC and The New York Times, there is no similar effort underway for that format. In fact, the person most influential in the RSS world, Dave Winer, has expressly disavowed such a role for the RSS Advisory Board he founded. Nevertheless, these are two useful technologies. Cheers to the Mozilla Foundation for embracing them both.
Firefox displays a rectangular orangs 'RSS' icon on the right side of the status bar when you're browsing a site with an RSS or ATOM feed. Click the icon, and Firefox will add the feed to your bookmarks. It treats feeds as a folder, and individual items as bookmarks. It seems a bit kludgey to me, as there isn't any obvious way to see only article summaries rather than hitting the sites themselves.
In Thunderbird, one sets up a 'News & Blogs' account and adds feeds there. One has the option of either loading descriptions or the full web page into the main window when one clicks on a headline, and can organize the feeds into folders. All in all, it's not unlike Ranchero Software's NetNewsWire Lite for OS X, or a primitive version of Newsgator, a feed reader extension for Microsoft Outlook on Windows.
Both programs, and the suite, sport numerous other smaller improvements and bug fixes as well. If you haven't tried either yet, it's time to give 'em a whirl.
A couple emailers have pointed out that Microsoft has apparently won a patent on tab navigation of links in a browser.
Theoretically, this means Microsoft could demand royalties from any company or organization whose browser allows users to find and navigate through links on a web page using the 'tab' key. Given that open source projects like the Mozilla Foundation lack the funds to pay such royalties, were Microsoft to enforce the patent it would seriously impact the accessibility of the web for people using their browsers.
In reality, despite the fear and loathing the news has engendered in the open source community, I find this outcome unlikely in the extreme. In the first place, as the article linked above indicates, the patent needs to hold up in court. As other outfits have discovered, that's hardly a slam-dunk when it comes to broad patents like this. Second, Microsoft's statement from the article linked above doesn't seem to indicate any aggressive intent on their part:
In response to a query from internetnews.com about how it planned to handle the patent, Microsoft said: "We respond to inquiries about our portfolio and typically have private collaborative discussions with companies about using our technology. Consistent with practice throughout our industry, we don't believe it's constructive to identify specific products and start labelling them as infringing or non-infringing."
It's not as reassuring as Microsoft's statements about their patent on CSS, but neither does it sound like the angry noises we heard from Eolas, or Forgent Networks.
To me, it seems the real impact of this patent is to drive home just how misguided the U.S. intellectual property regime has become, particularly with regard to software patents — and why Europe and Australia would do well to think twice before running headlong down that same path, as they appear to be doing.
This isn't really a 'standards' issue, to be honest. It's a social/legal/economic one. As such, I feel it's really outside the scope of Buzz. I'll just leave it now with the observation that lots and lots of bigcos amass patents as a hedge against someone coming after them for patent infringement: "Oh, so we infringed your patent did we? Well you infringed our patent first!" It's a seven-figure-settlement version of schoolyard squabbling, as in the recent Macromedia-Adobe brou-ha-ha. Distasteful? Sure. But it happens. And Microsoft would be remiss if they didn't try to protect themselves.
Paul Bellows has revisited the even-height CSS columns question discussed in a couple of previous posts.
Paul's method uses ECMAScript, the DOM and some non-standard properties to work it's magic. Personally, I'm not a big fan of using ECMAScript for basic layout. Neither is Paul, truth be told. But at the end of the day, some browsers need a bit of help doing the right thing.
Is Paul's method any better than the Eric Meyer/Douglas Bowman method? I'm not sure. A bit of ECMAScript doesn't seem to me any worse a hack than an oh-my-&deity wide background image or a lightweight layout <table>. Yet I'm not sure I like the idea that the layout relies on visitors having ECMAScript enabled on their browsers, particularly considering the well-publicized security issues some browsers have.
On balance, while I really haven't had a chance to work through Paul's method enough to form a firm opinion, I probably lean towards a pure-CSS or even the <table>-based solution.
Still, it's good that Paul has taken the time to explore this avenue and post his results. At the very least, it may form the basis for solutions in places where other techniques fall down. So if you have a free moment, spin by his site, look over — or try out — his code and send him some feedback.
While you're there, give a listen to some of Paul's tunes: like Scott Andrew Paul's a gifted musician and recording artist as well as a skilled scripter. Where do they find the time?
More sites jumping on the CSS + structural markup bandwagon:
Firtst, Happy Cog Studios has redesigned the Amnesty International USA site. Jeffrey Zeldman discusses the challenges of working with a large site on a tight budget and his experience with the dreaded 3rd party problem. He also has a more complete discussion of the issues here.
Next up, telecomms giant AT&T has pulled off a valid XHTML strict home page, complete with a Flash movie (yeah, SVG might be nice, but realistically browser and plugin support just hasn't reached practical levels). Go Ma Bell, go!
Finally, the open source Mono Project has come within three errors of validation. The gotchas are small: a non-standard attibute value, a missing alt attribute and a missing <li> tag. While not technically valid, they are using clean, semantic markup and that's the most important thing. Besides, I'm inclined to cut 'em a bit of slack on the site — replicating Microsoft's .Net framework as a cross-platform API is a momentous undertaking, and little details like three piddling validation errors on the site are bound to get overlooked here or there while they persue the greater goal.
Hat-tip to Eric Meyer and the 'Redesign Watch' item he's got tucked away in the right-hand column of his site.
Update: Joe D'Andrea of the AT&T web team emailed to say that we haven't seen nothing yet.
Joe's team has completed makeovers of some 30 pages on the site, and are slowly but surely working through the remainder. Meantime, he says they've got some serious DOM and ECMAScript goodness up their sleeves, and after that will be moving on to IFR and sIFR. Ultimately, Joe says they're aiming for zero inline ECMAScript and all page behavior defined through context and content. Woohoo!
It's great to see that a decade and change after spinning off Bell Labs, someone at Ma Bell is still keeping the spirit of innovation alive.
Fellow WaSP Dave Shea is compiling list of web design resources for beginners.
This project promises to be as useful in standards evangelism as Dave's Zen Garden. I'm really looking forward to seeing what he unearths.
The Poynter Institute has posted their EyeTrack III study, a fascinating look at how people view web pages facilitated by a technology that allowed researchers to track participants' eye movements as they surfed.
Hat-tips: Dan Gillmor and Steve Reubel via Robert Scoble
As part of his piece on best practices for online captioning, Joe Clark has also published a compendium of techniques for using <embed> and <object> with valid markup. This one's going in the bookmarks for sure.
Joe Clark, accessibility guru and author of the excellent Building Accessible Websites, has pointed me to 'Best practices in online captioning.' It's 21 chapters based on a government-funded university project, but don't let that fool you ;-): it's by far the most comprehensive work I've seen on online captioning.
I find myself doing more and more projects that call for advanced multimedia: audio, Flash, video and the like, as opposed to the more ordinary text + graphics. At the same time, many of my employer's clients are either in the financial industry or serve various government institutions — the sorts of clients for whom meeting accessiblity guidelines are a matter of law. Joe's work in this area will be a tremendous resource for me in the months to come.
Keep 'em coming, Joe!
Following up on the variable-width, even-height CSS columns technique he worked out with WaSP Douglas Bowman, Eric Meyer has added a couple of posts explaining the CSS table-layout properties.
Last week, an article on evolt called Ten CSS tricks you may not know made the rounds through the CSS blogosphere. CSS luminary and erstwhile IE 5/Mac developer Tantek Çelik is doing some peer review. A must-read, if only for information on IE/Win's support for multiple class selectors and why IE/Win sometimes appears to ignore !important declarations.
Andrei Herasimchuk has posted an excellent logo design tutorial based on his efforts to redesign the W3C logo. Andrei undertook the exercise after Dean Jackson asked him to lend a hand with an upcoming W3C ten year anniversary event.
Eric Meyer and WaSP Douglas Bowman have teamed up to develop a technique for creating multiple columns of equal height and variable width using CSS.
Eric's discussion also includes his thoughts on the expediency of the odd layout table, while Doug frames his explanation in a discussion of the advantages and drawbacks of liquid design.
Both are must-reads.
Update: WaSP Molly Holzschlag has posted her thoughts on the technique and the use of <table>s in such circumstances. Her verdict: sometimes you just can't beat a lightweight <table>.
She's got a point. Sometimes <table>s do just behave better. Lately I've mostly been marking-up designs done by others, and thoes others are firmly in the fixed-width camp, so I haven't had to worry so much about liquid columns. My problems tend to be vertical alignment: getting the tops — or worse yet, the bottoms — of variable-height elements to line up evenly. Perhaps I lack imagination, but I've not found any good way to do it. One can set a height for the elements that one hopes is enough to accommodate all the content, but one extra line of text and the whole thing goes bork.
Personally, I'm with Eric: the lack of a grid-based layout system in CSS is a head-slapper, what-were-they-thinking omission. It's the elephant in the corner when discussing the relative merits of <table>s vs. CSS for layout. The CSS2 table-layout properties would more or less mitigate the problem, I suppose. But with IE 5 Mac & Win still about in appreciable numbers, table-layout just isn't practical for most sites (yet).
Until it is, or until something better comes along, I fear the odd layout <table> will continue creeping into web designs. And semantics and separation of style and structure will continue to suffer.
WestCiv is holding a competition on developing CSS + XHTML templates for their StyleMaster web development applicaion.
StyleMaster is the brain child of John Alsopp, a former WaSP CSS Samurai.
Hat-tip: Jeffrey Zeldman.
A List Apart has a nifty piece on web design for handhelds. The article was written by Opera's Jorunn D. Newth and W3C Working Group invited expert Elika Etemad.
Tip o' the chapeau to Jeffrey Zeldman.
Somehow I missed this for nearly a month, but Dave Hyatt has added XSLT support to Apple's KHTML-based Safari browser.
Dave says he's working on an ECMAScript API for document transformations, and he's also asking for test cases using xml-stylesheet, as well as general feedback on XSLT support in general.
While I wasn't looking, Tantek Çelik, formerly a prime mover behind erstwhile CSS standard-bearer IE 5.x for Mac, has redesigned Technorati and Election Watch 2004 using structural markup and CSS.
Though Tantek too has been bitten by the dreaded 3rd party problem, his markup is still a treat.
Tantek gives a peek into his thought process during the projects on his blog, and promises a more thorough discussion of both the Technorati redesign and the in-progress redesign of his own site in days to come.
WaSP D. L. Byron recently observed that Macworld has hopped on the CSS layout bandwagon.
It's a nice design, and largely well-crafted, but is let down in the end by advertisement markup littered with FRAMEBORDER and MARGINWIDTH attributes and a mess of unencoded ampersands in various links. Looks like the "3rd party problem" I mentioned when discussing the RE/MAX redesign has claimed another victim.
Dodgy ad markup aside, Macworld appears to have done everything in their power to do it right, and their web team deserves kudos on a job well done.
After a sluggish spell — occasioned, no doubt, by the last-mile sprint to finish XP SP2 and recovery therefrom — IEBlog has started posting some gems.
Of particular interest to web developers is a post explaingin IE for Windows XP SP2's updated user agent string.
Also, be sure to catch Jeff Davis' recent post on the details of pop-up blocking in IE for Windows XP SP2.
WaSPs Meryl points to a slick new RE/MAX design, featuring valid XHTML and a (mostly) tableless, CSS-driven design.
A few of the sub-pages need some work, but most of the problems seem to derive from external service providers over which RE/MAX has limited control.
Such 3rd party problems are all too common, but as more site owners insist on valid, well-crafted markup for their own sites those service providers will undoubtedly find it in their interest to get their acts together.
Still and all, the RE/MAX team deserves a hearty pat on the back for a job well done. Keep it up!
