On Nov 11, 2015, at 3:25 AM, Simon Slavin <slavins at bigfraud.org> wrote:
> There's a story behind this. > > During the review leading to HTML5 the web consortium (the one which decides > which features should be introduced to HTML/CSS/JavaScript) listed a > 'webdatabase' specification so that browsers could maintain data in a > database-like structure. The specification boiled down to "Implement a > simple SQL engine with an API like [this].". > > Unfortunately all the leading browser publishers implemented this happily and > quickly. "Unfortunately" because they all did it by including SQLite in > their project. This is a problem because the only way to ensure > compatibility in that case is for the web consortium to include the full > specification of SQLite inside the specification for a compatible web > browser. Otherwise people would start doing things like calling PRAGMA > commands from web pages, and then an unofficial standard would evolve which > required PRAGMA commands to be supported from the web API. > > So they did. And it failed. And that's where we are today. Sorta-kinda. The standard was called WebSQL, and it was an extension to client-side data stores. W3C setup a working group and sent out a request for implementations, but as you said, everyone in the working group implemented the standard by basically wrapping SQLite. The standard was then killed due to lack of unique implementations. As you said, there was concern that without independent implementations, aspects specific to SQLite would ?leak? into the standards and commonly used APIs, so the standard was killed. The official working group notes are here: http://www.w3.org/TR/webdatabase/ -j -- Jay A. Kreibich < J A Y @ K R E I B I.C H > "Intelligence is like underwear: it is important that you have it, but showing it to the wrong people has the tendency to make them feel uncomfortable." -- Angela Johnson
