- From: Jesse McCarthy <mccarthy36@earthlink.net>
- Date: Thu, 8 Nov 2001 19:14:47 -0500
- To: www-dom@w3.org, keshlam@us.ibm.com
"Joseph Kesselman" <keshlam@us.ibm.com> wrote on 11/8/01 6:16:29 PM: > >The DOM is an API, not a set of specific variables. It expressed absolutely >no opinion on what your variable are called; it just says that if you have >an object which presents the Document interface, it will have thus-and-such >methods and properties and behaviors. > >If you want to assign a Document to a variable that you call custardPudding >-- assuming that the datatype of the variable is such that it will accept a >Document -- that's entirely your choice -- just as you can name your >integer variables anything you please. > >> If the purpose of the binding is not to establish actual object names, >>then I don't see what purpose it would serve. > >It isn't. It establishes the methods and behaviors of those objects once >you've obtained them. How you obtain them is out of the DOM's scope. I see your point. That leads me to my next question (I don't know if this is in your sphere of interest). I am a web developer, seeking to use the DOM with ECMAScript to author scripts that will function in any browser that complies with the standards for DOM, ECMAScript, and whatever version of HTML -- without any monkey business to find out what browser it is or do anything to accomodate any particular browser. Obviously the most fundamental interface that I will need to access is 'Document' or 'HTMLDocument'. If different compliant implementations of the DOM can provide access to that interface by objects of different names, how am I supposed to accomplish my goal?
Received on Thursday, 8 November 2001 19:42:19 UTC