Saturday, 29 October, 2005

Further to my previous rant, there is light at the end of the tunnel in the form of MathML. MathML does scale with the size of the text around it. Fantastic? Well not quite.. You see, you have the problem of how to detect that the browser actually supports it and this detection is not as trivial as it should be.

My first thought, would be for the browser to put something in the accepts header to say it can process that kind of content. I don't know why nobody has done this because it is blatently obvious to me as a web-developer that this is the entire purpose of the Accepts field in a HTTP request.

Instead, you have to reach for a complex bunch of Javascript to do the detection client side. This approach is imperfect and annoying. Firstly, sticking Javascript in your pages is something you should avoid at all cost. Let me digress for a paragraph, don't you think it's slightly mad that you're potentially letting a random developer somewhere else in the world execute any Javascript of their choice on your browser. What are the chances of you actually inspecting the code to see if it's safe? I've never done it and I'm as paranoid about security as they come. My advise to anyone is to live without Javascript or as a second best, run your browser as a non-privelleged user. There is a third option, which involves installing a plugin that blocks Javascript from all but a few white-listed sites. I don't like this because it's easy to subvert, if you don't believe me click this link (it's safe, trust me) to find out why.

I've had a realisation in the past week that the gap between the state of the art and what you can actually deploy seriously on a grand-scale is widening. Back in the days when HTML had just been invented (and I say this as somebody who was not old enough to write a web-page at the time) a new tag or a new feature quickly became adopted all over the world. Today is a stark constrast. It will take half a decade or maybe more before you are able to count on MathML being available in over 90% of the browsers that views your site. This isn't just unique to MathML but for any of the new innovative featurse in the browser world. Thanks to the Microsoft browser monopoly, the pace of innovation has crawled to a halt.

Will this change soon? Well the open-source browsers such as Firefox, Konquer0r et al, are bitting in to the Internet Explorer monopoly with some strength but Firefox still has less than eight percent of market share. While further growth for Firefox and the other open source browsers looks like a distinct possibility the rate of that growth is likely to be slow. Everybody who is tech savvy has already got Firefox. The people who will despise Internet Explorer just because it's from Microsoft have already got Firefox. The only people left are the people who are happy to live with Internet Explorer. Convincing them to switch is proving a lot tougher.

I think a lot of people misinterpret the important of Firefox. Firefox is not important because it may replace or substanitally reduce the Internet Explorer monopoly. Firefox is important because it forces Microsoft to innovate. The Internet Explorer team has been reformed and work is starting on the first serious upgrade to the product for years. This is a good thing for everbody, whether you like Microsoft or not.

Simon.

14:14:31 GMT | #Programming | Permalink
XML View Previous Posts