Join 3,564 readers in helping fund MetaFilter (Hide)


The State of the Web 2008
January 12, 2009 4:25 AM   Subscribe

The State of the Web 2008 is a report from Web Directions that includes details and analysis of all the responses to over 50 questions covering technologies, techniques, philosophies and practices that today’s web professionals employ. The survey was open for just under 3 weeks, from December 1st to 20th 2008. In total, over 1200 designers and developers from around the world responded to the survey. Respondents were likely to be self-educating, “early adopters” who keep abreast of developments in their field. Here are the tabular results.

In short, what did the survey find? Some quite surprising results include:
  • just how few of the respondents use any form of Internet Explorer for their day to day web use (with only 3 out of over 1200 respondents using IE8), and similarly how few use Google Chrome as their primary browser, despite the splash the launch of that browser recently
  • nearly half of respondents use Mac OS X as their primary operating system, and only 10% use Windows Vista
  • less than a third of respondents test their web sites with Internet Explorer 8 (while Mobile Safari comes in at 20%, and Chrome at 40%)
There’s also a great deal of interest in terms of the nitty-gritty of web design philosophy and practice, from the high percentage of respondents who use JavaScript (around 95%), to the very small uptake of Silverlight (around 2% of all respondents) to the very high percentage of database driven sites (96%), overwhelmingly run on open source databases (over 80%).

The number of responses, 1234 total, and the results themselves definitely provide both food for thought, and in many cases, cause for optimism that web development best practices are becoming more widely adopted over time.

Web Directions conducts two major conferences annually in Sydney, Australia and Vancouver, Canada, and covers the full range of interests for web professionals - web design, front-end and and back-end development, information architecture, interaction design, accessibility, data visualization and much more. Their more focused conferences home in on specific areas and segments, as seen from their podcasts, slides, and other presentation materials.
posted by netbros (7 comments total) 9 users marked this as a favorite

 
How are the sample audience supposed to be representative of?

"The pure and simple truth is rarely pure and never simple."
Oscar Wilde - as quoted by the w3schools listing of browser popularity.
posted by rongorongo at 4:39 AM on January 12, 2009


How are the sample audience supposed to be representative of?

Who was surveyed
posted by netbros at 4:47 AM on January 12, 2009


Could the mods fix the rampant misspelling in this post? It keeps saying S-U-R-V-E-Y instead of S-E-L-F S-E-L-E-C-T-I-O-N P-R-O-B-L-E-M.
posted by DU at 4:48 AM on January 12, 2009 [5 favorites]


This could actually be useful for someone heading into a job interview.

The 16% number for ASP.NET sounds awfully low, though, and the Mac OS numbers high (although I would expect them to be higher than the general population).

My guess is that there's a concentration on freelancers, on people who work in agencies, and on general 'design' types, less representation of people who work in-house, less representation of 'developer' types.
posted by gimonca at 5:42 AM on January 12, 2009


jQuery for the win.
posted by signal at 6:43 AM on January 12, 2009


I'm one of the respondents in this survey. I'm not so concerned about the selection problem here -- yes, it's a self-selected survey of people who read web design blogs, which means you're going to get more bleeding-edge people than general population -- but so long as you read the results with that in mind they could still be useful as an indicator of where things are headed.

But I wish they weren't so obviously determined to get the answers they want to get out of this. Most of them boil down to "do you use [new technique] or [older but still functional technique]?" which is obviously going to skew responses towards the newer technique, because nobody likes the implication that their skills are out of date. And if you leave out any options for "I use some of each, where appropriate," as they did for example in the tables versus CSS question, that's going to skew it even more.

They make the push-poll aspect of this pretty clear in their commentary:
First up, we asked “Do you separate your JavaScript from your markup (“unobtrusive JavaScript”)?”, a recommended practice. Just under 4% of developers said no, while 39% said always, and 42% said usually. It’s clear that respondents are at least on the whole aware of this recommended practice. You know, because we just told them about it in the question.

(I'm not disputing that the practices they are pushing are generally good ideas; I just would be interested to see what the numbers would look like from a poll that didn't so obviously have an intended "correct" answer to so many of the questions.)

jQuery for the win.
Except that they split the vote for Prototype / Scriptaculous by listing them as separate libraries.
posted by ook at 9:01 AM on January 12, 2009


I happen to agree, though; I was a die-hard Prototype user, but tried jQuery on a recent project out of curiosity and totally fell in lurve with it.
posted by ook at 9:03 AM on January 12, 2009


« Older A Mash-Up of Google Maps and the Proposition 8 Don...  |  Old Timey Player Pianos!... Newer »


This thread has been archived and is closed to new comments