I am now proudly powered by WordPress1.3. This version is still not in beta, so before you rush off to download it be aware that it is not quite ready for public consumption yet. I was intrested in testing some of its new features and thought the the redesign of this place was as good a place as any to give it a spin.

For anyone subscribed to feeds here, my apologies for the hideous mess you will have experienced this evening. I have no doubt that there are better ways of implementing a site wide CMS and CSS update, but live editing felt so much more exciting.

Version 2.0 has been christened Dockshite and nods to a building in Docklands, London, that I spent a bit of time in leading up to Christmas. It was tall, minimalist, and very shiny. All the graphics come from one photo taken in a glass lift.

Incidentally, if anyone has strong feelings about some of the content that has gone missing from The Dredge, drop me a lineand I will port some of it in to the new database. If not, I won’t bother because I’m not really missing it.

Some facts about sIFR: not up for debate

• sIFR does not require any changes to the (X)HTML code, all the work is done by Javascript, Flash and CSS.
• If the user does not have Flash installed or Javascript enabled then the (X)HTML text is displayed and styled by CSS.
• sIFR is scalable, and when rendered will adjust to the users font size settings.
• sIFR is compatible with all screen readers. No problems or issues have ever been reported.
• sIFR text is selectable with the mouse, although visual confirmation of the selection may be absent when selected with body text.
• sIFR does not affect search engine placement or ranking, nor does it hide textual content from search engines or users.

The conclusion should be that sIFR is an accessible, mature technology, and that its use must be given serious consideration by designers and developers.

When to use sIFR

Like all web technologies, it is important to understand the best way to use sIFR and to realise the situations in which it is most appropriately used. It’s about choosing the right tool for the job; and there is one job in particular where sIFR leaps from the toolbox and begs us to give it a go.

The scenario: A large sports news website decides to style all their headlines in a unique corporate typeface. News stories (with their headlines) are published by different people, in different places, all over the world, via some sort of content management software. They can’t employ someone to sit in front of Photoshop and create a headline image every time an editor wants to add a news story!

In this situation, sIFR is a tool that is unrivalled in simplicity, accessibility, and extensibility. Some news sites implement a solution involving the building of images on the fly via PHP or other server side jiggery-pokery. This works well as a time saver, but has many disadvantages when compared to sIFR:

• Images will not scale to the users default font size.
• Although images should be cached on the server, there is still a performance issue in generating the images.
• Each image must be downloaded separately creating server and bandwidth concerns.

With sIFR, all that is required is one Flash (.swf) file and one Javascript (.js) file to be downloaded allowing all headlines on the site to be rendered in the corporate typeface.

This example was not just plucked from the air. It was the exact scenario that led to Mike Davidson developing the original Flash replacement technique back in 2001, for a redesign of ESPN.com. Since then, with the input of Shaun Inman and others, it has grown in to the rounded and fluent technology we have today, and will likely have a major impact on web typography during 2005.

Using sIFR for links

The latest version of sIFR allows for linked text to be replaced. Although this is an exciting development it is almost always not appropriate to use it in such a context. This is due to the following accessibility issues:

• No browser right-click (contextual menu)
• No option/apple click
• No status bar information

Although these appear to be trivial issues, many people find broken functionality of these features quite frustrating. With the lack of status bar information it is impossible to get a clue as to where you may be heading next; and with the popularity of browsers such as Firefox and Opera, the contextual right click is becoming an increasingly used tool. Although sIFR provides a basic right click on links, the browser’s contextual options are unavailable.

Of course, these are limitations with Flash and not with sIFR itself. It is likely that some of these problems will be overcome in the future. The status bar for example can be controlled via Javascript, so it is not too hard to imagine some sort of functionality being added to display link targets. However, until Flash allows for a full browser contextual menu on links, I believe that sIFR is not quite viable for tackling this kind of text.

Anti-aliasing

While most of the interest in sIFR is centred around the ability to serve custom fonts, an important consideration is also the fact that Flash text can be anti-aliased. Web developers often forget this, partly because so many are working with Mac OS X and its impressive Quartz font smoothing. However, Windows users (despite having the ClearType option buried in a display menu somewhere) are not likely to have anti-aliasing enabled, and the ability to have anti-aliased headlines shown to these users as well as pre Windows XP and Mac OS X users is an important consideration.

Fine tuning

One issue that I have heard mentioned a few times, is that sIFR does not allow for the fine control of text that is possible with an image. This is perfectly true. With an image you have created in Photoshop or Fireworks, you can exactly control kerning, stretching, anti-aliasing, or any other features such as drop shadows very accurately. A graphic’s editor is a true WYSIWYG medium. sIFR rendered Flash is not.

If this level of control is required then an image is still the best way to deliver the text, and sIFR is not the right tool for these situations. However, if the requirement is purely to deliver a custom typeface, then sIFR beats creating images hands down.

Loading speed

The rendering speed of the replaced text is an important consideration when using sIFR, and although speeds have been significantly increased since early versions, there can still be a distinct delay if there are many instances of sIFR on screen at one time. This is perhaps the best example of why using sIFR in moderation (ie. one headline per page, or headline per post) is currently the ideal way to use the technique.

This is one of the most frustrating drawbacks of using Flash replacement. The temptation when first implementing sIFR is to replace too many elements on the one page. For this to be viable, loading speeds have to be greatly increased; and although a nice nippy server will help you out, the true culprit is the time that the rather bulky Javascript takes to run.

The conclusion

sIFR is not a rival to image replacement; it is a separate tool that does a different job. It is best used with text that would otherwise be displayed in one of the browser’s default fonts, and not to replace custom built graphics.

The ideal situation for sIFR is when you would otherwise use an image solely to display a custom typeface or anti-aliased headline. This is done very frequently on the web, and sIFR is by far a better option in these cases. It scales to the users default font size, is selectable, degrades gracefully beautifully, and can be used on thousands of pages with only two files required for download.

Summary

• Use sIFR for page titles or headlines.
• Keep use of sIFR limited to optimise loading times.
• Don’t use sIFR for links.

Useful links

• Mike Davidson: Introducing sIFR: The Healthy Alternative to Browser Text
• Mike Davidson: sIFR 2.0 RC 3 is Here… Come Get Some
• Dave Shea on sIFR
  

So here it is: another article on font sizing.

This one however, is not intended to expose the pros and cons of various sizing methods. In fact, it is not going to address any particular methods of font sizing at all. What it is intended to be is a base for further discussion. A central hub if you like, with many a spoke thrusting outwards to create a spinning wheel of controversy.

I want to talk about designers that lower their body text below the default font size. These ill-mannered, uncouth, hooligans that neglect standards, fly in the face of accessibility, and… oh, hang on, I’m one of them. Damn.

What’s going on?

I thought I was better than that. I thought I had a grasp of web standards that was allowing me to deliver tight semantic code, bandwidth saving CSS, and accessible sites to all. So why have I (and if you want to admit to it too, now is the time to do so) ignored one of the most obvious premises of usability?

First, a little history

From 2000 onwards all the leading browser makers adopted a default font size of 16 pixels (px). Before this time, default font sizes did not match up across operating systems, causing havoc with designers’ patience and sanity. So the introduction of a true standard default was hailed as a pretty good day for the web. Developers felt this would help solve the problem of delivering consistent font sizes across platforms, and there was much rejoicing by all.

Unfortunately it didn’t really work out that way. The 16px default font size is rather large. Not huge, but just a little too large for comfortable reading on an average resolution monitor. Macintosh users in particular were used to a smaller default. They took one glance at the new size and promptly switched their default back to the 12px they were used to seeing. All that good work, all that exciting promise: undone in three clicks of a mouse.

Meanwhile, web designers where looking at their pages and thinking, hmm… that’s a rather large font size for body text; I think I’ll drop it by about 10%.

With both groups (designers and users) lowering the font sizes, the result was inconsistency, confusion, illegible text, and a whole mess as big as the one we started with in the first place.

What’s the solution?

The true solution is obvious, surely? Either users or designers stop lowering default font sizes. Users shouldn’t have to do that of course. They want control over their web pages; to see them at a size that is comfortable (or indeed necessary for visually impaired users), without having to adjust the font size for every different site they come across.

So that leaves us. Designers. Simple really.

Every accessibility and usability rule in the book tells us to let the user control the experience. If they want to lower their default size, then let them. But don’t do it for them!

Right. Who wants to change?

The short answer to that is: no one.

I have spent the last 3 months developing this site, UsableType. And after not much soul searching and discussion, I decided to lower font sizes from the default size. Why did I do this?

For one reason: the site looked rubbish in every browser I tested on; and I wanted the site to look good.

That is the logic behind nearly every designers choice to lower body font sizes. It just doesn’t look good with line after line of the default font size. Accessiblity and usability can go hang… it just doesn’t look good!

Dirk Knemeyer recently published an article for Digital Web Magazine, called The End of Usability Culture, in which he claimed that the age of usability driven design was over. Although his argument was based on more general principals of design, in terms of font sizing his point is demonstrated very well indeed. Designers are reducing the level of usability in exchange for increased typographical control. In the same way as some designers set font sizes in pixels (despite their inaccessiblity in IE/Win), the bulk of desingers have decided that the pros of font size control outweigh the cons of reduced accessiblity.

A new default size

Downsized body text is so common on the web these days that we have gone too far to ever correct the problem. Users expect to see their default font size looking smaller than it is set in their browsers – whether they aware of the fact or not. Designers have forced a new default themselves by being so adamant that they want it smaller. It is now the norm for text to be set at anywhere between 70% and 80% for the <body> element and then other font sizes for title, etc, scaled from that base size.

As for users that want to see text larger, they will have figured out how to change the default on their machine and are probably browsing away happily as we speak. So it’s not all bad news, and although it is still not an ideal situation, the issue has settled a lot over the last few years. Designers have defninitely beaten the usability purists over the issue of font sizes. It is rare that you come across sites with default sized body text, and what’s more, my use of text-zoom is deployed only very late at night to reduce eye strain – so we must be getting something right!

Whether the purists like it or not, downsized body text is here to stay, and as designers we should no longer feel ashamed about our wish to present text at the size we want.

Until the <font> tag was introduced by browser vendors in 1995 there was no way of changing the font family except by changing the default font in your browser’s settings—which people rarely did anyway, as Times was seen as the obvious choice.

However, as the excitement of designating fonts on the web became more widespread—even becoming a W3C recommendation in 1997—designers rebuked Times fairly resolutely, and it became the emblem of early 1990′s web pages. The forgotten font of the web.

Typography Times

With the birth of CSS, web typography was given a new lease of life. Suddenly we had control of line-height (leading), letter-spacing, and word-spacing. But Times missed out on most of these revelations because hardly anyone was using it. At the same time, a school of thought that regarded sans-serif typefaces to be more readable on the screen, struck another blow to Times’s hope of a comeback.

Times are changing

In case you hadn’t noticed, this article uses the Times New Roman typeface. That might surprise you if you are only familiar with it in its un-styled form—the 1990′s emblem. I think this text is really rather pretty.

Let’s have a look at how we got from the plain, rather amateur look of the un-styled Times, to what you are reading now. Here is an example of your bog standard, rather embarrassing, default Times.

With that last addition we have the classic Times New Roman typeface looking as good as new, with the finished rule is as follows.

It would be great to see a resurgence of Times being used in 2005. Next time you need a serif font for the screen, instead of just typing Georgia without thinking, have a look at Times. Have a play with the CSS, and let’s see if we can restore it to its former glory.

I have no doubt that England will qualify for the World Cup in 2006. They have one of the best squads in Europe. Arugably two of the best midfielders in the world in Beckham and Gerrard, as well as two of the best defenders in Campbell and Cole. If they had strikers that could score at international level then they may well have won something in the last few years.

Let me make a few predictions…

Beckham will start for England as captain against Poland on Wednesday, as he will in the World Cup in 2006. He is one of the finest players in the world, a fantastic leader, and when on form can control and dominate matches.

David James will not start against Poland. I am sticking my neck out here big time, and may well be proved wrong – but I believe Eriksson may drop him in favour of Paul Robinson.

Eriksson will not be sacked by the FA. It would cost them millions of pounds to pull out of his contract, and lets face it – he’s done a bloody good job since he got here.

England will not win a major tournament in the next twenty years. Shock horror! For England to win tournaments things need to change in the English game. To begin with, the English Football Association (FA) need to start giving a shit. Their priorities are to the Premier League, the most popular and exciting league in the world, and probably rightly so. Football in this country has always been rooted in club football, and that is not going to change anytime soon. Where, for example, is the long awaited and long discussed ‘winter break’ that so many figures have been calling for? Yeah… right.

The players feel the same. Even the most dependable in say, Steven Gerrard. In an England shirt, is he ever going to hurtle himself around the pitch sweating blood from his pours, as he did when he single handedly propelled Liverpool into a European Cup spot last season?

Perhaps these are depressing thoughts for England fans. You don’t have to believe them though. Maybe I have it all wrong and the National game is in fine form. Either way, we have a long hard season of Premiership football to look forward too. Surely that is infinitely more appealing than worrying about Beckham and company as they trudge around Europe for these ‘tricky away ties’.

Update:
9/9/04

Well goodness, if I wasn’t proved right on a number of points. Beckham did start as captain and David James was dropped in favour of Paul Robinson. For the record, England won 1-2.

I have donated my remaining Gmail account invitations to him. So if you are still after one, please visit him, and consider a donation to the National Multiple Sclerosis Society. Let’s use the hype to do some good.

As most people rightly pointed out, most of the fuss over Gmail centers around the fact that you can’t actually get one yet. Google have cunningly tapped into a human drive that is with us from the day we are born until the day we die. The desire to want something we can’t have.

Now we all have Gmail, (and if anyone does not, I may – depending on when you are reading this post – be able to help you out with an invite) are we still mad keen on it, and is anyone actually using it?

My initial thoughts on the Gmail service is that it’s good, but I can’t yet see what advantages – apart from the 1GB storage – it holds over Hotmail or Yahooservices.

Gmail are pinning all their hopes on the way it is marketed. In their words, ‘Gmail is different’. They are massively plugging the 1GB of storage of course, but there is also the search facility, and the feature of showing messages in context as conversations. If Gmail does become successful it will be because these features catch the imagination of the public, and not because it is a superior email service to others.

My new Gmail address is andyhume@gmail.com.

This afternoon, I managed to post a rather sensitive password on to a very busy community’s web forum. A forum, that many WordPressusers may well be familiar with actually, as you’re all very busy on it.

Anyhow, after a 30 minute password recovery project – involving a quick visit to a server admin panel, ftp clients, and the like – I popped back to the ‘forum’ to see exactly what, in my stupidity, I had done.

Perhaps it’s just me , but there is something about the login interfacethat is slightly confusing. It seems a little tied up with the whole ‘new post’ thing. Maybe it could be moved above the ‘new topic’ <h2>, or perhaps to the bottom near the submit button. May suggest it to Matt.

What’s ironic of course, was that I was already logged in – hence the message being posted – and didn’t even need to enter any passwords.

What they are offering appears to be slightly different to your standard $5 (£2.72!) web design package – if there is such a thing. It is certainly feature packed, with options for e-commerce with intergrated merchant accounts, server logs, message boards, and the like. Of course, without signing up for a yearly subscription to Sam’s Club and then paying the extra for the website package (neither of which I was particularly attracted by!), it’s not possible to ascertain exactly what is available for $5 a month; and frankly, I’ve lost interest in finding out anyway.

What has been raised though, is the question of whether freelance designers like myself – and Scrivs I believe – be concerned about the rise of template packages such as these, no matter how featureless or feature packed they are?

There is no doubt that these offers are taking work from smaller design firms and the freelancer. That is unfortunate for whoever is losing the work; but at the same time it is pushing us to understand how we need to extend our skills as designers and developers.

Services such as Sam’s Club, which offer a web based interface allowing pretty much anyone to set up a website ‘in less than 5 minutes’, lack one pretty crucial function. No one is planning or designing the site. Users are just dumping content into various holes, specified by the templates. How are these sites really different from a small business owner, sitting down on a Sunday afternoon with a copy of Frontpage, and ‘designing’ a site for his business? Anyone with a reasonable amount of common sense and patience can do that, but it does not guarantee a successful online store, or customer service solution, or anything really.

My point is this: Yes, these services will take work from freelancers; but is it work that we would value highly anyway? I, like all designers, take a certain amount of pride in my work. I happen to specialise for a particular client type in my work, and I have a measure of self-esteem, and belief that I offer an excellent service in that area. The expertise I offer, as do all designers and developers, cannot be funnelled into some templates for a £2.72 website. What about information architecture, site goals, target audience, search engine optimization, not to mention standards.

These are facets of design that web professionals obtain through knowledge, experience, and practise. They are also skills that cannot be replicated by a template, the skills that we should continue to learn and develop, and the skills which mean we have a few more years left in us yet.

The philosophy behind the Sam’s Club service is, as Scrivs rightly points out, to increase the number of small/medium businesses with a web presence. So let’s hope, firstly for good old Sam’s sake, that many people sign up for it; but let’s also hope that within a few years, these businesses come to see the importance of a successful site, and realise what they have missed out on by buying a template service.

Remember the digital watch you bought from the petrol station for 99p thinking you’d found a bargain? Me too, but Rolex didn’t have much to worry about did they?

Still totally at a loss with what to do with the side bar here at The Dredge. I am playing around with offering a DOM based style switcher, allowing users to choose between two or three styles of the side bar, and seeing which proves most popular. I am quite happy with the rest of the design so far, and once I have tied the side bar into the overall feel, I think I will be well on the way to possibly having a finished Web site!

Implementing a site design while the site is live is quite an interesting concept. I remember Zeldmandoing it a few years back, and thinking it was a very bizzare affair. In hindsight, I imagine he found it quite exciting, and no doubt received hundreds of thoughts from visitors during the redesign with their individual opinions and ideas. Having the time to mull many concepts and notions for a design is not something Web developers are generally used to. Many projects tend to end up with some deadline or other looming dangerously close, and time to sit on a problem or mull over a concept is often unavailable.

I have enjoyed, and indeed, am enjoying, the ‘no pressure’ aspect of The Dredge’s development. For example, yesterday I added the little ‘Dredge the Archives:’ image replacement for the search interface. Yes, that’s all I did.

Anyway, having said all that, you probably realise I would encourage comments about any aspect of the site’s design. Infact I would like them very much, as I have no strong feelings about where to go with the design, and will probably mull things over for a good few weeks before making any more changes.

Isn’t that great. I’ll just wait and see what springs to mind… no pressure.