Tux news - Linux ubuntu fedora gentoo gnome news

KLEPAS.ORG (20 unread)

Meme: Reflections 2008

Posted: January 4th, 2009, 7:14am CET by Pascal
Copying Nathanael who stole if from Fiona, here’s a quick rundown of 2008, question and answer style:

Q: What did you do in 2008 that you’d never done before?
A: Left the country and moved overseas for half a year back to Germany to touch up my German, started freelancing properly, co-(un)organised Canberra’s very first BarCamp.

Q: Did you keep your new year’s resolutions, and will you make more for next year?
A: Much like Nat, I can’t remember any of my new year’s resolutions from last year.

Q: Did anyone close to you give birth?
No.

Q: Did anyone close to you die?
During this past year not, but Blacky passing away on Christmas Day last year is still fresh in our memories. We all miss the furry, barking bugger.

Q: What countries/states did you visit?
A: In Australia: Victoria for LCA, NSW a number of times for work; abroad: Germany (Frankfurt am Main, Mainz and Berlin), Turkey (Istanbul for GUADEC) and the UK for LugRadio Live.

Q: What would you like to have in 2009 that you lacked in 2008?
A: To have have attended a number of more conferences, or rather, not missed the ones with particular speakers, notably WDS.

Q: What was your biggest achievement of the year?
A: Becoming more independent as a result of my time in Europe and summoning the courage to get my most recent two piercings.

Q: What was your biggest failure?
A: Not managing to organise a more thorough trip throughout Europe alone through youth hostels and the like, partly due to money, partly due to time and partly due to my own negative misconceptions.

Q: Did you suffer illness or injury?
Nothing serious thankfully.

Q: What was the best thing you bought?
A: The AUD $1,800 return air ticket to Germany including health insurance, closely followed by my beloved black MacBook, Leno.

Q: Whose behaviour merited celebration?
A: Both Mum and Andy for overcoming more personal crap than any one person should put up with.

Q: Whose behaviour made you appalled and depressed?
A: Appalled and depressed are strong words, and I’m pretty laid back, so nothing really gets on my nerves much—and more importantly I try not to let something bother me for more than half a year. So, with this in mind, the name that rises to the top of the bad people cesspool in recent memory probably include Senator Stephen Conroy for his pathetic Australian mandatory clean feed idea.

Q: Where did most of your money go?
A: Travel costs and design books, most of which were typography books, naturally.

Q: What did you get really, really, really excited about?
A: My Europe trip, conferences—notably Canberra’s first BarCamp—and attending Kitty in Berlin. Oh, and currently still internally raving about the possibility of landing an exciting job (and thus stopping freelancing).

Q: What song will always remind you of 2008?
A: Krilatie Kacheli by Evgeniy Krilatov, this amazing Russian tune that was overlaid on video on a ModBlog (NSFW) video of the Russian Pain Theatre gang doing some awesome multi-tiered suicide suspensions.

Q: Compared to this time last year, are you:
- happier or sadder?—happy as a clam at the time of writing;
- thinner or fatter?—went up a kilo or two but still thin as usual;
- richer or poorer?—my bank statements would indicate little change, but I’ve accumulated new assets (books mostly).
Q: What do you wish you’d done more of?
A: Parkour, rock climbing, and traveling.

Q: What do you wish you’d done less of?
A: Nothing really… maybe sleep a little less?

Q: How did you spend Christmas?
A: With my Opa in Budenheim, near Mainz (Germany).

Q: Did you fall in love in 2008?
A: Big fat negatory.

Q: How many one-night stands?
A: Two.

Q: What was your favourite TV program?
A: The West Wing.

Q: Did you make a friend with anyone that you didn’t know this time last year?
A: Yes, after entirely coincidentally meeting both Söda and his girlfriend, Eugenie from Canberra all the way in Frankfurt. And to think in all the time I had known Söda he never mentioned his girlfriend!

Q: What was the best book you read?
A: The Elements of Typographic Style by Robert Bringhurst.

Q: What was your greatest musical discovery?
A: Vibrasphere, an amazing progressive psytrance group from Sweden.

Q: What did you want and get?
A: To improve my German and a MacBook.

Q: What did you want and not get?
A: An iPhone that wouldn’t become a paper weight when I were to take it home to Oz, and possibly a romantic relationship. The iPhone is being taken care of hopefully this week.

Q: What was your favourite film of this year?
A: Death Proof, part of the Grindhouse double set by Quentin Tarantino.

Q: What did you do on your birthday, and how old were you?
A: Celebrated it in Budenheim with my Mum, Opa, and German friends. Age: undisclosed.

Q: What one thing would have made your year immeasurably more satisfying?
A: To be with the many friends I left behind in Oz during my time in Europe.

Q: How would you describe your personal fashion concept in 2008?
A: Alternative whilst approachable and friendly? Just something different I guess.

Q: What kept you sane?
A: Mum and sleep.

Q: Which celebrity/public figure did you fancy the most?
A: I don’t really romantically fancy any celebrities, but was quite hot for seeing Obama get into the White House.

Q: What political issue stirred you the most?
A: The thought that even still in my lifetime the bollocks that’s going on in the Middle East probably won’t be solved. That, and the aforementioned proposed Australian Internet filter bollocks.

Q: Who did you miss?
A: Chewie our aging, barking furball of a lady, close friends from Canberra and Mum.

Q: Who was the best new person you met?
A: Eugenie Edquist, Mr. Söda’s lovely girlfriend.

Q: Tell us a valuable life lesson you learned in 2008.
A: It goes like this:

Find your obsession;
Every day, explain it to one person you respect;
Edit everything, skip shortcuts;
And try not to be a dick.
Get better at the above.

I’ve decided to skip the last question of which song lyric most sums up my year (I don’t know nor have the time right now to figure it out).

That’s it for ’08. What about you?
Evaluating fonts: kerning

Posted: December 14th, 2008, 4:01am CET by Pascal
There are thousands of fonts available and identifying the cream of the crop is not just an exercise of picking the most aesthetically pleasing. I constantly drool over new typefaces I come across; there’s usually a weekly dosage of five to ten that catch my eye, thanks of course to sites like I Love Typography.com. This has already caused a number of impulse-buys. That, and the review of a draft article I was peering over this evening got me thinking, what makes a good typeface?, or rather, what makes a good font? I’m distinguishing fonts from typefaces as the ‘physical’ (e.g. metal) or digital (e.g. font file) embodiment of the design; whilst typefaces are the design of the font, the ‘face’. Here are some rough evaluations, besides aesthetic merit:
Unicode character support
- Language support: how broad and faithful is it? (e.g. the correct glyphs for the right characters and availability of the typeface in other languages, such as Arabic or Cyrillic);
- Alternate characters: ligatures and stylistic variants (e.g. italic swash versions of z aren’t uncommon);
- Small capitals including small-cap figures (rare);
- Figures: hanging or “old-style” figures (for use in running text), lining figures, monospaced figures (for tabular work), superior and inferior figures, fraction figures;
- Mathematical operators and Greek letters (latter is also commonly used in applied sciences);
- And more, such as the International Phonetic Alphabet….
Format
The format the typeface is made available in should also be put up to consideration. It’s a subject I’m not as familiar with, but general advice reads opting for OpenType over other formats like
TrueType. As I understand it, many advanced features like automatic insertion of ligatures (e.g. fi to ﬁ), stylistic variants of characters and more are only accessible in OpenType (provided of course the program you’re using supports the advanced OpenType typography features).
Kerning table
Kerning is the horizontal adjustment of glyphs in a combination of glyphs (e.g. words) to ensure an optically correct fit, eliminating large gaps or tight collisions that occur with some compositions. Notably, f is frequently kerned to the right, as is W and V to remove undesired spaces that would appear on the bottom half of the glyph between common characters that follow it, such as the vowels.

Kerning tables are contained within font files and vary from non-existent to extensive. It’s important to note that good kerning tables should ideally accommodate for glyph compositions that occur in other languages.

Besides the drooling over the aesthetics, kerning is usually the other aspect I pay the next most attention to and if possible test. Most seasoned type designers and foundries will supply specimen sheets for their typefaces—often downloadable PDF files—with texts of various languages set in the fonts of the face, at various sizes, styles and so forth. As boring as it might be, it’s important to have a look over these, native speaker or not—you don’t have to understand the gobbledygook: just look out for large gaps between glyphs or ones that lie too close to one another. You might never write actively in more than one or two of the supported languages but nouns like places and people will almost certainly require setting.

If there aren’t various language specimens to check over or a specimen is lacking to begin with, ask to the supplier to throw the font(s) at a kerning test sheet and see how they do. A fun and handy exercise is to go through the fonts in your collection and see how they stack up; here are some of the blatant shortcomings of Linotype AG’s Times and Monotype’s Times New Roman, widely distributed on Mac OS X and Windows respectively:

Times kerning problems (view full size).

Times New Roman kerning problems (view full size).

As you can see Linotype’s Times has fewer kerning issues which is why I prefer listing it before and in addition to Times New Roman in my CSS font-family stacks.

That’s it. I’d be interested in seeing any kerning issues you find, particularly on widespread fonts, such as the web core fonts. Other than that, wish you a smashing Sunday.
Web fonts: the death of type foundries?

Posted: November 13th, 2008, 12:40am CET by Pascal

In the spirit of Halloween Jon Tan covered the ominous topic of @font-face linking in a comprehensive article on his beautiful website.

I’ve kept the implementation example to a minimum because it’s well covered by Jon, other websites, and the W3C. Instead, I wanted to voice my thoughts on the struggle between type foundries and designers. Feel free to skip the technical summary.
Implementation summary
‘Web fonts’ is currently a W3C CSS3 working draft. It allows for the direct linking to font files that are then used by the browser to set text in, broadening the relatively limited palette of ‘web-safe’ typefaces. Many designers have been longing for direct font linking since the earliest days of CSS. This is achieved with the @font-face element (fig. 1).
@font-face { font-family: 'Desired-Font'; src: url('Desired-Font.otf') format('opentype'); }
Fig. 1. After defining the @font-face element, you can directly refer to the linked font with the font-family property:
body { font-family: 'Desired-Font', …; }
Embedded OpenType
We’re all used to the major browsers not behaving in cohesion and web fonts is another victim of this. Presently @font-face is supported in WebKit (and consequently browsers that use it), Mozilla Firefox 3.1 beta and the development version of Opera. Internet Explorer (IE)—again, not surprisingly—does it’s own thing. Enter Microsoft’s proprietary Embedded OpenType (EOT). EOT was an endeavor by Microsoft to allow for font file linking on the web while safeguarding the typeface from being illegally copied by embedding the font, document to font and font to document. Linked fonts only are linked in one-way, document to font, allowing users to obtain the font with little difficulty. With Microsoft firmly opposed to supporting native OpenType linking as long as vendors object to it, EOT is set to be the only method available in IE for use with @font-face.

Technologically it’s not the greatest solution—see Jon’s frustrations with Microsoft’s ‘WEFT’ converter that produces EOT files from OpenType files (1, 2, 3) and web security analyst Chris Shiftlett’s thoughts on web fonts. There are a range of other and related issues that were discussed at a W3C meeting on EOT and font embedding. These problems will impact its chances of adoption as a standard. In 2007 Microsoft tried to push EOT to be part of CSS but this move was rejected, and they have since resubmitted it as a standalone submission. If EOT is accepted Microsoft would have to ‘open’ the technology to allow for the cross-platform creation of EOT files. This will greatly hinder the adoption of EOT by the W3C, and for good reasons. For EOT to truly catch on it would effectively have to be open sourced… …leaving that thought to die lonely at the bottom of a well. Perhaps I’m being somewhat too cynical; it would be a pleasant surprise if Microsoft were to open EOT.
Existing alternatives
Besides image replacement, there are a few other options that bring additional typefaces to the web which aim to retain text accessibility and degrade gracefully when viewed in unsupported user agents. One is sIFR and the second and more recent, , and VML">typeface.js. However neither is suitable for setting anything more than headings or a few lines; sIFR uses Flash, which isn’t a standard and quickly begins to affect page load times as well as resources; typeface.js meanwhile doesn’t allow the selection of the rendered text. These alternatives don’t compete with font linking if were to become a supported standard.
Finding opportunity in change
The technology implementation aside, direct font linking has raised licensing and copyright issues. Most type foundries that sell fonts, or rather sell licenses for fonts, naturally prohibit direct linking on the web. Foundries are afraid users will make copies of the linked fonts, putting an end to the foundry as the distributor. In the case of font linking, it would be simple for a user to extract the font of a web page: open the CSS, locate the @font-face definition and save the font file referenced in the src: property.

Type foundries are the product of a change in technology, specifically the invention of the movable type and the printing press. Foundries have for the past four and a half centuries thrived through further advances, such as lithography in the 19^th century, the invention of the typewriter and again adapted with the rising popularity of the personal computer in the 1980s. Gradually the entry barrier to printing was lowered. To print whatever you desired in 15^th–19^th centuries you either had to own a printing press or have some awesome friends; today anyone with a $1,000 computer and a $100 printer can print whatever they like and if you have access to public printing services (your public library for instance) you could even do it for free.

Type foundries (like other industries) were forced to adjust their business models with the invention of each new printing technology. Industries that don’t adapt generally whither, giving way to those that do. It’s silly to assume you can endlessly make money selling licenses to the same thing while everything about you changes. This is exactly what is happening here.

In the fear that their revenue streams will be undercut type foundries are reacting adversely to the topic of font linking and even font embedding. Instead of seeking new opportunities in the advent of these technologies, type foundries are reluctant to support web fonts. Illegal copying and sharing of typefaces is of course a bad thing for foundries and I empathise with them on that level, but I fail to empathise with them in regards to font embedding. Foundries need to realise that as typography becomes liberalised through the advancement of technology (namely the personal computer and the Internet), they must adapt and change with the rest of us. Respected typographer and web developer Richard Rutter wrote in July on the subject of font embedding with his article, The future of font embedding:

“…[I]t’s high time that font foundries and type designers stopped waving their hands in the air proclaiming the death of their industry, insisting that everyone will be pirating their fonts and installing them for free. Instead they should see this as an opportunity to be grabbed with both hands.”
Hosting & licensing service
Font embedding if implemented correctly as an open standard should be seen a new opportunity for foundries to sell fonts to a whole new market. One suggestion also advocated by Richard is for foundries to host embedded font files and sell them for use on websites. This would be in addition to the conventional model of selling licenses to weights and styles of a typeface individually or as a complete family based on the number of processors you intend to use them on (thankfully most foundries have realised the mistake in this and are now selling licenses per computer).

This web licensing and hosting service would be one dandy (and profitable) solution. I don’t have a clear-cut answer to this debate but ultimately in my experience the force of change is greater than any single proprietary interest and consequently always wins. Sometimes it takes a while and depending how involved parties react, things can get ugly which sucks for everyone. What excites me are those who find new opportunities that flourish with the new developments. I hope foundries see this as one of those opportunities and not the extinction of their industry—I look forward to ditching hacks in favour of standardised and open font embedding, because there are bigger fish to fry.

For anyone interested reading more, check out the W3C report on for and against standardising font embedding.
Whose Garamond is it anyway?

Posted: October 24th, 2008, 11:41pm CEST by Pascal

Flick through various foundry catalogs for a Garamond revival or adaption and you’re bound to discover more than garalde typefaces. Interspersed amongst the many Garamonds you’ll find erroneously titled baroque faces works by another type designer, Jean Jannon. I decided to investigate the affair and while doing so swept the dust off a little history of French printing.
Prelude: the birth of French printing
The European invention of letterpress printing with movable type by Johannes Gutenberg c. 1450 systematized the Latin alphabet into individual characters that could be physically composed and reused. Because Latin letters amalgamate into an alphabet—distinct from logographies and syllabaries—they are particularly well suited to be divided, cut, and finally cast into pieces of metal type that reside in wooden cases (fig. 1.). Majuscules (capital letters) lived in the “upper case” whilst the minuscules (small letters) inhabited the “lower case”. This is where we draw the synonyms for upper- and lowercase from.This was the basis for the revolutionary adoption in place of the meticulous copying of books by hand word-for-word and saw the establishment of an integral part of the printing trade: typography.

A reprint of a page of one of the Gutenberg Bibles, from Mainz, Germany. Gutenberg’s typefaces epitomize the strong gothic elements that we now classify as blackletter. The face in particular is a textura blackletter.

A sort is a single piece of metal type, a letter of one specific typeface and size. In digital typography it has been replaced by the term glyph as digital type does not physically exist until printed. Thus a sort of Bembo italic at 12 points is distinct from another of Bembo italic 10 points. Conversely those that semantically share the same letter but are stylistically different—even if of the same family and point size—are also classified as different sorts.Fig. 1. Metal sorts being set on a composition stick with many more organised in a job case underneath. Photo by Willi Heidelbach.

The trade of printing quickly spread throughout Europe. It was introduced to France in 1470 by Johann Heynlin and Guillaume Fichet—two professors of the Sorbonne, the historic university of Paris—who enlisted the aid of German printers to establish the first printing house. Three printers from Mainz helped them construct the presses, equipment and began cutting type. Still in the same year they printed France’s very first book, Gasparini Epistolae (“Letters”), written by the Italian grammarian Gasparino da Barizizza.
Traditional preferences
Gasparini Epistolae was set in a humanist blackletter (fig. 2). The face was a hybrid from a merging of roman and gothic elements. The style was rebuffed in favour of the traditional gothic properties of blackletter which French readers had grown accustomed to; unlike the humanist faces that were cut during the Italian Renaissance, humanist typefaces weren’t met as candidly in France as elsewhere in Europe. French typography remained moderately conservative, utilizing heavy blackletter faces of one style or another to set most printed material until the turn of the century.

Fig. 2. A sample of the humanist blackletter hybrid from Gasparini Epistolae, Haralambous, Y., 2007, Fonts & Encodings (p. 375), English edition, O’Reilly Media Inc., California, USA.

There are two noteworthy interludes between the end of the 15^th century and the cutting of garalde faces in the early 16^th century. In 1477 came the first book printed in French, an unusual and controversial venture as French was considered too vulgar to be set in print. Typically Latin was used for the written word, even in Germany at the time. It was set in cute bastarda (or Schwabacher) by Pasquier Bonhomme. Then in 1529 Geofrey de Troy, the personal printer to King François I, took the next steps that led to the broader adoption of roman with the printing of his book, Le Champ Fleury in which he put forth his theory that letter forms and human anatomy are closely linked.
The Garalades and Garamond
By the mid 16^thcentury printing was a solid industry. Certainly printing was still expensive, however it had become more economically viable and less time-consuming than the the previous option of employing scribes. Following the classical humanist typefaces, typographically in the 16^th century came the garalde (or old-style) faces, paying homage to Claude Garamond and Aldus Manutius. These featured a heavier weight and stronger emphasis on the downwards strokes than their antecedents (fig. 3). The weight can be attributed to a more oblique axis of the pen; garalde faces in no way loose their humanist elements. Today the best known garaldes faces are Garamond and Bembo, the former of which there are many versions of. Other notable digital renditions include Adobe Garamond, Granjon, Sabon, and Stempel Garamond.

Fig. 3. Top: 72 point Centaur, cut by Bruce Rogers in 1912–1914 and based on cuts by Nicholas Jenson made in the height of the Venetian Renaissance, 1469; bottom: 72 point Stempel Garamond, a true Garamondian revival by the Stempel Foundry, 1924. It was later digitized by Linotype.
The other Garamond
A century after Claude Garamond came Jean Jannon (1580–1658). Jannon was a French Protestant printer who began cutting type in the Protestant Academy in Sedan, France. He cut his type during the French Renaissance but did so illegally under the Catholic regime and consequently had his casts seized in 1641 by agents of the French crown, under orders of Cardinal Richelieu (who ironically used Jannon’s work to later to set his own memoirs, Principaux Points de la Foi). Jannon’s work sat locked-away for two centuries before seeing the light of day again. When they were uncovered they were misidentified as cuts by Claude Garamond, and hence named thereafter Garamond. Many digital revivals still carry on this error: ATF ‘Garamond’, Lanston ‘Garamond’, numerous versions of Monotype ‘Garamond’ and Simoncini ‘Garamond’. Jannon’s work is baroque in nature; it is easily distinguished from Claude Garmond’s; sharper serifs and an almost wild variation of axis and slope (fig. 4, 5).

Fig. 4. Top: 72 pt Monotype Garamond, digital revival based on Jean Jannon’s cuts; bottom: 72 pt Adobe Garamond, digitally revived by Robert Slimbach and based on Claude Garamond’s cuts.

Fig. 5. Top: 72 pt Simonici Garamond, another digital revival based on Jean Jannon’s cuts; bottom: 72 pt Stempel Garamond again, based on Claude Garamond’s cuts.
Historically befitting
So if you’re writing a piece on the introduction of French printing and want to set it in an appropriate typeface, a Garamond revival is most apt—of course you could pick a heavy gothic with which Johann Heynlin and Guillaume Fichet began France’s printing ventures but no one would comfortably read it today (and ultimately the blackletters are more Germanic than French). Conversely if you’re writing a piece set three centuries later, covering the French Renaisance, select one of Jannon’s revivals and pay tribute to a man almost forgotten by history.

Have a wonderful weekend everyone!
Times italic sorts

Posted: October 23rd, 2008, 1:28am CEST by Pascal

I’m still alive, just playing with sorts, that’s all.
Articles in limbo and ALA web survey

Posted: August 13th, 2008, 3:12pm CEST by Pascal

Whose Garamond is it anyway?
If you flick through various foundry catalogs for a Garamond revival or adaption, you’re probably bound to discover more than garalde typefaces. Interspersed amongst the many Garamonds by Claude Garamond (or Garamont) you’ll find baroque faces works by another type designer, Jean Jannon…

I wanted to cover this little typographic affair but quickly found the article gaining length and arguably bloat. It’s currently in limbo—I can’t decide whether to trim or polish it. Should I go ahead with a large but more interesting article (including a brief history French printing—which is interesting in my opinion) or go for the concise shorter version? Bah.

I also found that I’m short in garalde typefaces. Accompanying the article I’d prefer to offer garalde (and baroque) samples other than Garamond Premier Pro that’s on my Mac. Anyone with Bembo or another “Garamond” revival—either one by Garamond or Jean Jannon?
A List Apart The Survery, 2008

ALA is again running their annual survey calling all those who work in the process of creating websites. The survey closes on Tuesday the 26^th of August—have you taken part?
Putting Our Hot Heads Together
“If more of us are thinking ‘What can I contribute?’ instead of ‘Did I like this article?’ the entire conversation is transformed.”

Punctuation changes made to retain sequential quoting.

I also wanted to recommend article No. 265 Putting Our Hot Heads Together by Carolyn Wood. She offers a number of practical suggestions in nurturing collaboration in discussion and comment threads that appear obvious, but only after reading. Her writing style makes the piece a pleasure to read. Make sure to check that out.
Vivien’s review
In closing I also should pass on my thanks to Vivien from inspirationbit.com who kindly critiqued KLEPAS.ORG for me. Her thoughts and those that arose in the comments are going to help shape version 3 of this site; I’ve removed the clashing purple a:visited styling. Thanks again Vivien!
LugRadio Live and Silverback

Posted: August 2nd, 2008, 9:08pm CEST by Pascal

Yesterday’s entry regarding the great Australian Firewall was probably less interesting for anyone outside of Australia. The post was the result of a quick letter of concern that needed writing and consequently publishing (it’ll probably do more good online than in the inbox of the Minister I sent it to). That out of the way, I’ve been meaning to write about my time in England, in particular the visit to LugRadio Live ’08 UK.
LugRadio Live
LugRadio was the premier open source/free software podcast featuring the Internet’s Jono Bacon, famous web developer Stuart ‘Aq’ Langridge, and sysadmins Adam Sweet and Chris Proctor. The second season ended with a weekend live recording and conference, ‘LugRadio Live’, in Wolverhampton, UK.

Following the initial announcement that LugRadio were to conclude their show at the end of season five (to the dismay of the many listeners including myself), this year’s LugRadio Live (LRL) in the UK was to be the last ever given the podcast had come to an end. It was therefore a must attend.

LugRadio Live was literally a rocking event—rock music blared from the speakers half the time and there was a true “Chinny Raccoon” (LugRadio’s mascot), a custom made raccoon outfit as worn by Ben Thorp. Both the exhibitions and talks were all of a high calibre of which Bruno Bord’s talk “Baguette on Snails” probably being the top highlight. The show—the final episode of LugRadio—was of course also a good laugh, and for those who were by chance bored Bytemark’s portably gaming center ensured there was something to kill time with.

Of course the news is that due to the strong persuasion of fans LRL will continue to run annually despite the conclusion of the podcast. Given it’s the premier open source/free software event in the UK, it would have been a great loss to the community if this just past LRL was indeed the final one. To get a feel for what the event is like check out Flickr photos and make sure you try to make it next year.

Recordings of the talks should make their way online within the next few weeks. I presented on the Tango Desktop Project—particularly on the re-licensing into the public domain. I was originally going to demonstrate the creation of Tango-styled icons but decided in the last minute to alter the focus of the presentation. I think I disappointed some people with the move—if I make it to next year’s LRL I’ll submit another paper.
Silverback
I suspect that the word would have traveled far across the Intertubes that Silverback is now out and available (it was even mentioned by Leisa Reichelt at GUADEC), so I won’t mention it for long here. Silverback is a USD $49.95 program by Clearleft that turns a Mac into a mini portable usability testing lab—for the price it’s a must-have for anyone working in the fields of usability and accessibility in the software & web production industry. Although it doesn’t give you the amount of data that a specialised usability testing lab might provide, Silverback doesn’t cost an arm and a leg, it’s as portable as your Mac and suits every use case I’ve ever needed.

I did run into a small issue regarding licensing after my purchase: I used a friend’s credit card and with no option to purchase the license for another, my friend whose name was the credit card holder was the subsequent recipient licensee—not myself. For taxation purposes and simply for the sake of correctness I started a discussion on Clearleft’s GetSatisfaction.com pages regarding the issue. I received an email response within thirty-six hours of the post and had the issue sorted out in no time. Thanks to Sophie over at Clearleft for the help.

There are a number of reviews of Silverback—by Jonathan Christopher, .net magazine, and graphiceyedea—check them out if you’re unconvinced and otherwise have a great weekend.
Internet content filtering: outdated remedies applied to modern issues

Posted: August 1st, 2008, 3:31pm CEST by Pascal

Part of the hot-air during the Australian 2007 Federal Elections was the issue of Internet content filtering, which I waved off as just what I thought it was—hot-air—designed to impress and hopefully secure votes. The idea was shot down then and quickly was left alone … until now. Ashley Kyd drew this to my attention in his article Australia’s “Clean Feed”:

“The Australian Federal Government is powering full steam ahead on a $125.8 million dollar plan to enact a mandatory content-filter every single Australian Internet connection. Subtly different from the Great Firewall of China, the Australian version is to be implemented at ISP level, and will target not only illegal content, but also “inappropriate” content unsuitable for children.

The ACMA has submitted a report on the current state of the tech to Senator Stephen Conroy — so called ‘Minister for Broadband, Communications and the Digital Economy’ — who’s hailing it as a modern miracle, despite what can only be seen as a startling lack of comprehension of the real world implications.”

Punctuation changes made to retain sequential quoting.

So, I got off my butt and wrote an email to Stephen Conroy. Because I doubt it will ever get past his secretary and meet Mr. Conroy’s own eyes due to his likely busy schedule I am publishing this here in the hopes that any Australians—or even Australian permanent residents (I am one)—who see this might consider voicing their concerns to the Government. The letter:

Dear Minister Stephen Conroy,

As an employed Australian permanent resident—and web & Internet professional—I have serious reservations about your new mandatory “clean feed” filter initiative.

Pursuing an expensive $125.8 ~~billion~~ million dollar venture that entirely fails to address the security, speed, quality of any such a filtering system, as well as the rights of both website owners in and outside of Australia and finally the civil rights of the citizens of Australia is a waste of tax payers’ money and of serious concern to every Internet user in Australia.

Concerning security, I am shocked that the report claims six of the seven solutions tested filter the HTTPS protocol (HyperText Transfer Protocol over Secure Socket Layer)—the protocol over which Australians make secure Internet banking and e-commerce transactions. This communication protocol was inherently designed specifically to promise a set level of security and privacy for its users. Furthermore, as this is a secure protocol, any filter—by logical conclusion—must use a simple keyword, or “blacklist” method which has been proven both ineffective, slow, incredibly arduous to keep updated and also prone to blacklist (block) legitimate content.

In regards to speed, the July 2008 study clearly indicates that a high-speed, accurate filtering solution is not possible. Therefore implementing any such filtering at the ISP (Internet Service Provider) level will only hamper and slow down the current infrastructure—which greatly needs speed improvements, not declines.

As noted above a blacklist filter is of serious concern to security and speed, both of which are important elements of any such a system’s quality level. The higher the blocking accuracy of such a system, the more it well hamper the speed of the network infrastructure.

If such a system is implemented I believe the Australian Government will anger and frustrate many website owners both overseas and locally. Processing blacklist removal requests will be a major undertaking that will incur continuous monetary, time and implementation costs. I believe many web content owners will consider their legal options in regards any such a move by the Government. At the very least a one-size-fits-all filtering approach as the one proposed will cast the Australian Government in a very negative light, abroad and at home.

Finally—any most importantly—I believe the Australian Government has absolutely no role in monitoring and deciding what is appropriate for myself, our children—a fact that many Australians will undoubtedly agree with.

I believe the Internet is a technology that will test society and its legal systems—common issues that have arisen include content censorship (as on the table currently), copyright, licensing and patenting, universal access availability, and more. Many of our legal systems are outdated in respect to the advances of the digital age, and this current issue is just another hurdle to be overcome as we realise we must shift our attitudes; no body will be successfully able to control and filter the expanse of information present and growing on the Internet. Any such a venture is futile, expensive and will only cause distress.

The expansive resources—125.8 ~~billion~~ million dollars for example—available to the Australian Government instead be focused on providing up-to-date education on Internet technologies both to students and parents to ensure we can make well-educated judgements about how we spend our time on-line.

Our digital infrastructure has been the hot-topic before, during and now as it seems even after the election—I sincerely hope the Government will reconsider this initiative and make an informed decision that takes the needs of Australians at heart.

Your sincerely,

Simon Pascal Klein
Canberra, Australia
Opera Web Standards Curriculum

Posted: July 18th, 2008, 12:57am CEST by Pascal

Opera recently released a set of succinct educational articles on the foundation of the web, standard technologies and well-written introductions to vital web design topics, aptly titled the Opera Web Standards Curriculum (WSC).

At first I asked myself why something like this was necessary in the year of 2008—don’t students get taught web standards through college and univeristy these days? If not, aren’t there already are number of good resources available, free and commerical? I had a bit of a read, starting right at the beginning and found partially the answer in the fifth article:

“…The general feeling I get from looking at numerous university courses is that web languages like HTML, CSS and JavaScript are below the technical threshold of most computer science programs, and above the technical threshold of most MIS/New Media programs.

What I’m getting at here is that many educational courses don’t cover this kind of stuff in any great level of detail. I would be willing to wager that if you ask 10 developers that work with web standards where they learned how to use web standards that 9 of them would reply that they are self-taught (the other 1 won’t answer you because she’s too busy trying to get her site to render properly in IE6).”

Previously defined acronyms earlier in the quoted text have been redefined.

I started an IT course in college that focused on publishing content onto web-based mediums in early 2006. I was told we would not touch CSS until the end of the second year, and mention of web standards was not to found in the entire curriculum document. I promptly left the course. As Jonathan Lane points out, most standards-based web designers and developers are probably self-taught—I am.

But why another set of articles ontop of the many resources that already exist? Because commonly it’s scattered across the Internet or because you’d need to purchase books (and in the case of students, those university text books can be quite pricey). The neat thing about web design and developement is that articles like those from the WSC appear in the same canvas we work on and consequently offer interactive examples—and best of all are freely accessible. Books can be nice—particularly well set ones—but for someone just starting with web standards beginning with the WSC marks a good beginning.

The WSC doesn’t go into great technical detail. Instead, it offers the background information that is often omitted from the more technical articles on developer blogs or on related publications. Something like this would have been great in place of the IT course I entered a few years ago.

Apart from being an introduction to the creation of the Internet, how it came to be what it is now and why standards are good for everyone, the WSC also features introductory articles on information architecture, accessibility, usability and content context, colour theory, wireframing, assorted mockup tips and last (and best of all), an article on web typography.

The Opera WSC will probably be my future recommendation for newcomers to web standards, HTML, and the selected articles above. I would even recommend them to the experts—pick a topic you haven’t dealt with in a while and make sure you’re up to scratch (there are a few questions at the end of each article).

Kudos to Opera.
Tango to be public domain?

Posted: July 14th, 2008, 11:31am CEST by Pascal
In the previous post I hinted that the Tango Desktop Project icon set would be re-licensed into the public domain. The news came through at the start of GUADEC 2008 over the tango-artists mailing list. This somewhat lengthy article should set straight what Tango is, what it’s goals are, some of the hiccups that we’ve encountered along the way and the imminent future of a possible re-licensing into the public domain.

Fig. 1 Example of the 16×16 Tango icons
Tango?
The Tango icon theme was the first icon theme to follow the Tango style guide and use the freedesktop.org icon naming specification. The style guide was introduced to give artists a guideline to work from so that icons across the free desktop would have a unified and clean look, but also adhere to a common and well-defined set of visual metaphors. The naming specification which was an attempt to create a standard method of naming icons in an icon set such that various desktop environments and programs (GNOME, Xfce and even KDE) could use them, battling the otherwise gratuitous renaming of files.

The idea behind an icon set is to avoid redundancy. For example the “save” icon would be consistent across all applications that utilize the save function, and would only require one icon—in this case document-save, by the icon naming specification. Before each application would ship their own icons, which can cause redundancy and also forbids the user from changing the style of their desktop through different icon sets. Many applications still ship their own icons. Applications that ship their own hard-coded icons do so because there isn’t an icon theme-ing feature that would find the appropriate icon for the given feature for that desktop environment (or at least not without third-party theme-ing programs), because the developer does not know a theme and icon specification exists or because there is no entry for that function in the icon specification.
Licensing issues
The Tango icon theme was licensed under the Creative Commons Attribution-ShareAlike 2.5 License, which promoted the usage and editing of the icons, provided all use cases and derivatives were documented with an original author note and generally a link-back to the Tango Desktop Project homepage. In addition, derivatives had to be licensed under the same license as the original work i.e. derivative works of the Tango icons would also have to be licensed under the CC Attribution-ShareAlike 2.5 license.

Everything was dandy. The project received a good share of media attention and promotion—it even was on Slashdot a few times. After a while though some issues started to arise:
- The Free Software Foundation indicated that the CC Attribution-ShareAlike 2.5 license and their GNU General Public License (GPL) was not compatible—developers could not ship Tango icons with their GPL-licensed code. Furthermore the Debian Free Software Guidelines (DFSG) noted the same thing, and would shift free software shipping with Tango icons or derivatives of Tango icons (such as custom icons created for features that weren’t covered by the icon specification) from the general repository to the non-free repository.
- Artists could not mix Tango icons with icons from the GNOME icon theme that was being redone to follow the Tango style guide, as the GNOME icon theme is licensed under the GPL. I experienced some of the licensing issues myself as I created icons for Bazaar’s graphical front-end application, Olive (bzr-gtk) last year. (Bazaar is a distributed version control system.)
- Through the media attention Tango received, numerous websites including professional business websites and/or their software began using the Tango icons without attribution, in some cases fully claiming the icons as their work.
Tango and the public domain
Once the Tango icon theme was complete—i.e. there were icons at all required sizes for each entry in the icon specification—it was essentially deprecated. We recommended artists and developers alike to use and build on the GNOME icon theme, as it was licensed under the GPL; we cared less about third parties that wanted to use the icons on their website and in their products—after all the Tango icon theme was still there for them and all they were required to do was provide attribution and share-alike (as before).

Until now we have just been promoting Tango as an example icon set that complies to the icon naming specification and is based on the Tango style guide. However, this might now change! With the licensing issues we have for a while discussed the possibility of re-licensing the Tango icon theme to deal with the above mentioned problems. This was until now considered an arduous and unlikely task as numerous artists had contributed to the theme and each would have to agree to re-license their respective contributions (an issue that many larger free software projects face when re-licensing) and in particularly because some of the artists were working for Novell at the time of their contributions (who subsequently owned the rights to the work). Jakub Steiner had a long debate with the Novell legal and open source review board who have just agreed to re-license its share of the theme into the public domain (much of which was performed by Jakub himself, as he worked for Novell; kudos).
Implications
This fixes all of the licensing issues, but raised the question of usage. Ideally it would be nice to be attributed in some respect however legally anyone can claim authorship over the icons in derivatives and even simple use cases. I don’t think this is much of an issue as a large number of use cases lacked attribution in the first place and this was almost impossible to combat—we don’t want to hunt down unattributed use cases or derivativesy, and ultimately a library of free icons could still give Tango and free software a lot of public attention.

I discussed the re-licensing briefly with Andy Fitzsimon at GUADEC and he noted this would be great as there would be a large pool of public domain shapes and elements to draw from for future icons. Rodney Dawes made a similar observation and recommended we shouldn’t release the icon set as-is into the public domain, instead:

“…create a new repository which houses the Public Domain elements which make up the icons we offer in tango-icon-theme. This new module could be tango-icon-assets or something similar, and would contain just the assets in SVG form (in the one-canvas workflow style), such as ‘paper sheet’ and ‘error emblem’. Everything in this module would be Public Domain, and there would be no need for any build system work at all, as it is purely a repository of building blocks for creating full icons.

From there, we can make tango-icon-theme into an LGPL icon theme, made up of the public domain assets. This module would provide a clearer licensing scheme, for developers to be confident about using our icons. This would also provide a nice method of allowing contributors to submit icons based off the tango assets for inclusion, but who do not wish to relinquish copyright in doing so. There icons could be LGPL, yet the parts that comprise them, could be Public Domain, save for minor differences.

We could also pull over some of the Public Domain assets into other icons and themes, such as OpenOffice.org and gnome-icon-theme, without issue. The paper sheet for example would be a great place to start doing that.”

Tango’s future
Jakub has emailed all the other contributors of the Tango icon theme requesting their response regarding the re-license, and thus far everyone seems in agreement with Jakub’s and Rodney’s proposal. So, look forward to a public domain Tango assets library and TangoNG—a “next generation” Tango icon theme, better than the first!
GUADEC 2008, Istanbul

Posted: July 12th, 2008, 11:53pm CEST by Pascal
This marks the first journal entry from Istanbul! I’ve been here in Turkey for the past week attending GUADEC, which for the whole part has been a blast. It’s as community-focused as LCA—the largest free software conference in the southern hemisphere—and thus a whole heap of fun.

GUADEC is the premier GNOME conference, held in a different European city every year, where developers and users alike chat about GNOME and generally party a lot. GNOME is a free desktop environment built on the GTK+ toolkit.

Some highlights include:
- meeting Hylke Bons (hbons), Andreas Nilson (andreasn) and Kalle Persson (kallepersson) for the first time face-to-face—we’ve known each other for a good year through the Tango Desktop Project—it’s great putting a face to an IRC nick;
- catching up and having dinner with a different group of like-minded, awesome and friendly people every night at various restaurants on the Bosphorus;
- the three-hour boat party sponsored by Collabora on Thursday evening;
- Federico Mena-Quintero’s talk on document-centric GNOME where he proposed a journal-based history application—although I disagree with the implementation it caused quite a buzz and got the ideas flowing;
- Catching up with Lennart Poettering and watching him haggling and forcing taxi drivers to bid each other down;
- post-GUADEC dinner under Lefty’s guidance at a Japanese restaurant followed by a water-pipe smoking session whilst sipping apple tea in a back-alley café in Taxim.
I fly out on Monday before heading to London for LugRadio Live with a day in between back in Frankfurt to rest and recover. Between doors until the 24^th—should be fun!

On a final note, the Tango Desktop Project icons have been re-licensed into the public domain. I’ll do a proper writeup in the coming days on the implications and what this means for developers and artists.
Tintin and Lorem ipsum

Posted: June 29th, 2008, 2:04am CEST by Pascal

I was re-watching Hergé’s The Adventures of Tintin a while back and noticed the animators set the bodies in newspapers in Lorem ipsum as filler text. Lorem ispum is a placeholder or filler text used in typography, design and other publishing as means of testing the visual representation. Lorem ipsum features roughly the same letter distribution as in English.

Tintin and Snowie were favourite heroes of my childhood. This definitely made me smile.
Introducing Typenuts

Posted: June 25th, 2008, 4:16pm CEST by Pascal

A few weeks back I had the chance to page gently through an old German Lutheran bible from the mid 18^th century. I used the opportunity, quickly filling my camera’s SD card with photos of the bible and other things from my friend’s collection.

I noted I was thinking of starting a photoblog for many of these images to live and with an iPhoto typography collection getting larger and larger I figured I’d share the cream of the lot on a separate site.

With that, I’d like to introduce Typenuts, a small typography-related ~~photo~~ image blog (non-photographic images might make their way onto the site too). I’m planning to update it on a weekly basis, probably on the weekend. Logo is pending too—I’m thinking a squirrel with an acorn composed of the letters “typenuts” in small capitals or something similar….

Right now though Typenuts has it’s own yummy RSS feed and comes in two themes—a light and a dark theme allowing you to switch between them on the fly to best suit the image at hand. I am torn over which theme to set as the default—light or dark—what do you think?

Type on Typenuts is set in Minion, as designed by Robert Slimbach, 1990.
Give your content context

Posted: June 19th, 2008, 11:46pm CEST by Pascal

Giving links purpose such that they can be understood when read out of context—a feature many screen readers offer—has long been an accessibility recommendation. It’s comes as part of many accessibility guidelines, including W3C’s WCAG 1.0 and of course also remains part of WCAG 2.0 (§ 2.4).

A common mistake in thinking when approaching accessibility is that the work and time will only benefit a small number of your users. Quite the opposite: healthy font sizes, good contrast between elements and good descriptions for multimedia will benefit many more of your users.

This recommendation does not only come from accessibility gurus; information architects think long and hard about the most appropriate and politically correct attributes for labels—it’s a must when concerning contextual links to ensure users end up where they expect to.

“To be more representational of the content they connect to, contextual links rely instead upon, naturally, context. If the content’s author succeeds at establishing that context in his writing, then the label draws meaning from its surrounding text. If he doesn’t, the label looses its representational value, and users are more likely to experience occasionally rude surprises.”

Links aren’t the only labels that could do with a good description. Section titles, headings and sub headings in particular could benefit from good phraseology as these label the content that follows them. Users look to labels for structure as they browse your content, in particular helping skim readers who may be looking for a small scrap of information buried deep on your page.

This responsibility however often falls from information architects to authors. Style guides (e.g. names should always link to profile cards) and aesthetic visual distinctions for content markers through good typography (known as hierarchy) are the best ways of supporting authors in this process.

For more on hierarchy and web typography see A Guide to Web Typography on iLT.

Following this thinking, your can further add context to your links, labels and even obscure content through the page design and graphics. It may seem obvious to you that your website deals with your product or hobby topic, but consider a user landing on a page deep within your website. With strong direct linking to very specific content you should not rely on the author who provided the link to ensure the user knows exactly what they are in for.
Understanding community is paramount to the success of your Open Source project

Posted: June 13th, 2008, 11:39am CEST by Pascal
LugRadio recently featured a section on “community” in their latest episode—Finding Emo—discussing how many individuals and companies alike endeavor to form communities around their projects and products, and how in the end many of these fail. In particular with Jono Bacon on team it is a noteworthy listen (segment starts at approx. 27:00). I think the only thing that was missing at the end of the rather long segment was a conclusion of thoughts by Jono. Here’s my take, in order:
1. Don’t be a no-one: in order to start doing something yourself you’re the only one in need of convincing and inspiring that that something it’s in fact worth doing. Conversely, acquiring a community of contributers (developers, documentation writers, usability nutters and crazy “orange-sunglass-wearing” artists like myself) requires a bit more. To ensure others will jump onboard when you launch your project build a positive presence in an open source community; being known for your work gives you an air of trust and commitment that inspires others to lend an ear when you have something to say.
2. Be involved and listen: this should almost come automatically as part of having a presence in a project and as part of a community, but the last part is what really needs pointing out. Sometimes you have more to gain from listening to others than speaking yourself (and this applies far outside of open source communities). Listening will let help you understand the community you are apart of and the people within it, who ultimately are the easiest and possibly largest pool of possible contributers you can reach out to (at least at launch).
3. Become part of the community and launch from it: it is easier setting a stage within your community and communicating directly to its participants who already contribute to an open source project than from somewhere outside. Still address the world at large and invite anyone who is interested, but do it from a stage where you’re bound to have an audience.
These apply as much to companies that want to extend their products and build their services whilst contributing to open source as to individuals who just want to start their own project but could do with a few extra hands. Too many companies simply put up a web page offering links to a public version control repository and hope a mad flock of developers will form around their project.

Don’t be a no-one, get involved, and listen so that when you’re ready to kick-off your project close to the community you’ve become a part of you will pique the interest of contributors.
Out comes the inner-child

Posted: June 10th, 2008, 3:55pm CEST by Pascal

I’ve had an excuse to let my inner-child out with my recent tongue piercing as solid foods were off the menu for the past week. So at our recent visit to the supermarket I stashed up on soups and the like—perfect excuse to get some alphabet noodles.

Of course silliness always ensues with alphabet noodles—I was reading an article over lunch today on iLT:

Yum.
Comic Jens; no more excuses

Posted: June 9th, 2008, 4:58pm CEST by Pascal

Found this via ILT’s latest Sunday Type from fontblog.de:

Comic Jens is a looser and more aesthetic alternative to Comic Sans by Jens Kutilek. With a wholesome set of 303 characters in regular and bold, all available freely under a Creative Commons license in OpenType format no one has an excuse to use Comic Sans anymore.

Check out the cute extra glyphs:

The font also takes advantage of OpenType ligature features:

Download it from the Intertubes today!

Supported languages include Albanian, Basque, Catalan, Cornish, Danish, Dutch, English, Estonian, Faroese, Finnish, French, Galician, German, Greek, Icelandic, Indonesian, Irish, Italian, Malay, Manx, Norwegian Bokmål, Norwegian Nynorsk, Oromo, Portuguese, Somali, Spanish, Swahili, Swedish and Turkish.
Addendum
In retrospect I have to say that when setting body copy (even for only a poster) the follow-through of the cross/strikes on the “f” and “t” come to irritate reading slightly, particularly noticeable in the bold weight.
Biblia anno 1736

Posted: June 6th, 2008, 9:35pm CEST by Pascal

Turning the pages of a 250-year old book has a uniqueness to it that makes it quite different to flipping through what normally sits on by bedside table. Last week a friend kindly gave me the opportunity to peruse a German Lutheran bible, hand made for a Thomas CI Aussen (the “A” could potentially be another character) and completed in 1736.

Condition
The bible belongs to my friend’s collection of various antiquities, many of which hold within their covers (and sometimes even outside) scores of typographic treasures. The book itself is still in fairly good condition. It looks to have suffered some light water damage, wearing and folding of pages, namely the opening and closing pages. Further into the book are a few burn marks. I’d hazard that these were probably caused by a stray flying ember that landed on the page whilst its owner would have been reading it by firelight.

Some of the ink has faded, less in body text and more so in the versals and other ink-heavy areas. This is likely to be caused by the heavier saturation of ink on a relatively small area of the page and the subsequent weakening of the paper on that spot. This is usually exacerbated by pages not drying properly but in this case probably more the cause of just its age.

The binding is in good order, but careful handling never goes astray with older texts. The book is enclosed by wooden covers and protected by imprinted leather.
Typography
Type in mind, the first thing that stands out is the heavy blackletter. The bible would have been printed with a letterpress, undoubtedly not much different to that the one Gutenberg invented roughly two centuries earlier. I tried to identify the typeface or at least the type of blackletter used but was thrown off; the main body copy features different typefaces in its composition—I have a feeling the printer ran out of sorts. A sort is a single metal glyph of a given point size and style using in letterpress printing. This is most obvious in the most used characters, note particularly the characters “d”, “o” and “v”:

The two faces include a bastarda and a fraktur (see the rounded bowl of the “d” and shape of the “o”).

Capital letters amongst the body text are set quite ornately. Nouns and otherwise words that syntactically function as nouns are in the German orthography capitalised and this makes for a larger presence of capital letters in the running body of text than compared to, say modern English. Setting all of these ornately amongst a fairly thick and strong blackletter seems to interrupt rather than aid reading.

Margins are ruled (I’m not sure whether by hand or by the press), which today is quite unusual. I noticed also that the the versals are quite exquisite, and vary throughout the book in complexity and size.

Drawings and other large-scale artwork all received pages of their own and are incredible pieces of work. These would have most likely been woodcut relief printed, meaning the entire piece would have to be “negatively” carved into a block of wood at full size. The drawings were also, probably by some mistake, printed on slightly smaller paper and in many cased preserved better than those the larger surrounding ones.

And that’s it. Have a great weekend (und viele grüße)!
Versal letters on the web

Posted: May 26th, 2008, 5:30pm CEST by Pascal

Because it’s type-related I couldn’t resist waiting another day before publishing this…

…Yesterday I was going through an old un-named sketch book. Right at the back were these beautiful hand-drawn versal letters and an entire page of weaving ornaments:

Versals, drop capitals or raised capitals are large and often ornately decorated capital letters marking the opening of a running body of text or a verse.

Modern typography has discarded much of these classical decorations in favour of a more clinical page; these particular ornaments above aren’t as multicultural and certainly not as neutral as modern typography which aims to better cater for a larger and more multicultural audience.

I would like to see a revival of this form of decoration; I think a lot can be done with typography that currently scores of artists—particularly web designers—are trying to achieve with stock photography and ever popular vibrant, reflective and glossy elements.

Setting versals in CSS isn’t particularly hard. A common method is to use the pseudo-selector :first-letter to refer to the first letter of an element. You can combine this with other selectors to refer to the first letter of the first paragraph of a page, heading, or whatever else strikes your fancy (utilizing the DOM structure properly is helpful in making this work properly). Note Internet Explorer doesn’t recognise many of the CSS 2.1 selectors (IE does however understand :first-letter).

Another common method is to just manually wrap the first letter in span tags with a specific class of desired styling. Whatever you choose, the styling might look something like this:
div#content h2.entry-title + p:first-letter { font-size: 4.2em; float: left; display: in-line; text-transform: uppercase; }

You may also want to set margins and line-height to better place the versal amongst the rest of the paragraph. The float: left; is optional—it wraps the paragraph around the versal—if you want the versal to stand on the baseline of the opening line of the paragraph then don’t set the float.

Targeting the first letter and styling it was simple; browser consistency is the hair-pulling part. Internet Explorer in particular complicates things because it doesn’t understand the preceding selector (element + first-child). If your versals are vital to the page, see Bill Weaver’s article, Wrangling Drop Caps, which goes into closer detail on the subject including a JavaScript method.

When printing a page in Internet Explorer featuring a versal specified using the pseudo-selector :first-letter IE bizarrely displaces other floated elements in proximity to the paragraph. See Nicholas Rougeux’s article Drop caps in the wild for an example.

Finding versal letters with further decoration done other than through size and placement is a rarity on the web. Replacing the first letter entirely with a graphic isn’t impossible but you’re bound to run into many accessibility issues using image replacement on the fly (and getting the right letter—that means designing twenty four versals and some scripting). I did however see a simple method last week of adding a bit more to versals by a bright Polish designer:

Feeds

KLEPAS.ORG (20 unread)

Posted: January 4th, 2009, 7:14am CET by Pascal

Posted: December 14th, 2008, 4:01am CET by Pascal

Posted: November 13th, 2008, 12:40am CET by Pascal

Posted: October 24th, 2008, 11:41pm CEST by Pascal

Posted: October 23rd, 2008, 1:28am CEST by Pascal

Posted: August 13th, 2008, 3:12pm CEST by Pascal

Posted: August 2nd, 2008, 9:08pm CEST by Pascal

Posted: August 1st, 2008, 3:31pm CEST by Pascal

Posted: July 18th, 2008, 12:57am CEST by Pascal

Posted: July 14th, 2008, 11:31am CEST by Pascal

Posted: July 12th, 2008, 11:53pm CEST by Pascal

Posted: June 29th, 2008, 2:04am CEST by Pascal

Posted: June 25th, 2008, 4:16pm CEST by Pascal

Posted: June 19th, 2008, 11:46pm CEST by Pascal

Posted: June 13th, 2008, 11:39am CEST by Pascal

Posted: June 10th, 2008, 3:55pm CEST by Pascal

Posted: June 9th, 2008, 4:58pm CEST by Pascal

Posted: June 6th, 2008, 9:35pm CEST by Pascal

Posted: May 26th, 2008, 5:30pm CEST by Pascal