W3C home > Mailing lists > Public > public-webapi@w3.org > August 2006

Re: XmlHttpRequest IDL non normative

From: Anne van Kesteren <annevk@opera.com>
Date: Thu, 10 Aug 2006 11:37:59 +0200
To: José Manuel Cantera Fonseca <jmcf@tid.es>
Cc: "Web APIs WG (public)" <public-webapi@w3.org>
Message-ID: <op.td2adlfh64w2qv@id-c0020.oslo.opera.com>

On Wed, 26 Jul 2006 08:52:46 +0200, José Manuel Cantera Fonseca  
<jmcf@tid.es> wrote:
>>> It really sounds strange to me. To specify something in IDL that is  
>>> not OMG-IDL-conformant but you are going to use the bindings of  
>>> OMG-IDL.
>> I'm not sure what you mean by "us[ing] the bindings of OMG-IDL", but I  
>> don't think we are. The "IDL" in the draft is there because it's  
>> intuitive to people who are used to the DOM specifications and the  
>> such. We're not trying to conform to OMG IDL simply because it's not  
>> powerful enough to capture what we need to express.
> Bindings is the word that is used in the document :-)

Actually. "Language bindings" is what's used in the document.

> I think that a document that tries to standardize something and itself  
> doesn't conform or adhere other standards is simply nonsense.

The IDL is simply non normative and illustrates the API. The actual  
language bindings (not yet included in the document) will define the exact  
mapping to ECMAScript and perhaps other languages.

I don't see the problem.

>>> If you are not going to use the sintax and semantics of OMG-IDL it  
>>> could be better not specifying the object in IDL. You could do it  
>>> directly in EcmaScript.
>> I'm not sure Ecmascript would be a good option here, but I don't have a  
>> strong opinion. The best option would be to document a "Web API IDL"  
>> but that's quite a lot of work.
> Why ECMAScript is not a good option? All the DOM developers and Web  
> Developers know the language and in fact the APIs you are trying to  
> standardize are yet defined in browsers and developers are used to use  
> them from EcmaScript.

As far as I know ECMAScript has no notion of IDLs... Anyway, I agree with  
Robin, we need a Web API IDL document. Unfortunately, I have no clue as  
how to write such a thing.

Anne van Kesteren
Received on Thursday, 10 August 2006 09:38:18 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:16:21 UTC