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

Re: No arguments to XMLHttpRequest.send (ACTION-58)

From: Jonas Sicking <jonas@sicking.cc>
Date: Fri, 03 Mar 2006 11:41:38 -0800
Message-ID: <44089BF2.9020400@sicking.cc>
To: Jim Ley <jim@jibbering.com>
Cc: Web APIs WG <public-webapi@w3.org>

Jim Ley wrote:
> "Maciej Stachowiak" <mjs@apple.com>
>> For things that some but not all UAs implement, and that we'd like to 
>> require in a future version, we can add an informative note, or make  
>> it a MAY or OPTIONAL level requirement.
>> How does that sound as a general approach?
>> I thinkingthe no-arg version of send() would fall under this  
>> category. It does seem like something we want eventually, but could  
>> be MAY-level or an informative "some implementations allow this" note  
>> for XMLHttpRequest 1.0.
> I much prefer the reverse, "some implementations are broken w.r.t." 
> especially as Jonas has already said it will be fixed shortly - it would 
> be pretty odd to have a spec which says 1 implementation is broken, when 
> that implementation isn't even broken.

I think there is need for some perspective here. Mozilla isn't broken in 
that .send doesn't work at all or that we in some cases have very broken 
behaviour. We simply follow DOM convention and have all the parameters 
to the function required even if they in some cases are not needed.

I agree this is bad because it doesn't follow convention with IE which 
was has been the goal of everyone implementing XMLHttpRequest.

Even if I fix this today, there will be around 6 months before a release 
with the fix is out, and another 6 months to a year before everyone has 
upgraded. So authors will have to provide an argument for a significant 
time if they want to work in the major browsers.

Another problem is how do you write the spec the other way around? We 
can't say that you MUST make the argument optional and at the same time 
say that you MAY require the argument.

 > I am more
> concerned about implementors not making the mistake.

If this is the only concern then I think we could just mention in the 
spec that "implementors are advised to make this argument optional since 
there is a lot of content on the web today that calls the function with 
no arguments".

/ Jonas
Received on Friday, 3 March 2006 19:39:50 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:16:20 UTC