Sunday, August 05, 2007

Microsoft's IE6 browser dying a slow and (for me) painful death.

I'll preface this post with the fact that statistics lie, are unreliable and can be fudged to support almost any conclusion, but the general trend here is what I am interested in rather than specific numbers, so that said, let me get started here...

IE6 has been the poster child for the need for Web standards for many years, with its broken box model implementation and numerous tedious problems, such as margin doubling on floats, peek-a-boo, guillotine and other mostly hasLayout related bugs, resulting in the need for all kinds of nasty hacks and workarounds in our CSS.

The Web community awaited IE7 hoping (against all hope :) that perhaps cross-browser problems would soon be a thing of the past. And while it still does not implement some of the features of other standard compliant browsers, IE7 a vast improvement over IE6.

The problem is that the adoption of IE7 is proceeding very slowly. A year after IE7's launch, browsers stats from thecounter.com for July 2007, (which is widely used by corporate types and therefore possibly likely to give a fair picture of the outside world as whole - see disclaimer above) show IE7 with less than 20% share and IE6 still at 51%. In other words, over half's the world's users (according to this single data point - disclaim, disclaim), are still on an out-dated, non-compliant browser.

w3schools.com's browser stats tell a similar story - only 19% adoption for IE7 by July 2007, although they show a lower percentage for IE6 - 37.3% and more - 34% - for Firefox, which is what you might expect from a site that attract people who are into creating for Web and therefore willing to make the effort to download a more standards-compliant browser.

What's amazing is that IE7 had hit this adoption (19-20%) level, according to The Counter, in January of 2007, and have only increased 6% (13% > 19%) according to w3schools. So adoption is either stagnated or is pretty darned slow, depending on your point of view and sources of data. At this rate, I will still be coding around IE6 in 2012.

So what can hasten the demise of IE6? I've got a couple of suggestions.

First, Microsoft does not take active steps to tell users that the 'new' IE7 browser is available. By contrast, it seems not a month goes by that Firefox doesn't pop up a little notice telling me that an updated version is available and would I like to download it, which it does in seconds and automatically opens - minimal pain, and I am always current. Microsoft does not do this - I would guess that most current users of IE6 don't even know that IE7 is available, and most that have IE7 didn't upgrade simply got it with a recent computer purchase.

If Microsoft wants to show the world it is really behind Web Standards, then any XP patches (which it does inform users need to be downloaded) should include a pop-up for IE6, with a message that reads "Hey, do you realize your current browser is outdated and you could have a much better one in a couple of minutes?", or words to that effect, and a Download Now link.

The other reason that IE7 isn't bringing about the rapid demise of IE6 is that Microsoft forces would-be IE7 users to prove that their system software is legal before allowing them to download it. Now I do NOT condone software piracy, but you would think that in the interests of showing they give a hoot about Web Standards, and the fact that the browser is supposedly free, that Microsoft would allow anyone to download IE7 - clearly, it's not free if you have to spend $200 before you can run it. I can understand enforcing that the user's system software is legal before installing Microsoft Office or other similar pieces of their magnificent and expensive software, but a free browser? Pu-leez!

So, Microsoft execs, I am appealing to you to get an "Upgrade your Browser" message into the next patch of XP, and stop forcing the legal system verification process onto people who want to upgrade to IE7. We need to retire IE6 fast, and I hazard to guess that you are not getting more revenue by forcing people to prove their legal ownership of XP. What you are getting is sluggish adoption of IE7.

Perhaps the most profound effect of all this is you are slowing down the move to Web Standards that you (finally) committed to champion, to say nothing of keeping me and countless other Web developers up at night, struggling to find suitable ugly hacks to add to code that worked first time in Firefox, Safari, Opera - I estimate that as much as 70% of my CSS development time is spent working around IE6 problems. IE7 is a big step towards walking the Web Standards talk, but what use is it if it's not actually getting on to people's machines?

Thanks for listening, Microsoft. You are listening, right...?

Saturday, August 04, 2007

Back on the blog

Well, I started this blog with the best intentions, and after a few posts, I got bogged down in my previous book Codin' for the Web, my PHP and mySQL tutorial book, and then moved to the East Coast and started a new job. So nothing has been posted for over a year. But enough of the lame excuses; I am back on the blog and thanks to all of you who have posted while I have not.

I am working on a new version of Stylin' with CSS, my Cascading Style Sheets tutorial book, and I will blog about that soon, as I have a lot of new material, particularly code examples, for y'all, as they say around here.

Sunday, March 26, 2006

SXSW and Disneyland - there is a difference...

It's been almost two weeks since SXSW and once again it was an excellent show, mostly because of all the amazing people I got to meet and all the great things they are doing - not many boring people at SXSW... I have just got back from Disneyland where my two daughters had the time of their lives, and I had a lot of fun too, so it's been quite the month for exposure to media. Now I have to get back to working on my new book and get it finished by the end of April. Doesn't leave much time for blogging. Oh well, thanks for all the mail I got about the CSS presentation in Austin and Stylin' in general and I will write back back to everyone as soon as I can...

Saturday, March 11, 2006

Here at SXSW

Wel, here I am sitting a seesion on Podcasting - great quotes from the audience "Podcasters are people who know the technology - rather than anyone with the content". Laura Swisher said - "Podcasting is for people who aren't going to make in conventional "hollywood" ways" -
sommeone just said - "it's about getting the content up first and finding the advertisers afterwards"
"A targetted audience is difference between podcasting and radio"

Meantime I am busy IMing with Christopher Schmitt who is in a session next door - that's how you make sure you are int the right session.

Tuesday, February 28, 2006

SXSW - CSS Problem Solving panel update

For those of you attending everyone's favorite geek-fest in Austin in March, the panel for the CSS Problem Solving panel is now Christopher Schmitt, Dave Shea, Ethan Marcotte, myself and a recent exciting addition, Tiffany Brown, web maveness and writer of her (in)famous feminism blog. She may even talk about CSS! This is an awesome group of people and I am very happy to be up there with them - I am going to deconstruct some all-CSS multi-column layouts, and then sit down and watch these talented people talk about the really difficult stuff.

Sunday, February 12, 2006

Helpin' with Codin'

Been working on the new book for while now - the pitch is that I'm going to help designers understand how web sites - the real ones that are connected to databases and can let you shop, blog and do other interactive things - are actually built so that you can work effectively with programming teams or actually start learning to program yourself. So if you have input about what you want to know in this area or what is hard to make sense of, there's still time for me to drill down into areas that really matter to you. After all, this isn't for me, but you.

Thursday, February 02, 2006

In the code zone

Writing code is like spinning plates - after about 20 minutes I get them all going - all those variables, arrays and what's in all the objects is laid out like a map and I can type as I think. But it just takes one interuption - a phone call or some vulture coming in my office to pick my brain, and everything comes crashing down. So I am a miserably anti-social programmer - no music, no talk radio and most of all, no people - just me, the code, and a cup of tea to sip from as I compile and review. So the rest of the time, I'm my usual friendly self but if you come in and I'm facing that screen - don't bug me, I'm codin'...