At the beginning of the year Microsoft celebrated the death of Internet Explorer 6 (or IE6 for short) with a cake at their headquarters in Redmond, Washington. More specifically they were celebrating that in December 2011 IE6 usage dropped below 1% of all traffic in the US.

Microsoft have been keen to see the back of IE6 for some time now as it’s one of the biggest things holding up further development of web technologies at the moment. It’s also very buggy and difficult to code websites for, not great adverts for Microsoft. Unfortunately usage has not dropped that low in the UK – Microsoft puts usage here at 1.4%.

The software giants aren’t content to sit back and let events take their course however, for example Google has stopped supporting IE6 with many of their web apps. It’s Microsoft itself though that has taken the biggest step, they recently released a Windows security update that automatically updated users from IE6 and 7 to version 8. I run IE6 on my computer to test sites on and when last month it was automatically updated to IE8 I watched the stats of the websites we monitor excitedly waiting for IE6 users to disappear… but they didn’t. In fact there was virtually no change in the percentage of IE6 users at all.

So who are all these IE6 users that aren’t applying Windows security updates? We have another tool which we can use to help us solve this problem. Analyzing the stats from this year of our biggest sites, we can see the domain names of IE6 users’ ISPs (Internet Service Providers). The top 10 are:

• Unknown (17.5%)
• fsa.gov.uk (5.4%)
• nhs.uk (5.3%)
• gsi.gov.uk (5.1%)
• police.uk (4.3%)
• chase.com (3.9%)
• messagelabs.net (3.7%)
• southwark.gov.uk (3.7%)
• edfenergy.com (2.5%)
• mod.uk (2.0%)

I think there is a perception that the majority of IE6 users are stubborn consumers who have no inclination to upgrade, and most campaigns – including Microsoft’s own banner campaign – rely on educating people to upgrade there and then. However our data paints quite a different picture, government users alone account for over a third of known IE6 users, and corporate users account for most of the rest. Most people therefore are using IE6 because they have to due to decisions made at management level.

If our (very non-scientific!) analysis is correct then the main thing stopping companies developing next generation web technologies in the UK is the government itself.

It’s not all doom and gloom though, as we can see from some of our data (compiled from almost a million users) IE6 usage has been shrinking rapidly for a long time now, so there are signs that big businesses are switching to more contemporary technologies and we have it on good authority that even the new government intranet won’t support IE6, the switch is coming! In the mean time unfortunately IE6 users represent a rich segment of the market and our sites all still support them.

But we can say goodbye to IE6… Just not quite yet!

Image credit: issaco

As HTML 5 gets closer to fruition some are beginning to question whether it finally removes the need for Flash altogether. None are fighting the battle to banish Flash to the annals of history more than Apple and this week the conflict between Apple and Adobe (who produce the authoring tools for Flash files) flared up again.

Historically Apple and Adobe have had great relations; Adobe is responsible for producing the graphics programs that drew creatives to Macs for many years when they weren’t quite mainstream yet. No one (not even Adobe) expected a phone to run full Flash – Adobe tried to make a stripped down version called Flash Lite but it never really took off – so it was never questioned that the iPhone should be able to run it.

The debate started when Apple released the iPad. Most predictions expected it to run a modified version of OSX, instead it ran an uprated version of what is now called iOS. When Apple announced it was “the best way to experience the web” many could not help but notice the absence of Flash. What followed was a very public debate between the two companies: Apple arguing that HTML 5 now offered everything that Flash could do but that it’s more ubiquitous, open-source and efficient; Adobe arguing that Flash is ubiquitous, that it’s very capable of being used on smart-phones and that they are very open-source (the last statement being a little bit of a lie, but Apple is hardly a pioneer of open source code).

The debate didn’t really go anywhere but over the last few weeks it has hotted up again with two important announcements.

Unlike the browser Google Chrome, which includes Flash functionality by default, Apple’s browser Safari only includes Flash as a separate plug-in. Last week they announced that they would no longer be supplying it with their new computers. This is clearly a move designed to force developers to create alternatives to Adobe Flash content but until this happens it means that a brand new Mac, which always made a point of working straight out of the box, will not show videos on let’s say the BBC website until Mac users download a separate plug-in from a third party. I have to say I don’t think that Apple are putting their users first in making this decision.

The second announcement perhaps makes it a little more clear as to why Apple are being so aggressive towards Flash. Last year Adobe announced that it has done a deal with TV chip manufacturers allowing them to ship “Flash applications” to televisions and opening up a huge new area of the market. This week they announced that Adobe AIR 2.5 is out which will enable Flash based applications to run on Google Andriod, RIM’s blackberry playbook and Samsung’s latest TVs.

It is yet to be seen how this conflict will pan out, whether it will just be a messy confrontation between companies with no real winner or whether Apple will succeed in sinking Flash. Ultimately it seems this will rest on whether Adobe and AIR 2.5 can successfully break into the App market and retain its development community, perhaps this in itself is an interesting insight into the software market in 2010.

Some people love Apple, some people are indifferent but one thing that’s become difficult to deny is that they have produced some revolutionary products that have changed the face of many industries. They’ve kept Microsoft and many PC manufacturers on their toes with computers and operating systems, the iPod – which received a cold reception from some quarters when it was launched – blew the portable music player industry apart, but surely it’s the iPhone that has changed the way people perceive electronics the most. It has not only changed how phones look and work but also peoples’ perception and expectation of how operating systems can be.

So another year, another iPhone. And this one has not been without its controversies! But I’d like to focus in on one feature which I think is a hugely important step in consumer electronics: the screen, and in particular the pixel density.

What is pixel density and PPI?

So firstly what is the pixel density, how is it measured and why is it important? The pixel density is the number of pixels in a given area of a screen. You can measure the pixel density in a number of ways but a common if arcane one for the moment is PPI, which is the number of pixels per linear inch on a screen and it is very similar to DPI: the dots per linear inch printed on a piece of paper. The screen on your computer is probably about 100 PPI and therefore it is probably pretty easy for you to see an individual pixel. This means that if you try to make text with just black and white pixels it comes out looking noticeably blocky like this:

To get over this blocky look programmers use a range of techniques, the simplest is called anti-aliasing, which is the method Flash uses. This involves sampling the text at a sub-pixel scale and then using mid tones to represent pixels which are partly full, which looks like this:

The method you’ll probably be most familiar with for making text as sharp as possible uses the fact that each pixel on a computer screen is made up of a red dot, a blue dot and a green dot, and it displays text  in multi colours so that it uses these sub pixels.

But what PPI would you need so that you wouldn’t need to do all this clever stuff with aliasing? How small would a pixel have to be so that your eye no longer has the optical resolution to resolve a single pixel? Well the answer is the same for screens as it is for print: 300 DPI. And what is the pixel density of the iPhone 4 screen? 326 PPI.

The iPhone 4 screen

I think the screen looks stunning, it’s amazing how detailed the icons are and it’s slightly perplexing that you can’t see the pixels no matter how close you get. To give you an idea of how many pixels this screen has check it out pixel for pixel on your own monitor:

This allows Apple to produce a reading experience that is much closer to and may become preferable to reading on paper, and I think it is a vision of the future.

Now before everyone says it, I know this isn’t the first device with a high pixel density screen, but it’s probably the first one that most people will see and although Apple may not be the first to bring these features to market, they produce the device that changes the market.

The future of computing and websites?

I think it is inevitable that there will begin to be demand for these high pixel density screens on desktop, laptop and netbook computers in the coming years, but it poses two huge problems: a hardware problem and a software problem.

The harware problem is a simple one, for a 22 inch screen you would need a resolution of about 5550 x 3450 pixels, which is going to be difficult to manufacture and expensive to buy.

The software problem is a little more fundamental. Historically when screen resolutions have increased so have the physical size of the screens, which has kept the PPI approximately the same. So this gave you more screen real estate but kept the physical size of – let’s say – the start bar the same. You can see how little change there has been in pixel density by looking at this screen shot of Windows 95 from 15 years ago!

This stability of pixel density has allowed developers to use the number of pixels as a stable measurement that corresponds to a physical size on the screen. This basically means if you went to the Igloo site on one of these imaginary screens the Igloo site would measure just 10cm across, which is clearly not how it was intended… Even things like the text size are not defined in terms of their final size like they are on paper, but on their height in pixels.

So what does this mean for the future of the web? Well obviously it doesn’t spell it’s demise! Old websites could just be scaled to be three times bigger (or nine times the area!) and they would still be blocky but they would be the physical size they were intended; the text could then be rendered at a higher quality but a similar physical size. That is the approach that Apple took when viewing legacy iPhone apps on an iPhone 4. But what about new websites? Will browsers have a feature that scales new websites down for older computers? Or will developers have to produce two versions until the user group using these old fashioned screens is small enough to ignore? That will depend on the operating systems and the browsers and ultimately only time will tell.

I would like to see a step change in the way that websites and indeed operating systems work, whereby image and text sizes are measured in mm and points instead of pixels, and will have a “DPI”. Each element would then be rendered for each person’s screen. This would be a major step in freeing up hardware manufacturers to produce high pixel density screens and ultimately would be a great step towards the paperless office.

Perhaps the change will be so complex to make that by the time the technology is ready the world wide web of hyper-linked web pages would have been superseded by the next generation of internet pages/programs.

For me though the change to high pixel density screens (or HPD screens to coin a term) can’t come soon enough.