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

Re: XMLHttpRequest progress events

From: Maciej Stachowiak <mjs@apple.com>
Date: Sun, 23 Apr 2006 23:38:14 -0700
Message-Id: <98F409D1-9DF2-426E-9FDB-A06DF42547B4@apple.com>
Cc: Robin Berjon <robin.berjon@expway.fr>, "Web APIs WG (public)" <public-webapi@w3.org>
To: Boris Zbarsky <bzbarsky@mit.edu>

On Apr 23, 2006, at 4:43 PM, Boris Zbarsky wrote:

>>> Does this seem acceptable?  If not, are there counter-proposals?
>> I'm not sure what (and who) exactly you're asking here :)
> I'm asking the WG whether it's acceptable to them for us to add  
> these events in the null namespace for XMLHttpRequest or whether  
> that conflicts with planned work on the part of the WG or with the  
> XMLHttpRequest implementations of other participants of the WG.  I  
> realize that you guys may not know what you're planning to  
> standardize in version 2, and I'm not proposing we add progress  
> events to version 1, but if there _are_ concrete plans to add  
> progress events to version 2, I might as well implement those now.

It is likely the WG will add progress events in version 2, but there  
are no concrete plans. I think the most constructive way to proceed  
would be to make a specific proposal for how they should work, even  
though the WG is not working on version 2 yet (much the way Ian made  
his post-1.0 proposal for cross-site XHR).

> If there aren't, I'll just implement what I feel like and then if  
> progress events are brought up in version 2 I'll likely try to get  
> our implementation to be the spec.  ;)

I understand that as an implementor you don't want to sit around  
waiting for standards bodies to include features in a spec before  
implementing them. However, since there is an active, open and  
healthy standards process going on here, you may want to try  
participating in it more actively before taking that approach.

> So I'm not asking the WG to sanction this implementation decision,  
> but I'm asking whether any of the WG members can think of obvious  
> reasons this is a bad idea in terms of where they want to take the  
> spec or in terms of interaction with their implementations.

It would be much easier to judge this given a more specific proposal.  
Pretend you are writing a rough draft the spec language. Questions  
your original proposal did not answer include:

1) What interface would the "progress" and "uploadprogress" events  

2) Are there any requirements on when implementations should fire  
these? (Is it required that there be at least one? Do you get one for  
0 progress? etc...)

Received on Monday, 24 April 2006 06:41:54 UTC

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