July 2009 Archives
Molly E. Holzschlag is a writer, teacher, public speaker and opera singer based in Tucson (who enjoys a paycheck from Norway and a life at 35,000 feet…).
An author, with a hand in over 35 books related to the craft of web design - including Transcending CSS and The Zen of CSS Design: Visual Enlightenment for the Web - she has inspired many of those at the forefront of today’s Web Standards movement. A movement she has continued to passionately lead since her recent appointment as a Web Evangelist at Opera Software.
There are few evangelists working within any industry known simply by their first name alone. Molly is one.
We asked Ms Holzschlag a dozen questions.
Where did you learn your craft?
I lost my roadmap, took a turn, and ended up here. What happened in between, I honestly cannot remember.
Who inspires you?
People who demonstrate courage, which is displayed in such a variety of ways, it’s difficult to categorise. I suppose I’m interested in anyone who challenges the status quo with a reasonable argument.
What are your influences?
Alan Turing. And sex, drugs and rock ‘n’ roll.
In ‘[Why] Web Standards Aren’t’ 1 you reference the standards that ease the lives of engineers on oil rigs, stating by comparison that, “What we have today, on the web, are not standards in the truest sense.” How important are standards, and can you envisage a time when web designers will enjoy standards in the truest sense of the word?
Web Standards are the closest form of organised anarchy with the goal to keep the web open and as free or low-cost as possible. It’s a mandatory fight.
Without a Web Standards movement, proprietary formats would move in too quickly at this sensitive time of the web’s life. This doesn’t mean proprietary formats are bad, it just means we need practical tools to carry out the true vision of the web, which is for everyone.
Which is more important: “valid and conforming” or “useful, usable, accessible and [what] really works”?
The latter. What works should always trump what’s specified. It goes back to the organised anarchy idea. Even if it’s
<marquee>. Would we have progressed to a richer web without them?
The W3C’s recent announcement that, “work on XHTML 2 is expected to stop at the end of 2009 to focus resources on the development of HTML 5,” sparked a great deal of - at times very heated - discussion. As a web standards advocate, what are your views on this announcement, the subsequent reaction(s), and its impact on the direction the web might take as a consequence?
This is the ‘elephant in the room’ question.
If I were a fortune teller of merit I would give you an answer, however, my honest answer is I have no idea what these changes will bring.
What I do know is that XHTML lost momentum despite its promises. In some ways, this is due to lack of browser support for XML MIME types, but in other ways this is due to no clear pathway as to what “Extensible” really meant. So I could write my own DTDs? I liked that idea, to be honest. I don’t want XHTML as XML to die per se. I’d like to see it really work and then make a decision.
But like it or not, HTML 5 is here, and as I said, and have said before, what is implemented trumps what is specified. HTML 5 is being implemented. XHTML, other than 1.0 or 1.1 served as
text/html, is not a cross-browser answer, despite over nine years in development.
HTML 5 represents a move beyond markup, what implications do you envisage this will have for designers and developers working on the web?
Ideally, HTML 5 will enrich the web by providing developers with a strong application and rich web platform. There’s elegance in HTML 5, but also chaos and confusion. I fear the latter, I embrace the former.
As a web standards advocate you’ve spoken about the importance of interoperability and being platform agnostic. In your role as a Web Evangelist at Opera Software - a company innovating within the browser space - do you believe interoperability is a goal that can be attained?
At night I often dream of a world where all platforms and user agents perform equally, if uniquely magnificently, and leave everyone satisfied with their experience. Then I wake up.
Universities are often criticised for being too slow to react to the rapid pace of change on the web. As an educator, do you believe web standards can be taught?
Absolutely. Just add students.
Universities are slower to move toward new requirements, usually. In the US we have state and community colleges, and often they are more capable of responding quickly to new curricula.
This is an enormous issue, however. At Opera Software, one of the tasks we have is to address this very thing, which we do in tandem with WaSP Interact Curriculum and Opera Web Standards Curriculum, run by Chris Mills and many other colleagues.
What’s your favourite typeface?
All of them. Except for Comic Sans. Unless a child sends me something in Comic Sans. Then it’s OK.
What’s your favourite plain text editor?
What’s your favourite tea?
Assam first flush brewed at least three minutes and served with 2% milk. Although Earl Grey strong and then iced and served with lemon and orange slices is very refreshing during Arizona summers.
Further details are available in the “carefully designed” Calluna PDF Specimen (272 KB).
The landscape for typography on the web appears to be changing on a near daily basis. Typekit, Kernest, EOT Lite, Tal Leming and Erik van Blokland’s .webfont Proposal (now at Version 2.1)… all are evolving week by week, day by day, hour by hour.
Fontdeck, a partnership between Clearleft and OmniTi, promises to be “a web service delivering real fonts to your web site.” (Coming soon.) With the talented Richard Rutter, of The Elements of Typographic Style Applied to the Web, behind it, it looks like one that should be well worth watching.
Sign up and you’ll be notified the moment it’s ready.
In terms of design (and (designed) information hierarchy) it might be a little old school, even intimidating, but if you’re a typographer on the web, Tal Leming and Erik van Blokland’s .webfont Proposal is well worth reading. As Leming puts it:
We’re hopeful that this is a good format for everyone. It gives users smaller file sizes. It gives the font vendors a simple format that allows them to include information about the font. It doesn’t require entirely new technologies from the browser developers.
With the current interest surrounding Typekit and yesterday’s announcement by David Berlow of Font Bureau of a proposal for a Permissions Table for OpenType, the issue of typography on the web is, thankfully, appearing to reach a tipping point (not before time).
Leming and van Blokland’s proposal is interesting for a number of reasons. Firstly, it appears to have the backing of a sizable number of the world’s leading type foundries. In barely a week it has attracted support from, amongst others, FontFont, H&FJ, Emigre, House Industries and Process Type Foundry. Those are some impressive names and, as Typegirl puts it (with just a hint of wit), “With this much weight behind it, it has to be taken seriously.”
Secondly, its approach to the issue of what to display when encountering a potential license issue is encouraging. Rather than refusing to display the typeface intended and fall back on another specified typeface, the browser renders the page with the typeface, but displays a simple, unobtrusive alert about the discrepancies in the font’s domain information.
Philosophically this embraces a Digital Rights Assistance (DRA) approach as opposed to a Digital Rights Management (DRM) approach and is consequently less likely to alienate end users.
So, what exactly is the proposal?
A Little Deciphering
To save you filtering through 500+ messages at the W3C mailing list (and to spare you from some of the more esoteric discussion) we’ve listed some of the key posts below. Brace yourself, however, the language can - at times - be less than user friendly.
So, in plain language, what exactly does this proposal mean?
If we’re interpreting it correctly (and feel free to comment if you think we’re not), what this means is as follows. A typeface in Leming and van Blokland’s proposed .webfont format consists of a compressed file containing two files with the following names:
The first file,
info.xml, contains information about the font including meta data supplied by the foundries. Critically it contains an
<allow> field which contains a list of URLs allowed to use the font (for open source fonts, this could be set to ‘any’, allowing the font to be used on all domains). This allows a font to be licensed for a specific domain. The second file,
fontdata, contains the actual font file.
Authors can specify a font using
@fontface in CSS, referencing a .webfont uploaded to the web site’s relevant directory. Simple. (And legal.)
What’s appealing about this proposal is its reliance on an existing standard -
@fontface - that removes the need for workarounds and hosted solutions (one of the key criticisms of the Typekit proposal).
The Tipping Point
Once again we find ourselves moving quickly in a debate about the future of web standards and the tools we use as web designers and developers to design for the web (witness the recent XHTML DOA WTF debate). It’s encouraging, however, to see that the issue of typography on the web appears to have reached a tipping point, with a huge amount of talent now addressing the issue.
Tim Brown’s recent experiments with open source typefaces, licensed for use on the web - Glyphs of Steel and All Aboard - showcase the potential of well-crafted type coupled with
@fontface. When the major foundries sign up to a proposal and it’s completed, designers will enjoy the many benefits.
Leming and van Blokland’s proposal looks, alongside a number of other recently announced proposals, to point in the right direction. We can only hope the issue is resolved, sooner rather than later.
Celebrating the 40th anniversary of the Apollo 11 Lunar Landing, We Choose the Moon is a beautifully crafted site featuring “a real-time, interactive re-creation of the Apollo 11 mission, 40 years after Apollo 11 made JFK’s dream a reality.”
The site is launched today - July 16, 2009 at 9.32 am EDT - exactly 40 years after Apollo 11 lifted off and allows users to follow Apollo 11’s progress in real time on its journey to the moon.
As the dust begins to settle on the debate sparked off by the W3C’s recent announcement that work on XHTML 2 is expected to stop at the end of 2009 to focus resources on the development of HTML 5, it’s worth pausing to reflect upon what this means to web designers and developers moving forward.
Judging by the often heated comments made over the last fortnight, this is clearly a topic that many feel passionate about and weren’t afraid to express an opinion on. Indeed, at times it appeared as if this were the only thing on many notable standardistas’ minds.
The original announcement quickly drew comment from Jeffrey Zeldman, sparking a protracted, and at times less than productive, discussion in the comments of his post XHTML DOA WTF. It even gave rise to a short poem - This Is the House, courtesy of Mark Pilgrim - again, unsurprisingly, sparking a heated discussion.
Watching from the sidelines it’s debatable what many of the the cheap shots being traded left, right and centre achieved. Surely what most web designers and developers want, is to do the best that they can with the tools they have at their disposal? A point Eric Meyer summed up neatly.
First things first - and worth noting, given the often repeated, post-announcement mantra that, “XHTML is dead” - XHTML is not dead, it still exists as XHTML 5. If you don’t believe us, ask the doctor, he couldn’t be any clearer:
HTML 5 + XML = XHTML 5
So, XHTML 5 is alive and well, but XHTML 2 is dead? Admittedly this might appear confusing, perhaps it’s best to give this some context.
A Little History
To understand the current situation it’s best to get a little background, specifically to ask, what was XHTML 2 all about and how did it differ from HTML 5?
One of the main differences between the development of the XHTML 2 and HTML 5 standards lay in the XHTML 2 Working Group’s controversial decision to discard backward compatibility and create a language - based solely on XML - that was unconstrained by current browser implementations, limitations and history.
This decision to break with the past in an attempt to create a theoretically ‘pure’ vocabulary arguably resulted in XHTML 2 becoming an increasingly theoretical proposition, lacking pragmatism, and is perhaps one reason the W3C reached its recent conclusion.
HTML 5, on the other hand, has been developed with current and future browser development in mind and, as a consequence, has recently been gaining significant momentum. Driven by pragmatism, it has recently emerged as a the W3C’s choice for “accelerated developement”. John Allsop summarises it best:
One of the lessons the web continues to teach us is that it values pragmatic development over theoretical perfection.
The pragmatic approach taken when developing HTML 5 perhaps helps to explain why it has emerged as the frontrunner in the web’s evolution, with the W3C stating that, through its recent decision, “it hopes to accelerate the progress of HTML 5 and clarify [the] W3C’s position regarding the future of HTML.”
So, XHTML? Was it all just a waste of time?
Is XHTML DOA (WTF)?
Herein lies the confusion.
The W3C might have chosen to focus its future efforts on supporting the development of HTML 5, however, this does not mean that the benefits of learning (and using) XHTML are suddenly redundant.
For one thing XHTML’s strict rules - lowercase markup, quoted attributes, correctly nesting elements, closing all opened tags… - are a lot easier to follow and remember than the rather more lenient rules of HTML 4.01.
Not only does XHTML’s stricter syntax leave some feeling nice and comfy, it’s also easier to learn, one reason many are still embracing it. Drew McLellan states, “In my experience simple, strict rules are much easier to learn … rules that require no thought, and result in uniform, predictable markup.” Ian Lloyd agrees, “I still believe that it’s better to teach XHTML syntax rather than HTML, so that beginners learn rules rather than (arguably lazy) shortcuts.”
This emphasis on the creation of well-formed markup might be argued as the catalyst that, “taught a generation of developers what good code should or could look like,” as Jonathan Snook (an HTML 4.01 aficionado) puts it. It also offered the “opportunity to rethink markup,” as Jeffrey Zeldman summarised in his follow up post In Defense of Web Developers:
XHTML’s introduction in 2000, and its emphasis on rules of construction, gave web standards evangelists like me a platform on which to hook a program of semantic markup replacing the bloated and unsustainable tag soup of the day. The web is better for this and always will be, and there is much still to do, as many people who create websites still have not heard the call.
At the end of the day it boils down to choice, as Jeremy Keith points out in his excellent article Misunderstanding Markup:
The death of XHTML 2 does not mean the death of XHTML syntax. If you want to continue to close all tags and quote all attributes, you can do so. You can either use the existing XHTML 1 spec or you can use HTML 5.
The bottom line? What you’ve learned to date will form a solid foundation on which to build moving forward.
We operate in an industry that changes rapidly. Keeping abreast of developments as they emerge is essential to maintain an understanding of contemporary practice moving forward.
The following links are worth setting aside a few hours for and reading, if only to enable yourself to fully appreciate the complexities of the current debate from the different participants’ perspectives.
It’s perhaps best to start with the two posts by Mr Zeldman that started the ball rolling: XHTML DOA WTF and In Defense of Web Developers, both are well worth reading, particularly their extensive comments.
Lachlan Hunt’s A Preview of HTML 5 at A List Apart is still well worth reading, despite being published in 2007; John Allsop’s more recent Semantics in HTML 5, also at A List Apart, is an excellent and insightful exploration of semantics in HTML 5. Lastly, as HTML 5 takes hold, it’s worth checking in periodically with the HTML 5 Doctor.
Tim Brown of Nice Web Type likes Museo and Museo Sans. To prove it he’s created All Aboard the Exljbris Express, a carefully crafted type specimen showcasing Museo and Museo Sans and demonstrating
@font-face in action. Lovely.
Mr Zeldman on Apple’s web standards secret sauce:
Webkit + Apple = Secret Sauce
A secret sauce that’s selling web standards to a new generation of, in many cases mobile, consumers and developers.
If you appreciate Helvetica, you’ll appreciate Helveticons: 245 royalty-free vector icons based on Helvetica Bold. Lovely.
With a lovingly crafted interface, the Hobnox Audiotool enables you to produce electronic music 100% within your browser, all within an intuitive, ruthlessly-attentive-to-detail environment.
With exquisite emulations of Roland’s TR-808 and TR-909 drum machines, a choice of Roland’s iconic TB-303 sequencer or a ToneMatrix synthesizer inspired by Yamaha’s Tenori-on, a wealth of effects boxes and the ability to record WYHIWYG (what you hear is what you get) direct to your hard drive, prepare to lose an afternoon ‘researching its potential’…
A simple style sheet (which includes a number of CSS3 specific elements), SlickMap CSS allows for the rapid creation of fully functioning site maps from unordered lists in HTML. Suitable for most web sites it’s well worth downloading and experimenting with. See it in action at the SlickMap CSS Demo.
There’s been a great deal of debate since the W3C announced that the XHTML 2 Working Group [is] Expected to Stop Work [at the] End of 2009 to focus resources on the development of HTML 5. It’s a debate all web designers and developers should familiarise themselves with.
Mr Zeldman’s recent articles XHTML DOA WTF and In Defense of Web Developers are a useful point of entry into the (often heated) discussion. Both articles are well worth reading, in particular the lengthy discussion they generate in the comments.
The future of HTML 5 (and how XHTML fits within that future) will affect most designers and developers working on the web today, consequently it pays to keep abreast of developments. Watch this space for an easy-to-parse roundup of the discussion which we’ll be posting in the next few days.
With a focus on “progressive enrichment” and attention to detail, Handcrafted CSS - a partnership between Messrs. Cederholm and Marcotte - looks set to be a must have purchase when it hits the shelves on 10 August, 2009.
It’s refreshing to see the foregrounding of craftsmanship in (web) design, an often overlooked aspect of the design process that can make - or break - a design. Get your copy now.