« April 2007 | Main | July 2007 »
June 30, 2007
Data-Mining for Web Coders
Rene Saarsoo's 2006 thesis, Coding Practices of Web Pages presents histograms of CSS usage, attributes and their values. One cool finding; the top 3 fonts used in the CSS sites sampled are sans serifs: Arial, Verdana, and Helvetica.
Google did a similar study back in 2005, resulting in Google Code's Web Authoring Statisics. Unfortunately, their results only cover the most popular HTML and CSS elements.
Rene's thesis brings up some interesting points. Given that we're roughly 30 years (depending on who you ask) into the mainstream computer revolution and HTML will be with us for a very long time, there's value in using such studies to simplify future web standards.
Posted by Devon at 10:03 AM | Comments (0)
June 19, 2007
Why Apple Released Safari for Windows
Some possibilities:
- To make it easier to develop apps for the iPhone. It's still a PC world, among developers even more so. Putting a binary in the hands of developers knocks down a big barrier to entry, and will make a big impact, the same way ADC membership quadrupled in the year since Apple shifted to Intel. Apple will innovate in Safari, and these innovations will run on Windows, Mac, and iPhone.
- Lock-in by proprietary extensions. For example, the web services on my corporate intranet all but require IE due to the use of proprietary features and extensions. FireFox and other browsers simply don't work 100% of the time, which is the kiss of death when using mission-critical apps in that context. Replace "corporate intranet app" with "iPhone-aware web app" and it's possible that the cool, flashy new sites built for the iPhone will be a driver for folks to use Safari on their desktops.
- Lock-in by bundling. iTunes is de facto for PC iPod owners, regardless of what their favorite MP3 player app is. Safari may become de facto for PC iPhone users. The iPhone will integrate with a desktop, it's not clear yet what form this will take. Even if Safari isn't the primary mechanism for sync'ing, there's value in having, e.g., shared bookmarks, which will likely work better with Safari than any other browser.
- Own the desktop. iTunes and QuickTime are Apple's beachhead on Windows. If Safari gains equivalent traction, the Windows desktop looks and works more like a Mac desktop, knocking down another reason not to buy a Mac.
- Adobe already did it. They showed that it was possible, even easy to (effectively) release Safari on Windows. AIR, formerly known as Apollo, integrates WebKit, and has been released for Windows. The WebKit library forms the underpinnings of Safari.
Posted by Devon at 10:33 PM | Comments (0)
June 16, 2007
Marc Andreessen's blog ain't bad
Don't know why I'm surprised, but Marc's blog is a terrific read. He jumps from a list of the top 10 sci-fi novels of the '00's to a paper claiming that 48% of the late 90's dot-com startups survived.
My new metric for blog quality is when I find myself led to the same blog for multiple good articles in the same sitting. Given the sameness of blog layouts, it's all too easy for these cookie-cutter CSS layouts to blend indistinguishably together.
Posted by Devon at 02:23 PM | Comments (0)
