Its fairly odd that a design tweak like the New York Times’ website overhaul should be news, particularly since CNN did pretty much the same thing with less fuss a couple of weeks ago. And it took the times nearly a year!

Nerertheless, something interesting is at work - first, sites are now ignoring smaller screens for the first time in years - 1024 pixels wide is becoming the standard. More importantly, by ignoring the low end they can also ignore large screens in a way that 800 pixel wide designs didn’t really cut it. They are bypassing the ridiculous ‘holy grail’ three column CSS layout that geeks with no graphic design sense use in favor of fixed column, paper-like designs used by web designers.

Lastly, with RSS and Ajax, the notion of a page impression is gone - and yet that, rather than just impressions is what is often measured for advertising.

I wonder how long the page impression metric will last?

The New York Times - Breaking News, World News & Multimedia

The Apple Newton showed that form factor is everything. The Newton was pretty cool and ahead of its time - the first PDA. But it was the wrong size.

The Palm Pilot did less (it didn’t even try to do full handwriting recognition), but it was the right size. The same form factor as Walkmans and cigarette packets and wallets and iPods, it slipped into existing shirt and jacket pockets, to be carried everywhere.

Good design is about the right choices not technical wizardry. The Segway looked like magic - a two wheeled stable vehicle. But it needed gyros and computing power because its two wheels were next to each other. If its wheels were inline then it would have been a bicycle and the conservation of angular momentum alone would have kept it aloft. The bicycle is a better design for a two wheeled vehicle than a Segway.

The Origami is remarkably similar in size to the Newton - and therefore, even if it packs as much technical innovation as the Segway, unless people start wearing sporrans, the world over, it will die.

Intel shows Origami-like device | CNET News.com

Windows Live Local - Virtual Earth Technology Preview

In 1994 lot of people thought VRML was cool, but it wasn’t cool compared to offline video games and it wasn’t useful compared to regular web-page search and browse.

This mockup of Microsoft’s Local Live’ reminds me of VRML. As a video game, it’s crude by the standards of 1994 and as a web app it has none of the design sensibility of Google Maps.

The mockup is from a multi billion dollar company whose most obvious online avenue of attack against Google is local advertising - and the end result is a maps application that allows you to choose a view that superimposes crappy vignettes of the interior of a ‘race car’ or ’sports car’, as you ‘drive’ around maps. The product substitutes kitsch and gimmickery for ergonomics and usefulness.

I love San Francisco, but when you realize that the Castro Safeway and the Sony Metreon are landmarks, you realize that its a small city. It is the worlds best sleepy seaside town.

One of the two landmarks has some architectural merit, the other is the Metreon, a failed attempt at a French style Mediatek - a multimedia library/art complex.

In the end the Metreon, complete is a shopping mall, and now its being sold to a mall developer.

Metreon’s shattered dreams via kottke

Jakob Nielsen says that the new UI paradigm to replace Apple’s will come from Microsoft:

“Macintosh-style interaction design has reached its limits. A new paradigm, called results-oriented UI, might well be the way to empower users in the future…The next version of Microsoft Office (code-named “Office 12″) will be based on a new interaction paradigm called the results-oriented user interface”

Results-oriented UI turns out to be templates. Because there are too many options in MS Office to have individual commands the idea is that the results of groups of them are displayed.

It is, perhaps, a bit rich for anyone to champion Microsoft over Apple in terms of design at the moment, but design is subjective, I guess. Where Nielsen is provably wrong, however, is where he confuses User Interface with User Interaction (isn’t he supposed to be an expert in Interaction?):

“rather than typing in commands and parameters, users select commands from menus, freeing them from typing errors. Menus, toolbars, and dialog boxes operate on the screen’s visual objects, which faithfully represent user goals. This is known as WYSIWYG, or What You See Is What You Get.”

No. WYSIWYG refers to the fact that what you see on the screen is what it looks like when printed or nowadays, printed on the web. It has nothing to do with dropdown menus rather than typing commands, it is just that command line interaction sits in a UI that is not WYSIWG. This is just a confused metaphor.

I used to work with a CAD system that had several thousand menu items because templates don’t tend to work for professional software. Templates are often the ‘hide the crap under the carpet’ approach of UI design avoiding stripping out superfluous features for true design elegance and providing the worst UI experience of all when templates can’t be adapted intuitively

This would be OK, but for the fact that from ‘clippy’ to the automatic insertion of things like bullet points in Word, Microsoft has a terrible record at UI which is results oriented, i.e. tries to guess what results you want and groups commands together.

What’s currently badly designed about Office is not the details or the bloat, but the premise. Yes, the WYSIWYG UI is dead, but only in the sense that it refers to printing a document from what you see on screen.

Paper documents, ledgers, and overhead projector slides were the metaphors that, Word, Excel and Powerpoint were based on. But we live in a realm of email and weblogs, websites, shared web updatable financial data and multimedia mashups. Office is not designed to deal with these types of documents. What is wrong with Office is Word, Excel and Powerpoint not the principal of menus.

R.I.P. WYSIWYG - Results-Oriented UI Coming (Jakob Nielsen’s Alertbox)

Marc Canter had it absolutely right when he cautioned about the fuss over Ajax.

Perhaps Ajax is a meme more than a ‘thing’, and like all good meme’s something that is spreading because the environment is ready for it.

When I first used Gopher or WAIS and then downloaded Mosaic I was impressed by the architectural simplicity of Internet applications, so much so that I stopped being an architect and started working on web stuff.

Here was something in computing that was seemingly a retrograde step - one window instead of many. I spent most of my day at the time in front of a twin screen CAD application that had several hundred palettes. But because that one window opened onto a world of other computers, like a unix terminal, it was so much more elegant. (My favorite new experience with UI has been finally using VI, a text editor with Zen-like elegance that has evolved over 30 years while maintaining simplicity).

In fact the one window paradigm has persisted in an almost inevitable way, as the parasites of the web have driven people to use pop-up blockers, thus keeping it from becoming more like windowS, plural, apps.

The chunkiness of the web continued when forms were introduced, and people started to create applications, but with the new restrictions of statelessness and non event-driven UI.

Restrictions are sometimes held up as the mother of elegance and the web had all of them: bandwidth restrictions, browser compatibility restrictions, restrictions maintaining state, procedural user interface restrictions and the one window paradigm.

From this, the web has, in my opinion, improved the quality of software design.

But the web now exists in an environment where some of those restrictions are easing and perhaps the time is right for event driven UI (which as Marc pointed out, has been available for years with Flash, but just within the wrong community) so perhaps Ajax is, like better bandwidth, a natural evolution.

And, unlike the poorly thought out attempts at useability standardization, Ajax is an elegant way to lay down some specifics for richer web apps - even if there is absolutely nothing really new in it.

Continuing on my anti-CSS rant - 3 columns, with fixed-width left and right and a fluid center column, are ofter referred to as the ‘holy grail’.

The problem is that this is an obsolete solution, when people increasingly have massive screens where any fluidity breaks the design if people auto-expand windows - which they do.

Mike Golding breaks the mold and argues the case very well for fixed width CSS layout: notestips.com :: The benefits of a fixed width design

Tables may suck, but CSS is no improvement. Yet web designers who have never used page layout tools for offline printing, or object based CAD software are still brainwashed by it.

I just came accross this classic: BlueRobot “Many a talented web designer has struggled with CSS-based centering. Though CSS vertical centering eludes us, two techniques for horizontal centering are BlueRobot approved. Take your pick”

First one: “Unfortunately, IE5/Win does not respond to this method - a shortcoming of that browser, not the technique” Fair enough, but then why recommend it (This is still one of the largest browser versions in use).

Second one:

#Content {
position:absolute;
left:50%;
width:500px;
margin-top:50px;
margin-left:-266px;
padding:15px;
border:1px dashed #333;
background-color:#eee;
}

All this to avoid ‘align=center’.

When I arrived in the US from bad teeth land one of the first things I asked my dentist for was a set of American teeth.

Unfortunately I was told there was nothing that could be done. However at dinner with friends last night I was introduced to ‘Rota Points’, the best toothpick in the world.

The bits in between my teeth are gone now, even if I’m still on page 27 of the Great Book of British Smiles.

Intradental Cleaner

Color scheme shows a base color with alternate color schemes based upon base, complement, split complement and neighbors.